2021-03-01 04:55:05 Hi, I see a bunch of download options. Which should I go for? ...noting that ethernet is not an option, wifi only. 2021-03-01 05:02:43 I'm wanting to use it as my main OS on what used to be a win10 "cloudbook". 32gb flash only, 2gb ram./ 2021-03-01 07:07:12 Ok, I got extended on a thumb drive. I added bootia32.efi to secure boot list and I'm in shell. I type bootia32.efi into the efi shell and hit enter. It returns with no error and does not boot. What am I missing? 2021-03-01 07:20:34 find . -type f -exe file {} \; | grep -i efi ... reveals that that is the only EFI boot file. I don't know what to do if the only boot file doesn't boot. 2021-03-01 07:30:21 ...and non-EFI legacy mode boot just freezes. nice. 2021-03-01 07:39:17 ...also locks up with the x865_64 on a diff thumb drive. #QualitySoftware 2021-03-01 07:46:58 piousminion: alpine doesn't support secure boot, afaik 2021-03-01 07:49:06 also tried dd mode. freezes as syslinux boot prompt as well. 2021-03-01 07:49:24 456 people in here. crazy 2021-03-01 07:55:21 crazy is that I'm trying to help all these NG newcomers 2021-03-01 07:56:14 ng? 2021-03-01 07:56:30 New Generation :D 2021-03-01 07:56:32 NG - new generation 2021-03-01 07:56:48 yea but like, of what 2021-03-01 07:56:50 but actually 'Not Good' 2021-03-01 07:56:55 did alpine just have a new release or something 2021-03-01 07:57:08 and most of them cant even survive 10 minutes on IRC lately... 2021-03-01 07:57:12 or do you mean that whatshisface the bald shakespeare guy is using alpine now 2021-03-01 07:57:28 twb: it is not hard to visit alpine web site and read 2021-03-01 07:57:49 mps: I just don't understand what your referring to 2021-03-01 07:58:03 twb: np 2021-03-01 07:58:25 If you had said "FNG" I would have understood perfectly :-) 2021-03-01 07:58:55 FNG, F**king NG :) ? 2021-03-01 07:59:07 Yes it's a reference to the Resistance War Against America 2021-03-01 07:59:37 heh, learned something new this morning 2021-03-01 08:00:07 Normally troops are rotated out in large batches, but USA was bad at war so they ran out of troops and were replacing them one at a time 2021-03-01 08:00:21 So existing group got a "FNG" who was probably going to die before you learned his name 2021-03-01 08:10:07 Has anyone noticed Firefox tabs crashing with some video meeting applications like Google Meet? I'm used to Chromium tabs constantly crashing but Firefox is new to me. It doesn't crash with Jitsi, but on an average 10 minute video call with Google Meet it crashes at least once and often twice 2021-03-01 08:10:11 Cogitri: maybe? 2021-03-01 08:10:27 PureTryOut[m]2: firefox esr or firefox regular? 2021-03-01 08:11:00 Regular 2021-03-01 08:37:53 PureTryOut[m]2: Hm, haven't had that happen to me, but I don't use Google Meet. I only use Jitsi occasionally in Firefox and all the Zoom stuff via their client 2021-03-01 08:38:24 Well give it a shot, it's really annoying lol. I constantly have to rejoin the meeting 2021-03-01 08:39:22 Does the tab crash or the entire browser? 2021-03-01 08:41:05 Just the tab. In case of Chromium all the tabs, but for Firefox just the one 2021-03-01 08:42:23 Hmm, does it print something useful in the console? 2021-03-01 10:05:56 Hello everyone, I would like to update a package on my alpine 3.13 but the package is only updated in "edge". when will it be available in 3.13? or what is the strategy of alpine linux about it? the update of the package seems quite minor to me though. i hope I shouldn't wait for alpine 3.14 2021-03-01 10:06:55 GreyXor: what package / version 2021-03-01 10:07:40 @ikkde ; duplicity 2021-03-01 10:07:51 @ikke 2021-03-01 10:08:24 Cogitri: I haven't had the console open, or do you mean terminal (didn't launch it via terminal either but ok) 2021-03-01 10:09:57 GreyXor: it's a patch / bug fix release, so it can be backported 2021-03-01 10:10:54 ikke indeed, and then it's an automatic process or the user (me) has to make the request to alpine linux? 2021-03-01 10:12:11 i will looking for the alpine package updating strategy 2021-03-01 10:12:54 PureTryOut[m]2: ah, I was referring to the terminal. If the thing crashes so often it should be easy to reproduce 2021-03-01 10:14:05 I'll run it from terminal next time, but I only use Google Meet once a week on monday so it'll take a bit 😅 2021-03-01 10:15:54 Ah, thought you could open a test meeting to check it 2021-03-01 10:24:50 Oh maybe. Idk I never host meetings myself, I'll give it a shot 2021-03-01 11:09:22 hi all 2021-03-01 11:10:32 i'm trying to figure out how to do unattended installation from any linux cloud image using cloud-init. i'm able to download AL kernel/initrd, copy stuff via apkovl, to use kexec to restart kernel (eg. on Ubuntu) and then...? 2021-03-01 11:11:38 I read initrd init shell script... but my question is if I replace initrd init with my installation stuff, how to do reboot? should/can i call exec busybox reboot in the end? 2021-03-01 11:12:22 or... i was thinking to let initrd init be default and customize /etc/inittab via apkovl... any idea? 2021-03-01 11:14:46 I also see a problem, apkovl is extracted before adding alpine-base in initrd init, thus it would replace my extracted files 2021-03-01 11:15:43 (btw in OpenBSD custom files in siteXX could have script which is called very late, maybe such feature - to rm -rf some files after installing pkgs in initrd init - would be handy) 2021-03-01 12:59:11 f*ck i have to go in a teams meeting but firefox is not supported. any quick fix 2021-03-01 12:59:14 ? 2021-03-01 13:01:40 chromium 2021-03-01 13:02:50 ah yeah, saw it on their page. microshit 2021-03-01 13:46:30 it froze all the time. microshit has to work on performance lol 2021-03-01 14:35:44 Hi,i'm sorry to bother you, i would like to make a question about Alpine installation on an arm board that i got, i hope to be in the right channel to make this question, otherwise i'm sorry. I would like to install Alpine on an AML-S905X-CC board and i would like to ask if the "Generic ARM branch" (or one of the other branches) support that board. Thanks for listening! 2021-03-01 14:41:15 unfortunately i have not enough knowledge to build uboot and related stuff, that's why i'm currently using Armbian, simply because they seems to be the only distro that explicitly support it. Some time ago, they they have switched to mainline kernel, i was wondering if i can install Alpine on it. Honestly i don't like Debian based distros 😁 2021-03-01 14:43:05 * unfortunately i have not enough knowledge to build uboot and related stuff, that's why i'm currently using Armbian, simply because they seems to be the only distro that explicitly support it. Some time ago, they they have switched to mainline kernel, so i was wondering if i can install Alpine on it. Honestly i don't like Debian based distros 😁 2021-03-01 14:44:03 what do most people use here, pip or compile your own ? 2021-03-01 14:45:39 uniformbufferPin: can you give more info about board 2021-03-01 14:46:14 what SOC it is based on 2021-03-01 14:46:28 g0r3: just fake user agent with settings or some extension 2021-03-01 14:46:47 i used to fake chromium on openbsd as chrome on linux 2021-03-01 14:47:49 mps: sure, the board is an Amlogic S905X with 2GB of ram,it has 4 cortex a53 with a mali 450 mp3 gpu 2021-03-01 14:48:09 * mps: sure, the board has an Amlogic S905X soc with 2GB of ram,it has 4 cortex a53 with a mali 450 mp3 gpu 2021-03-01 14:49:04 it is from a company called Libre Computer, their product page is this: https://libre.computer/products/boards/aml-s905x-cc/ 2021-03-01 14:49:42 found it, thanks 2021-03-01 14:50:23 alpine aarch64 userspace should work out-of-the-box 2021-03-01 14:51:05 uniformbufferPin: do you know does board use UEFI to boot, or need u-boot 2021-03-01 14:52:31 mps: Armbian (that i'm currently running) seems to be using u-boot 2021-03-01 14:53:00 aha 2021-03-01 14:53:13 i don't know if also support uefi 2021-03-01 14:54:13 all arm64 boards should support uefi, but again some don't 2021-03-01 14:56:06 from their website, the company offer an ubuntu 18.04 with uefi, so maybe it is supported? (https://libre.computer/2019/02/18/ubuntu-18-04-2-lts-images-for-le-potato-with-uefi/) 2021-03-01 14:56:22 if you persist here for some time and help us we can look what is needed to add for this board 2021-03-01 14:57:49 sure, i will try to help as i can 2021-03-01 14:57:50 for now, you can use u-boot and kernel from Armbian (as you already have it) and install alpine aarch64 on separate root partition 2021-03-01 15:00:06 I'll look at alpine kernel to see do we need to enable some drivers or options for this board 2021-03-01 15:00:48 thanks a lot for the help and the dedicated time 2021-03-01 15:03:48 np 2021-03-01 15:10:29 uniformbufferPin: unfortunately this board is not enabled in alpine kernels 2021-03-01 15:11:10 we should and we will (I think) in next kernel upgrade 2021-03-01 15:11:17 ncopa: ^ 2021-03-01 15:12:17 have to look in u-boot source to find what needs to be done there to enable it on alpine 2021-03-01 15:15:08 thanks again for the invested time and effort 2021-03-01 15:20:27 but to repeat: for now, you can use u-boot and kernel from Armbian (as you already have it) and install alpine aarch64 on separate root partition, or if you have spare mmc you can replace armbian root FS with alpine 2021-03-01 15:22:34 yes, i have a second sd card where i would like to install Alpine, so i will follow your suggestion to replace the root filesystem 2021-03-01 15:23:49 that is how I started to use alpine on arms, using armbian and arch linux kernels and u-boots :) 2021-03-01 15:25:37 mps: what kernel config knob is it? 2021-03-01 15:27:52 ARCH_MESON, and probably some drivers, LIMA DRM 2021-03-01 15:28:09 have to look, but don't have much time right now 2021-03-01 15:45:30 would be good with a ticket with the details 2021-03-01 15:45:43 with category kernel 2021-03-01 15:45:56 i ted to froget things that is only mentioned on irc 2021-03-01 15:51:00 amen 2021-03-01 15:51:13 for work it's s/irc/slack/ 2021-03-01 15:51:17 but so much that 2021-03-01 15:53:31 uniformbufferPin: looks like we already have u-boot for this board, named as u-boot-libretech-cc 2021-03-01 15:53:51 mps: Want to hear something interesting? 2021-03-01 15:54:03 You'll definitely find it interesting.. 2021-03-01 15:54:29 After buying my espressobin I noticed that netgate's newest line of arm based pfsense devices looked VERY much like my espressobin 2021-03-01 15:54:36 fys: yes, I want to hear old sayings, for example 'nothing new under the sun/sky' :) 2021-03-01 15:54:46 but their arm version isn't open source or available for download 2021-03-01 15:54:58 ... so i bought one for the license. 2021-03-01 15:55:21 hmm 2021-03-01 15:55:41 im going to do some... playing 2021-03-01 15:55:43 :) 2021-03-01 15:56:08 fys: thank you for yet another 'nothing new under the sun/sky' :) 2021-03-01 15:56:15 teehee 2021-03-01 15:56:37 im part apache, my native name is very special 2021-03-01 15:56:40 Runs With Scissors 2021-03-01 15:57:05 its okay, i actually am part apache so i get to make that joke :P 2021-03-01 15:57:13 hehe, nice name but be careful 2021-03-01 15:57:23 (: 2021-03-01 15:57:50 don't run too much 'with your name' 2021-03-01 15:57:54 i meant to play with netgate stuff earlier this month but i spent time in the hospital 2021-03-01 15:58:59 im sure based on the information ive given you know exactly what im planning to do 2021-03-01 15:59:02 ;) 2021-03-01 16:00:11 heh, maybe you will need more tools, hammer? :D 2021-03-01 16:00:35 Haha. 2021-03-01 16:05:24 btw, I'm pleasantly surprised to this talk, because I had impression that Apaches are extinguished, and now see that my impression was wrong 2021-03-01 16:05:56 I am a member of the Mescalero 2021-03-01 16:06:31 Geronimo is supposedly a distant relative of mine 2021-03-01 16:06:47 these are 'sub tribe' of Apaches, iiuc 2021-03-01 16:06:52 Yes 2021-03-01 16:07:31 ok, from now on I will call you Geronimo 2021-03-01 16:07:35 :D 2021-03-01 16:08:06 It's funny to explain to people how all that works because.. 2021-03-01 16:08:19 While I'm an Apache which is designated "Native American" 2021-03-01 16:08:26 We're hispanic. 2021-03-01 16:08:34 lol 2021-03-01 16:09:32 I'm just as much a "native mexican" as a "native american" 2021-03-01 16:09:41 yes, people like to interbred 2021-03-01 16:09:42 Life is strange. 2021-03-01 16:09:56 Well it's more that New Mexico used to be plain old mexico 2021-03-01 16:10:06 and we became "native americans" in the process 2021-03-01 16:10:09 aha 2021-03-01 16:10:26 heh, process called occupation 2021-03-01 16:10:31 Indeed. 2021-03-01 16:10:39 I'm not bothered by such things. 2021-03-01 16:10:47 Borders rise, fall, change, meh. 2021-03-01 16:11:04 but good you are not extinct in that time 2021-03-01 16:11:14 Haha, I like to thank so. I'm happy to be alive. 2021-03-01 16:11:15 ;) 2021-03-01 16:12:33 but we are OT here 2021-03-01 16:12:47 aye 2021-03-01 16:12:50 2021-03-01 17:34:16 hello ppl 2021-03-01 17:34:52 I got an incredible issue today on one of my VPS servers upgrading to the latest edge 2021-03-01 17:35:17 i ended up with the openrc scritpt in /etc/init.d/networking DELETED 2021-03-01 17:35:30 so my network would not start 2021-03-01 17:35:45 and since it was a vps, things got pretty hairy 2021-03-01 17:36:06 This was a temporary situation, which only happened on edge 2021-03-01 17:36:19 I had to use a rescue disk, find a working /etc/init.d/networking and put it on again 2021-03-01 17:37:05 it was very weird situation :) 2021-03-01 17:37:24 wyk72: doesn't your vps provider have a console you can log into? 2021-03-01 17:37:40 yeah, it did, but was very limited 2021-03-01 17:37:58 and without network I could not scp/find somewhere anything 2021-03-01 17:38:47 You could just do ifup tried many times 2021-03-01 17:39:00 but it did not work at all 2021-03-01 18:59:00 eut installer google-chrome? 2021-03-01 19:08:15 is it possible to install google-chrome??? 2021-03-01 19:08:46 You can use chromium 2021-03-01 19:09:32 then, no way for google-chrome??? 2021-03-01 19:09:45 google-chrome is built against glibc 2021-03-01 19:10:02 okay, thanks a lot 2021-03-01 19:10:26 If u rlly want chrome there might be a flatpack 2021-03-01 19:11:02 thank you dear 2021-03-01 20:01:46 is it intentional that the files lbu creates for /etc persistence are world-readable by default? i couldn't find any issues, mails or documentation regarding this, so maybe i'm missing something here? 2021-03-01 20:49:48 Hello, i would like to encrypt my system, i run rpi 4 in tmpfs. Anything i need to know as a alternative to lvm luks setup guide? I have already installed the system, and want to migrate to luks 2021-03-01 23:11:21 Hello 2021-03-01 23:11:41 Can i run luks, tmpfs on rpi? 2021-03-01 23:12:54 Sure 2021-03-01 23:13:17 Can i follow the lvm on luks guide for rpi, 2021-03-01 23:13:18 ? 2021-03-01 23:17:51 I suppose you could try, haven't setup a Pi yet 2021-03-01 23:31:58 luks on tmpfs does not make a lot of sense 2021-03-01 23:36:47 Luks on LVM says it needs to be installed on new, but can i migrate instead? 2021-03-02 00:20:15 Back, trying to not disconnect 2021-03-02 01:12:53 Can I do this for multiple repo's at once ? apk --no-cache -X http://dl-cdn.alpinelinux.org/alpine/edge/testing ? 2021-03-02 01:48:03 Hi Lutin 2021-03-02 01:49:17 I already have extended and standard. What can I expect before booting if I begin reading documentation and wikis 2021-03-02 02:04:20 Ok. How far can I get with the default repositories in standard and extended? 2021-03-02 03:01:58 If I do a "apk manifest \*" I have two entries for /etc/profile. Why are there two entries, and how would I determine where the second one came from? 2021-03-02 03:02:45 One is in alpine-baselayout. Not sure where the other one came from. 2021-03-02 03:03:55 Speaking of which, where are the apk/package metadata file stored in alpine linux? 2021-03-02 03:24:28 never mind. for some weird reason it appears that alpine-baselayout is showing up in the manifest twice. 2021-03-02 03:24:50 dunno why, but the hash is the same so am gonna ignore it. 2021-03-02 08:16:38 hey there 2021-03-02 08:18:21 👋 2021-03-02 08:19:21 I'm trying to use alpine in a docker. I only need to run rabbitmq, so I use CMD to start it with rc-service, but since that process ends immediately, it also forces the container to close (CMD ["rc-service", "rabbitmq-server", "start"]). What is the correct way to do it? 2021-03-02 08:52:10 Start it in the foreground 2021-03-02 14:44:33 What do i need to mount a "raw VM disk image" 2021-03-02 14:45:09 Have tried diffrent modprobe but it does not work 2021-03-02 14:45:59 https://tpaste.us/QWvz 2021-03-02 15:55:31 Anyone have experience with Plymouth on Alpine? 2021-03-02 17:27:11 (ask someone who like simple things in text mode :) ) 2021-03-02 17:46:43 mps: lol its not for a desktop - it's for an appliance im building based on alpine 2021-03-02 18:23:18 Excellent. 2021-03-02 18:23:34 HTPC remote I bought for my Alpine Kodi/Plex box worked right out of the box. 2021-03-02 18:23:37 Zero config needed. 2021-03-02 18:29:57 I do however want to customize it a bit, but it "just works" out of the box. 2021-03-02 18:30:00 :) 2021-03-02 18:30:09 Alpine is a GREAT solution for Kodi/Plex 2021-03-02 18:49:29 link for remote? 2021-03-02 18:49:36 <-- lazy 2021-03-02 18:50:24 that sounds like an ideal solution. <3 alpine on the pi, so kodi there sounds nice. I'm not so happy with libreelec atm 2021-03-02 18:50:34 hold on and i will link you 2021-03-02 18:50:53 https://www.amazon.com/gp/product/B06XHF7DNQ/ref=ppx_yo_dt_b_asin_title_o00_s00?ie=UTF8&psc=1 2021-03-02 18:51:06 That is 100% the reason I built my box. 2021-03-02 18:51:11 I hate LibreElec 2021-03-02 18:51:25 I've got Alpine doing Kodi + Plex now. 2021-03-02 18:51:32 + That remote. 2021-03-02 18:51:33 we've had a lot of issues with it over time, its time for it to go 2021-03-02 18:51:54 sweet 2021-03-02 18:51:56 thank you 2021-03-02 18:51:59 np 2021-03-02 18:52:19 I kept notes but basically all you have to do is setup_xorg_base, apk add kodi, put kodi in xinitrc 2021-03-02 18:52:22 and startx 2021-03-02 18:52:24 and you're done 2021-03-02 18:52:33 nice 2021-03-02 18:52:46 you can also get kodi to autostart with some initttab hackery 2021-03-02 18:52:47 :) 2021-03-02 18:52:57 I will probably do that 2021-03-02 18:53:03 im at that point, just want to add plymouth now 2021-03-02 18:53:06 and it'll be done 2021-03-02 18:53:07 lol 2021-03-02 18:53:10 nice 2021-03-02 18:53:21 it really feels like a "product" 2021-03-02 18:53:43 4K X265 hardware decoding direct from my server. 2021-03-02 18:53:50 It's pure love. 2021-03-02 18:54:22 nice 2021-03-02 20:09:58 I love the fact most of my house runs on Alpine at the moment.. 2021-03-02 20:10:12 My router, HTPC, file server, etc etc 2021-03-02 20:10:13 lol 2021-03-02 20:10:24 So diverse. 2021-03-02 20:48:31 my home runs mostly (only?) FreeBSD. 2021-03-02 20:48:50 I wonder if people run Alpine on their laptop 2021-03-02 20:49:22 i do 2021-03-02 20:49:54 and my desktop pc runs freebsd :D 2021-03-02 20:50:23 it seems like a nice alternative to most. no systemd, a nice package manager, uses OpenRC. what's the package support on it? 2021-03-02 20:50:33 I run FreeBSD on a laptop, never looked back. 2021-03-02 20:50:54 there would be some bonuses and downsides with alpine i thik 2021-03-02 20:51:19 what's the downsides? 2021-03-02 20:51:44 the having musl instead of glibc can be annoying if you want some closed source things or somethign that really demands glibc because of quirks 2021-03-02 20:51:52 the big one for me is android studio 2021-03-02 20:52:05 but i fixed that by installing it with nix package manager 2021-03-02 20:52:17 which just throws glibc in a chroot ish thing, basically 2021-03-02 20:52:31 yeah I can't have Android Studio on my FreeBSD laptop as well, but my workaround was using a VM and USB passthrou to a phone. 2021-03-02 20:52:55 and setting up graphics i found to be a bit of a pita due to fonts not being pulled in as dependencies 2021-03-02 20:53:00 once that's done it's done tho 2021-03-02 20:53:40 and freebsd is just impeccably tidy and coherent compared to anything linux xD 2021-03-02 20:54:19 I want to use Hyperbola GNU/kFreeBSD 2021-03-02 20:54:33 or Guix GNU (w/ Hurd) 2021-03-02 20:54:44 yeah, that's the main reason I moved, but now I'm sticking because of Jails, OpenZFS and DTrace. I cannot believe there's no dynamic tracing in the Linux world yet. 2021-03-02 20:54:51 are there people that actually use hurd? 2021-03-02 20:54:54 oh yes, GNU Hurd, is it still beta? 2021-03-02 20:55:01 yep 2021-03-02 20:55:20 I used it once back in school. but didn't know much. it's architecture is amazing to be honest. 2021-03-02 20:55:55 Guix recentlyish added initial support for it (but the blog post only showed it in a VM IIRC) 2021-03-02 20:56:19 this channel is to help alpine users 2021-03-02 20:56:25 sorry 2021-03-02 20:56:36 I'll move to #alpine-offtopic 2021-03-02 20:58:57 mps, sorry! 2021-03-02 20:59:29 antranigv: np :) 2021-03-02 20:59:58 sometimes we all are 'out of subject' 2021-03-02 21:01:21 all questions about alpine are ok here and difference with other OSes, but discussing other OSes here - not sure how it is useful 2021-03-02 22:34:48 in some way I like that Alpine places phpX in /etc but in some way I like it better if they do in /etc/php/ 2021-03-02 22:35:53 pardon? 2021-03-02 22:36:10 I don't fully understand. 2021-03-02 22:36:27 You'd prefer /etc/php/phpX/ ? 2021-03-02 22:39:33 ? 2021-03-02 22:47:27 fys yes the Ubuntu/Debian was 2021-03-02 22:47:30 way 2021-03-02 22:47:35 I think centos does the same 2021-03-02 22:56:26 ln -s . /etc/php 2021-03-03 00:00:21 Hello71 ? 2021-03-03 00:00:47 Hello71 I think that is not the right solution 2021-03-03 00:07:33 fys / 2021-03-03 00:07:34 ? 2021-03-03 03:30:32 why would Alpine not load installed modules by default for PHP ?They are in conf.d 2021-03-03 05:33:38 which SAPI? 2021-03-03 05:35:15 oh, gone 2021-03-03 09:25:48 is Nathan Johnson in here ? 2021-03-03 09:38:41 is it intentional that the files lbu creates for /etc persistence are world-readable by default? i couldn't find any issues, mails or documentation regarding this, so maybe i'm missing something here? 2021-03-03 10:07:43 How do i mount a disk file with btrfs in alpine linux 2021-03-03 10:08:00 wrong fs type, bad option, bad superblock on /dev/loop0, missing codepage or helper program, or other error. 2021-03-03 10:08:33 I did a mount -t btrfs imagefile /mnt/mount 2021-03-03 10:13:22 Jenkler: mount -o loop .... 2021-03-03 10:14:04 hey. apk upgrade end with "1 error; 3692 MiB in 622 packages". how can i display the error? 2021-03-03 10:14:38 g0r3: try 'apk fix' 2021-03-03 10:15:25 ah, thx 2021-03-03 10:16:00 mps does not work 2021-03-03 10:16:09 but maybe the img is broken 2021-03-03 10:16:45 Jenkler: maybe file have partition table 2021-03-03 10:16:47 I will try with another img that i know works in ubunto 2021-03-03 10:16:58 loop file, image 2021-03-03 10:32:06 mps trying to unpack with bsdtar now and see if thats the issue 2021-03-03 10:32:31 ubunto and bsdtar worked on the file I am unpacking now 2021-03-03 10:36:17 Jenkler: unpacking FS image file with tar!? strange 2021-03-03 10:53:22 mps, the issue was not using bsdtar ;) 2021-03-03 10:53:46 Now it works with mount -t btrfs image /mountpoint 2021-03-03 10:53:48 ;) 2021-03-03 12:14:59 i can't boot into vmware player with virt iso 2021-03-03 12:57:12 Are the only headless dlna servers on Alpine, minidlna and gerbera? 2021-03-03 17:40:40 How can I reach a package maintainer from an Alpine package ? 2021-03-03 17:41:55 email 2021-03-03 17:42:06 should be listed in the APKBUILD file 2021-03-03 17:42:34 but some are quite unresponsive, or uninterested in people even sending them patches directly 2021-03-03 17:47:18 Or gitlab issue 2021-03-03 17:47:41 yZ5vlALg86lP OK, I didn't find anything so far, but indeed. it's pretty annoying when they don't respond 2021-03-03 17:50:10 how you mean > didn't find anything so far. you mean you looked in the APKBUILD and there was no maintainer listed? that is kinda impossible, given the scrutiny of reviewers before merging patches 2021-03-03 17:52:35 yZ5vlALg86lP: there are pkgs without maintainer 2021-03-03 17:54:02 yZ5vlALg86lP no maintainer is listed 2021-03-03 17:55:19 wow 2021-03-03 17:56:06 why are those not in unmaintained then? 2021-03-03 17:56:53 how did they get into the repo in the 1st place without a maintainer? 2021-03-03 17:57:11 no one stepped for this, or no one wants to take these 2021-03-03 17:57:27 but how did they get in the repo without a maintainer? 2021-03-03 17:57:34 yZ5vlALg86lP: these are usually 'orphaned' pkgs 2021-03-03 17:57:58 and why are they not moved to unmaintained then? 2021-03-03 17:58:33 but I also noticed few times 'merge/push' pkgs without maintainer iirc 2021-03-03 17:59:06 yZ5vlALg86lP: maybe they are usable or some other depends on them 2021-03-03 17:59:14 is it possible some of those packages had maintainers but were alienated somehow? 2021-03-03 17:59:47 yes, afaik all had maintainers 2021-03-03 18:00:13 and is there any info why they were alienated? 2021-03-03 18:00:30 that sounds like something quite important to figure out 2021-03-03 18:01:38 reasons are different, as people are different 2021-03-03 18:04:58 yeah, but there might be reasons that can be fixed/corrected and maintainers return. 2021-03-03 18:05:05 anyway this: > find main community -name APKBUILD -print0 | xargs -0 grep -iL "^# Maintainer: " | cut -d/ -f 2 | wc -l 2021-03-03 18:05:09 gives me 188 packages 2021-03-03 18:08:26 ahm, but some of these packages are still maintained by other developers and contributors, though I don't have any statistic 2021-03-03 18:09:47 i still propose to move all of them to unmaintained 2021-03-03 18:13:45 we will also have to move dependants, which will be not small number of pkgs 2021-03-03 18:15:35 only if noone steps up to be maintainer 2021-03-03 18:25:16 > find main community -name APKBUILD -print0 | xargs -0 grep -i "^# Maintainer:$" | cut -d/ -f2 | wc -l 2021-03-03 18:25:22 gives another 184 packages 2021-03-03 18:37:00 ash doesn't have pushd/popd combo? 2021-03-03 18:41:20 No 2021-03-03 18:43:34 any trick to 'emulate' these two? 2021-03-03 18:52:07 Subshell 2021-03-03 18:52:51 (cd foobar; do_something;) 2021-03-03 18:53:10 in APKBUILD? 2021-03-03 18:53:15 Yes 2021-03-03 18:53:34 hmm, will consider. thanks 2021-03-03 18:55:26 In an APKBUILD file you can ofcourse split it over multiple lines 2021-03-03 18:56:34 that sounds good then 2021-03-03 22:44:04 Hi everyone, s6-overlay will be backported to 3.13 ? 2021-03-03 23:04:47 engine chill... next up launch 2021-03-03 23:13:31 T minus 1 min and counting... 2021-03-03 23:15:30 LIFT OFF! 2021-03-03 23:18:53 Apachez: want #alpine-offtopic ? 2021-03-03 23:21:41 OH MY F**KING ROFLCOPTER!!! ITS STANDING!!111oneoneone :D 2021-03-03 23:22:42 asdflkj: will try next time but this is history :) 2021-03-03 23:22:57 cartoonbased spacerocket launched and landed without going in pieces :D 2021-03-03 23:23:40 it's not the first rocket to do that though, is it? 2021-03-03 23:26:10 its the first SN rocket who does that 2021-03-03 23:26:25 the first two attempts went into "rapid disassemble mode" 2021-03-03 23:26:36 "rapid unscheduled disassemble mode" 2021-03-03 23:30:19 oh f**k it tipped over :D 2021-03-03 23:41:12 Flisk: at a guess, it just uses umask 2021-03-03 23:42:59 it seems like it should probably umask 077 before writing that file instead of putting your private ssh host keys in a world-readable tarball 2021-03-03 23:45:02 i mean tar doesn't either 2021-03-03 23:45:22 but it could probably print a warning at least if it doesn't already 2021-03-03 23:46:01 tars scope isnt creating a snapshot of your /etc though 2021-03-03 23:48:49 can be 2021-03-03 23:49:03 same with cp, vi editing a new file, etc 2021-03-03 23:49:14 yes but these are general purpose tools 2021-03-03 23:49:22 but like i said it could probably print a warning at least 2021-03-03 23:50:03 yeah i might open an issue about this since no one yelled at me for missing anything obvious :P 2021-03-04 00:45:51 Apachez kicked out of ##networking or so ? 2021-03-04 02:54:27 Lutin: nope, why do you wonder? 2021-03-04 02:54:47 or you meant that you were kicked out? 2021-03-04 02:56:19 Apachez no, I know better resources then that channel :D 2021-03-04 02:56:31 good for you 2021-03-04 02:56:50 Indeed! 2021-03-04 10:36:06 Are the only headless dlna servers on Alpine, minidlna and gerbera? Any recommendations or packages I may have missed? 2021-03-04 10:55:39 Hi, I've a problem with my lenovo yoga 920 with alpine. Sometimes it just hangs. The screen become black but there's still backlight. Only way out of it is a reboot. If I'm connected over ssh to the machine, the ssh connection also hangs. That is the machine is completely dead. I've previously had a hard time debugging this but now it happens half of the times i run make on a source tree 2021-03-04 10:56:15 I've updated alpine to latest edge, but I suspect this might be a hardware error. However I've no idea on how to debug this 2021-03-04 10:59:51 Hm, I suppose the logs in /var/log don't have anything useful in them either? 2021-03-04 11:02:21 iveqy: intel gpu? 2021-03-04 11:34:38 MY-R: yes 2021-03-04 11:41:24 iveqy: did you have that problem before on different kernels than 5.10.x? 2021-03-04 11:47:49 MY-R: I'm not sure. I use the rolling release, the edge release and I'm not quite sure when the problem arised. Before I went on edge, I did not have those problems 2021-03-04 11:48:11 is there a huge change in 5.10.x? 2021-03-04 11:50:48 iveqy: for me I couldnt use kernel 5.10 with intel haswell anymore because of hangs/artifacts etc BUT found solution like add to kernel command line this: i915.mitigations=off 2021-03-04 11:51:11 and everything working fine again like on older kernels 2021-03-04 11:52:04 there were many changes in intel drm in 5.10, enough if follow issues here: https://gitlab.freedesktop.org/drm/intel/-/issues 2021-03-04 11:53:34 thanks, I'm running memory tests on it now, but after that I'll check which kernel I use and try with i915.mitigations=off 2021-03-04 12:25:03 hi, i would like to use a raspberry pi 3b+ as media center and i would like to install Alpine armv7 on it. Unfortunaltely seems that there is no armv7 build for Kodi. On the APKBUILD there is a comment saying "`flatbuffers` missing on armhf and armv7", but there is an armv7 build of that package available on the repository, so i was wondering if maybe that package was unavailable some time ago and now that is 2021-03-04 12:25:03 available, the APKBUILD was simply left unchanged 2021-03-04 12:28:30 the reason to have an armv7 build instead of an aarch64 one is `inputstream-adaptive` plugin that, if it detect an armv7 build, it will "borrow" the Widewine executable (to reproduce drm protected content) from chrome os recovery image. I also hope the executable is runned in the recovery image environment, so there is no difference if the system is musl or glibc 🤞 2021-03-04 15:08:41 Hello! I was reading the official guide about sway and towards the end these two steps are recommended: 2021-03-04 15:08:49 set the XDG_RUNTIME_DIR to a suitable location (e.g. /tmp)". Then run Sway from the Linux console: 2021-03-04 15:09:08 $ XDG_RUNTIME_DIR=/tmp sway 2021-03-04 15:09:51 Does this command set both XDG_RUNTIME_DIR to /tmp and start sway? 2021-03-04 15:16:28 ewzyk: yes but better way to set up XDG_RUNTIME_DIR is to use this script if you not using elogind of course: https://wiki.alpinelinux.org/wiki/Wayland 2021-03-04 15:23:03 Understood. Instead how do I set Sway to start automatically on TTY login? With Arch Linux I had to add this to /etc/profile 2021-03-04 15:23:16 if [ -z $DISPLAY ] && [ "$(tty)" == "/dev/tty1" ]; then 2021-03-04 15:23:17 exec sway 2021-03-04 15:23:17 fi 2021-03-04 15:23:38 Doss this also apply to Alpine? 2021-03-04 15:24:04 i mean, even on arch that is kinda a shitty way of starting Sway 2021-03-04 15:24:43 but i see that is actually in the wiki haha 2021-03-04 15:25:15 i would've tried to create a systemd service file to do it, but maybe that's not possible? 2021-03-04 15:25:23 Indeed I read that in the wiki. 2021-03-04 15:26:14 I don't know how to do that honestly. 2021-03-04 15:26:24 seems like the whole thing is jank right now 2021-03-04 15:26:53 like wtf, /etc/profile is the best way? they say not to use systemd services, and no display manager is "officially supported"? lol 2021-03-04 15:27:21 what a joke ... :( 2021-03-04 15:33:03 you could also just add that to the local config for whichever shell you use (.bashrc, .zshrc, etc) 2021-03-04 15:33:33 I don't think there's necessarily a better way to do it if you don't want to use a login manager 2021-03-04 15:50:04 the alpine standard iso doesnt have network firmware for my card 2021-03-04 15:50:13 would the extended iso have the drivers i need? 2021-03-04 15:50:39 does apk handle alternatives? like vim/nvim? 2021-03-04 16:12:37 deldenzen: depends on your card 2021-03-04 16:12:56 i think its a realtek card 2021-03-04 16:13:41 if the firmware is in upstream linux-firmware then it is packaged 2021-03-04 16:14:15 it probably is 2021-03-04 16:14:32 i'll connect it to a phone with a cable to use usb hotspot and install linux-firmware 2021-03-04 16:14:35 would that work 2021-03-04 16:15:09 it should, if firmware is in upstream 2021-03-04 16:16:12 Piraty: no 2021-03-04 16:20:43 mps: thanks 2021-03-04 16:22:01 i asked before before about a nice place for one-time commands that i want to be run *before* default runlevel, on other system's i would use /etc/rc.local for that. the stuff i have in mind don't justify a service 2021-03-04 16:26:38 Piraty: it is not hard to create one and put it in sysinit level 2021-03-04 16:27:09 There is a local service that would execute scripts in /etc/local.d/, but that's in the default runlevel 2021-03-04 16:28:05 /etc/local.d/ is after default 2021-03-04 16:28:26 (at the end of default) 2021-03-04 16:28:37 hm, sorry 'after *' 2021-03-04 17:32:33 is there a way to get podman rootless to work on latest alpine? I cannot set kernel.unprivileged_userns_clone ; maybe there is a boot command line tweak to enable it? 2021-03-04 17:33:34 I know there are people here who use podman 2021-03-04 17:35:57 yeah, I thought i had it working months ago, just tried it recently and i'm having issues; my kernel is 5.10.19-virt ; CONFIG_USER_NS=y ; but I still get issues just simply doing podman pull fedora and I have entries in /etc/subuid and subgid 2021-03-04 17:36:40 I think that config just means that user namespaces are available 2021-03-04 17:38:52 Hey, I was trying to cross compile qt5-qtmultimedia from aports for aarch64 and it fails. Is that expected to work? 2021-03-04 17:39:18 (I needed to apply a patch for a bug) 2021-03-04 17:39:21 yeah, I'm looking around for why the corresponding sysctl was missing ... testing on another dev box now ... 2021-03-04 17:40:00 any1: We only cross-compile the basic toolchain, the rest is all built natively, so we do not test if cross-compilation works for all packages 2021-03-04 17:40:33 https://security.stackexchange.com/questions/209529/what-does-enabling-kernel-unprivileged-userns-clone-do 2021-03-04 17:40:51 "such a setting doesn't even exist in the mainline Linux kernel. " 2021-03-04 17:41:19 that explains it :-) 2021-03-04 17:42:09 ikke: Thanks. Wouldn't webengine need to be crosscompiled for armv7? I would assume that it needs some crazy amount of contiguous addressable memory linking like chromium does 2021-03-04 17:42:23 for linking* 2021-03-04 17:42:37 any1: Our builders are pretty sizable 2021-03-04 17:43:05 they are aarch64 machines running in 32-bits mode 2021-03-04 17:43:13 with armhf / armv7 userspace 2021-03-04 17:44:00 interesting 2021-03-04 18:04:57 regarding rootless podman on alpine, found solution: #1 (uid mapping starting at 100000 is too small, changed to 200000 in /etc/sub[ug]id files) #2 (need to load fuse and tun kernel modules - modprobe tun and fuse ) works perfectly now 2021-03-04 18:27:38 uniformbufferPin: do you know for any option which is needed to be enabled for Amlogic S905X, I'm just working on linux-edge ugrade and enabled defaults for this SOC 2021-03-04 18:54:31 Piraty: you can add "local" service to "boot" level and it will run at the end of "boot" but before "default" 2021-03-04 18:58:43 Piraty: https://tpaste.us/1Jw4 2021-03-04 19:01:04 thanks MY-R but what is that paste supposed to tell me 2021-03-04 19:02:06 Hello friends 2021-03-04 19:08:40 Piraty: that the order is like I said in previous message (local.d start before default) :P 2021-03-04 19:09:19 well then. 2021-03-04 19:09:19 ACTION reads on about init.d runlevels 2021-03-04 19:09:43 still using a 'service' is bad, just like so-called one-shots in systemd land 2021-03-04 19:23:13 til: your 27th drive on a linux box will be called sdaa 2021-03-04 19:32:38 fys: the solaris scheme was cleaner imho : c0t0d0s0 -> ctds :-) 2021-03-04 19:39:36 aquamo42: wow 2021-03-04 19:39:37 lol 2021-03-04 19:50:18 and crytic, newer liked this on solaris 2021-03-04 19:50:27 cryptic( 2021-03-04 19:50:35 cryptic* 2021-03-04 19:51:51 yeah, controller -c, target - t, disk -d and slice -s 2021-03-04 19:52:20 yes, I remember :) 2021-03-04 19:52:59 when I used solaris there were no hot plugable 'disks' 2021-03-04 19:53:27 actually yes, but not often 2021-03-04 19:54:06 you are making me remember some things i had forgotten - i seem to remember having to run drvconfig -a or something like that to detect new disks manually 2021-03-04 19:55:22 I didn't used solaris for long time, few years, and then converted sun boxes linux 2021-03-04 19:55:38 to linux* 2021-03-04 19:56:30 though also tested solaris on PC but didn't used it seriously 2021-03-04 19:59:23 i preferred SunOS 4.x - transitioning to Solaris was disruptive 2021-03-04 19:59:57 sunos 3 and 4 was more like Alpine, Solaris was like RHEL 2021-03-04 20:01:33 I was already on linux when I 'contacted' first time solaris, never was happy with it. and never worked with SunOS 2021-03-04 20:02:12 but had to administer some sun/solaris boxes for few years 2021-03-04 20:24:14 Hello, I am installed python2 and python3 from edge (patched for CVE2021-3177) on arvm7l alpine image base (3.12), but when I run python2 and python3 I am getting errors and can't use python 2021-03-04 20:24:50 https://www.irccloud.com/pastebin/MJlu87EZ/Edge%20Python2%2FPython3%20output%20on%20armv7l%2FAlpine3.12 2021-03-04 20:47:21 leo8989: Installing binaries from edge on 3.12 is broken 2021-03-04 20:48:27 As in always unsupported since it breaks like this 2021-03-04 21:09:36 I tried for arm32v7l/alpine-3.13/3.11/3.10 they seem all having issue installing edge packages, in this case how can I patch python on armv7l images? 2021-03-04 21:11:07 leo8989: combining edge with stable releases is not supported 2021-03-04 21:11:08 Yes, installing edge packages on anything but edge is unsupported and broken 2021-03-04 21:11:10 especially in this case 2021-03-04 21:11:30 because since 3.13, packages have been rebuilt with time_t being 64 bits 2021-03-04 21:12:57 leo8989: it should have been fixed on all stable branches 2021-03-04 21:13:03 Yup 2021-03-04 21:13:06 leo8989: https://gitlab.alpinelinux.org/alpine/aports/-/issues/12448 2021-03-04 21:13:28 so just making sure you have the latest version of python available for that version should be enough 2021-03-04 21:14:16 python2 is not affected 2021-03-04 21:14:30 (And EOL anyway) 2021-03-04 21:30:02 It seems python2 is affected https://security-tracker.debian.org/tracker/CVE-2021-3177 2021-03-04 21:30:53 Hmm, the description only mentions python 3.x 2021-03-04 21:32:16 Probably because 2.x is EOL, so the description does not mention it 2021-03-04 21:34:40 could be, but yeah python2 is affected https://security-tracker.debian.org/tracker/CVE-2021-3177 2021-03-04 21:38:46 https://www.activestate.com/blog/latest-python-3-vulnerability-affects-python-2/ 2021-03-05 11:53:27 mps: sorry for the late reply. Unfortunately i don't know. This seems to be what Armbian use to configure the kernel for this board: https://github.com/armbian/build/blob/master/config/kernel/linux-meson64-current.config . The specific components that the board has are Lima for gpu and MesonVDEC for the hardware decoding. The last one. afaik it is not working correctly neither in Armbian, but they have enabled 2021-03-05 11:53:27 it. 2021-03-05 11:55:00 uniformbufferPin: np 2021-03-05 11:56:38 yesterday I upgraded linux-edge and enabled ARCH_MESON SOC/platform which auto-enabled some drivers and options, I didn't enabled some other which I'm not sure if needed 2021-03-05 11:57:12 you can try to install linux-edge 5.11.3 and check will it boot 2021-03-05 11:58:44 also, you can open issue/wish for linux-lts and assign it to ncopa, so he will not miss it on next upgrade of -lts kernel 2021-03-05 11:59:04 https://gitlab.alpinelinux.org/alpine/aports 2021-03-05 11:59:41 yes, I forgot, thanks ikke 2021-03-05 12:02:08 uniformbufferPin: but without your testing we can't do much about this SOC, I don't think that any alpine developer have it 2021-03-05 12:02:26 mps: thanks a lot for the dedicated effort, i will try it right now, let me prepare the sd card 2021-03-05 12:02:46 mps: i can be the beta tested ^^ 2021-03-05 12:03:15 uniformbufferPin: good :) 2021-03-05 12:03:51 I will look later at above link you posted to see what else maybe needs to be enabled 2021-03-05 12:04:28 now I'm busy with $day_job 2021-03-05 12:04:52 mps: thanks again for the dedicated time 2021-03-05 12:06:05 uniformbufferPin: np 2021-03-05 12:06:50 and you can look here https://arvanta.net/alpine/install-alpine-on-acer-r13-chromebook/ for hints how I installed alpine aarch64 on mediatek chromebook 2021-03-05 12:07:00 only hints, not guide 2021-03-05 12:07:34 thanks! 2021-03-05 12:09:04 uniformbufferPin: btw, there is u-boot-libretech apk, I think it should work with your board 2021-03-05 12:23:43 I'm trying to install https://pkgs.alpinelinux.org/packages?name=kubernetes&branch=edge on a new 3.13 vm. Running `apk add kubernetes --repository=http://dl-cdn.alpinelinux.org/alpine/edge/main` Gives me "Installing kubernetes" It's supposed to contain https://pkgs.alpinelinux.org/contents?branch=edge&name=kubeadm&arch=x86&repo=testing but /usr/bin/kubeadmin just doesn't exist. What could be wrong? 2021-03-05 12:27:31 Output of the `apk add` https://pastebin.com/ukGJxka2 I would have assumed it should show it installing the sub packages. But no. 2021-03-05 12:41:26 you can't mix repos 2021-03-05 12:41:36 howdy 2021-03-05 12:59:35 * mps: i can be the beta tester ^^ 2021-03-05 17:56:24 hi, how do i find out what versions of a particular package exists in repo? 2021-03-05 17:57:31 https://pkgs.alpinelinux.org/packages 2021-03-05 17:57:37 Or `apk search` 2021-03-05 17:58:11 https://pkgs.alpinelinux.org/packages only shows the latest version 2021-03-05 17:58:53 There's only ever one version of a package in the repos 2021-03-05 18:00:38 oh 2021-03-05 18:01:05 Cogitri: you mean, only 1 version per release of alpine? 2021-03-05 18:01:20 Yes 2021-03-05 18:01:36 where can i find the version release dates of alpines 2021-03-05 18:01:39 -s 2021-03-05 18:02:02 https://alpinelinux.org/releases/ 2021-03-05 18:02:08 thanks 2021-03-05 18:31:28 mps: i'm sorry to cother you again, unfortunately i'm not able to make Alpine boot on AML-S905X-CC Libretech board. I'm not able to get to the uboot screen, so it is very likely i have builded the sd card in the wrong way. Honestly i have not understood the way uboot works... 2021-03-05 18:31:39 * mps: i'm sorry to bother you again, unfortunately i'm not able to make Alpine boot on AML-S905X-CC Libretech board. I'm not able to get to the uboot screen, so it is very likely i have builded the sd card in the wrong way. Honestly i have not understood the way uboot works... 2021-03-05 18:38:50 ACTION sent a long message: < https://matrix.org/_matrix/media/r0/download/matrix.org/OSHSsqITNbEzGfOpoNmNeGAR/message.txt > 2021-03-05 18:39:48 ACTION sent a long message: < https://matrix.org/_matrix/media/r0/download/matrix.org/lhAvVabeySilrHJJRJHpfHNg/message.txt > 2021-03-05 18:40:05 ACTION sent a long message: < https://matrix.org/_matrix/media/r0/download/matrix.org/zPMPeMQjBTrtCsQUlTiMFSXz/message.txt > 2021-03-05 18:40:38 ACTION sent a long message: < https://matrix.org/_matrix/media/r0/download/matrix.org/TrrdJqeBMNeYlPzDRSTcoxLx/message.txt > 2021-03-05 18:41:04 Uniformbuffer[m]: please don't do that 2021-03-05 18:41:04 so, there is obviously something that i miss on the way uboot is loaded.... 2021-03-05 18:41:19 we are in IRC not matrix 2021-03-05 18:42:28 mps: i'm sorry, i made something wrong? 2021-03-05 18:42:29 if you have longer text to send use paste service, tpaste.us or ix.io 2021-03-05 18:43:12 oh, ok, sorry, i didn't know long messages can cause problems 2021-03-05 18:47:00 it is interesting for company called libre*tech that don't have any technical guide for their products (except marketing babble) 2021-03-05 18:49:18 isnt matrix doing what it does with long messages a kind of paste service? 2021-03-05 18:49:39 I found short guide in u-boot source in file doc/board/amlogic/libretech-cc.rst 2021-03-05 18:49:57 that is, how to build and 'flash' u-boot 2021-03-05 18:51:52 not simple, and I can't help because I don't have board and can't try to test how it works 2021-03-05 18:52:28 but, having serial to usb adapter could be of great help for such devices 2021-03-05 18:52:56 yZ5vlALg86lP: yes, but ugly to read 2021-03-05 18:53:26 and people don't use it appropriately, usually 2021-03-05 18:53:44 found it, thanks. Even if it is difficult, i have something to work on 2021-03-05 18:54:30 matrix users 'blindly' type long messages without thinking it will be 'sent' to matrix 'paste' urls 2021-03-05 18:54:31 a lot of guides on how to get this board working refer to the guide to the khadas vim board that has the same soc 2021-03-05 18:54:43 yeah, agreed 2021-03-05 18:55:34 Uniformbuffer[m]: that can help, if the SOC is same and board similar, then reading other guides can be useful 2021-03-05 18:55:51 but then your recommendation is incorrect. i think you wanted matrix users to send each line seperately, instead of in bulk as a matrix paste 2021-03-05 18:56:14 yZ5vlALg86lP: well, you are right 2021-03-05 18:56:20 but i shutup, sorry for nitpicking. 2021-03-05 18:56:28 np :) 2021-03-05 18:57:11 mps: honestly i didn't know that long messages could be problematic. I will be more careful next time, sorry 2021-03-05 18:57:53 np :) (also to you) 2021-03-05 18:58:25 we all make mistakes with something we don't have experience or knowledge 2021-03-05 18:59:54 Uniformbuffer[m]: yes, u-boot docs 'shows' that khadas-vim is similar 2021-03-05 19:08:03 as i have understood, the `u-boot.bin` provided by `u-boot-libretech` package is the result of the uboot compilation 2021-03-05 19:08:44 so should i only need to write it with dd on the first sector in the point specified by the guide, right? 2021-03-05 19:33:11 Uniformbuffer[m]: I'm not sure is it built correctly on alpine but you can try and see 2021-03-05 19:34:05 Uniformbuffer[m]: do you have serial to usb adapter/cable 2021-03-05 19:34:57 unfortunately no, but even then, i have never used it 2021-03-05 19:35:39 anyway i have tried to write the u-boot.bin file where the guide says, but still no boot 2021-03-05 19:37:18 ok, we have no luck 2021-03-05 19:38:03 i'm trying to undestand what armbian does to boot the board, so appling the same tecnique 2021-03-05 19:38:16 though, good idea is to obtain serial cable for anyone using or playing with SBC devices 2021-03-05 19:39:17 armbian guides is quite sparse/short, how to dd image to mmc usually 2021-03-05 19:40:21 I can't ever read the word "armbian" without thinking of sleep medication for pirates. 2021-03-05 19:53:14 question -- the 'Local Backup' option on the acf website isn't working anymore; it just hangs there indefinately -- nothing in /var/messages and nothing output if I run mini_httpd in the shell with debug -- any ideas? 2021-03-05 19:53:39 I've trying removing it and reinstalling it, that also hasn't helped 2021-03-05 19:53:48 s/trying/tried 2021-03-05 19:55:00 Guest49: what version is? I could try to reproduce it but I could delay a while 2021-03-05 19:55:18 Alpine is on 3.5 2021-03-05 19:55:35 I lost remote on the server, so I'm not sure what version the acf files are 2021-03-05 19:56:13 uhM 2021-03-05 19:56:29 we enabled the backlog for lbu; I'm assuming that's not a problem 2021-03-05 19:56:49 3.5 is absolute EOL 2021-03-05 19:57:00 https://alpinelinux.org/releases/ 2021-03-05 19:57:04 I know :P 2021-03-05 19:57:18 you should upgrade for security 2021-03-05 19:57:27 but trying to fix a problem there is a waste of time... 2021-03-05 19:57:33 absolutely 2021-03-05 19:57:36 also 2021-03-05 19:57:43 we'll wrap up to the latest and greatest 2021-03-05 19:57:55 is there some problem with upgrading it? 2021-03-05 19:58:02 but we've had issues doing an in-place kernel update on the Pi run-from-ram mode 2021-03-05 19:58:48 I see... 2021-03-05 19:58:53 no problem with making a new build from scratch though; it's just a hassle swapping out SD cards in dozens of units 2021-03-05 19:59:44 and your problem is that the 'Local Backup' page is not loaded? 2021-03-05 19:59:46 we also have some customers who are version-change adverse, so there's that too ;) 2021-03-05 19:59:54 yeah 2021-03-05 19:59:58 every other page loads fine 2021-03-05 20:00:41 but loads something? the menu options, design...? 2021-03-05 20:01:37 yes 2021-03-05 20:01:43 so here's the landing page 2021-03-05 20:01:44 https://i.imgur.com/7IkQRKG.png 2021-03-05 20:01:55 About to click... 2021-03-05 20:01:57 https://i.imgur.com/NT4pCqH.png 2021-03-05 20:02:46 Stuck loading "list of changes"? 2021-03-05 20:02:47 https://i.imgur.com/m0wZdOs.png 2021-03-05 20:02:52 if you look top is something running? 2021-03-05 20:03:58 maybe? 2021-03-05 20:03:58 https://i.imgur.com/ZhhNnAa.png 2021-03-05 20:04:38 am I just out of RAM? 2021-03-05 20:06:35 see 'dmesg' 2021-03-05 20:06:51 if you are out of RAM you should see some OOM messages 2021-03-05 20:07:42 hmm, no, but lots of fsck errors 2021-03-05 20:08:07 uhM.. 2021-03-05 20:08:50 try "dmesg -w" and refresh the page 2021-03-05 20:09:04 if you see new errors probably thats your problem 2021-03-05 20:09:29 busybox dmesg does not have a -w option :) 2021-03-05 20:09:50 I got the idea though 2021-03-05 20:10:59 hehe 2021-03-05 20:11:19 is lbu working fine with command line? 2021-03-05 20:11:28 yes, that works fine 2021-03-05 20:12:00 and the apkovl isn't terribly large 2021-03-05 20:14:16 maybe you can try to comment some lines on acf files to see if something is breaking it 2021-03-05 20:14:40 on the lua source files? 2021-03-05 20:14:46 I gonna check on last version 2021-03-05 20:14:51 it should be pretty similar 2021-03-05 20:19:18 what package is acf-lbu? 2021-03-05 20:19:42 acf-alpine-conf 2021-03-05 20:19:52 ok 2021-03-05 20:20:15 perfect loading here :D 2021-03-05 20:20:28 could try loading 2021-03-05 20:20:34 https://alpine/cgi-bin/acf/alpine-conf/lbu/status 2021-03-05 20:20:37 Works for Me (tm) 2021-03-05 20:20:51 ok 2021-03-05 20:21:04 checking 2021-03-05 20:22:14 if works try 2021-03-05 20:22:16 https://alpine/cgi-bin/acf/alpine-conf/lbu/config 2021-03-05 20:22:45 and probably it fails with https://alpine/cgi-bin/acf/alpine-conf/lbu/listchanges 2021-03-05 20:24:18 did you check /var/log/acf.log? 2021-03-05 20:24:29 status and listchanges doesn't work at all, config and expert shows the menu and then nothing 2021-03-05 20:24:53 it's empty 2021-03-05 20:25:00 is there a flag to enable logging? 2021-03-05 20:25:18 IDK 2021-03-05 20:26:08 haha 2021-03-05 20:26:10 sigh 2021-03-05 20:26:25 is rare that status load fails 2021-03-05 20:27:15 yeah, it's not something I expected would fail either 2021-03-05 20:28:41 here /usr/share/acf/app/alpine-conf you have all the source 2021-03-05 20:30:02 I suppose that some func in '/usr/share/acf/app/alpine-conf/lbu-model.lua' is hanging 2021-03-05 20:30:46 status only calls getstatus() 2021-03-05 20:32:02 it calls getconfig()... 2021-03-05 20:32:36 try commeting some code 2021-03-05 20:32:48 if it tries to run some undefined func it should log it in /var/log/acf.log 2021-03-05 20:33:13 I'm sorry but I have to go, good luck Guest49 2021-03-05 20:33:17 thanks 2021-03-05 20:33:22 thank you for your help 2021-03-05 20:33:47 np, try to upgrade :P 2021-03-05 22:46:35 Hi all. When pulling the 3.13 image from dockerhub, I get a message that no trust is available (using DOCKER_TRUST=1). I don't have this issue with older images. What am I doing wrong? 2021-03-06 11:10:34 Hey there, I'm stuck on something. I got a fitlet2 mini-PC with Alpine Linux on it, and I need to use serial to admin it. BIOS, GRUB, printk's and inittab work, which means I get the BIOS, GRUB, kernel messages and a login prompt on my serial terminal, but I'm missing a few seconds of output between the last kernel message and the login prompt, 2021-03-06 11:10:34 usually when Alpine would show OpenRC and what it's doing. After "Mounting root: ok.", I get nothing for a few seconds, when I think I should be seeing output from OpenRC. This is the kernel param I'm using to get serial output: "console=uart8250,mmio32,0x280000000,115200". 2021-03-06 12:19:28 do you have any other console parameters 2021-03-06 12:20:01 some output only goes to the first one listed (or maybe the last one) 2021-03-06 12:20:06 Nope. That's the only one 2021-03-06 12:20:18 whats the full grub line? 2021-03-06 12:20:36 usually there are three stages: bios, kernel, userland 2021-03-06 12:21:16 one option is to just use rc_log 2021-03-06 12:22:11 Here's the GRUB line: /boot/vmlinuz-lts root=UUID=5b328069-ab55-4da3-beba-e7ca42b199b2 ro modules=sd-mod,usb-storage,ext4 rootfstype=ext4 console=uart8250,mmio32,0x280000000,115200 2021-03-06 12:23:24 trying to build a package with meson 2021-03-06 12:23:32 dependency "cairo" not found tried pkgconfig and cmake 2021-03-06 12:23:34 but i have cairo installed 2021-03-06 12:23:39 any other packages i have to install? 2021-03-06 12:24:13 ... cairo-dev? 2021-03-06 12:26:39 hello 2021-03-06 12:27:16 I have a huge problem with Alpine 3.13 into QEMU, my network does not go up 2021-03-06 12:27:31 the interface is set up as a tap bridge 2021-03-06 12:27:50 Hello71 now its "wayland-client" not found 2021-03-06 12:28:04 what kills me is that when I boot the same exact qemu with windows 7/8/10 the interface works perfectly 2021-03-06 12:28:32 but with alpine I get ioctl 0x8914 faild: Address not available 2021-03-06 12:28:37 deldenzen: yes... and? 2021-03-06 12:28:53 Wyk72: this is probably unrelated 2021-03-06 12:28:56 and i installed wayland-dev for that 2021-03-06 12:28:59 works now 2021-03-06 12:29:04 thanks for the help 2021-03-06 12:29:08 did you run setup-alpine 2021-03-06 12:29:12 same with ubuntu, the iterface is down 2021-03-06 12:29:15 yep as usual 2021-03-06 12:29:30 I can't find the cause of the problem 2021-03-06 12:30:36 can you point me to something I can do to diagnose the problem ? 2021-03-06 12:30:53 the module seems ok (virtio_net) 2021-03-06 12:31:02 I got no error in dmesg or elsewhere 2021-03-06 12:31:08 but the iterface does not go up 2021-03-06 12:32:43 ip link reports interface as qdisc noop state DOWN qlen 1000 2021-03-06 12:32:49 but does not go up 2021-03-06 12:35:35 my host is an alpine-edge machine running various quemu virtual machines 2021-03-06 12:36:01 windows VM with latest virtio drivers works fine 2021-03-06 12:36:29 firmware problem? 2021-03-06 12:36:36 did you install linux-firmware 2021-03-06 12:36:58 I am using alpine-virt, but tried alpine-standard too (3.13) 2021-03-06 12:37:20 I used versions 3.10, 3.11, 3.12 never a problem 2021-03-06 12:37:29 I am puzzled 2021-03-06 12:42:43 even 3.10-virt  reports "ioctl 0x8914 failed: address not available" 2021-03-06 12:46:46 if I use vmxnet3 driver, the interfaces seems up but does not communicate at all 2021-03-06 13:02:32 when did it work last time? 2021-03-06 13:13:54 Wyk72: virtio-net-pci 2021-03-06 13:14:23 Wyk72: what arch and distro is your guest 2021-03-06 13:16:44 alpine linux edge x86_64 2021-03-06 13:16:59 the device is passed as a tap/bridge 2021-03-06 13:18:22 the incredible thing is that in windows guest network is working but  linux guest (any distro) is not 2021-03-06 13:18:52 look here for hints https://arvanta.net/alpine/install-arm-under-qemu 2021-03-06 13:19:29 NETWORK="-netdev tap,id=hostnet0,ifname=alp,script=no,downscript=no -device virtio-net-device,netdev=hostnet0,id=net0,mac=52:55:00:d1:53:01" 2021-03-06 13:20:21 exactly as I defined it 2021-03-06 13:20:52 MAC=$(cat /sys/class/net/eth0/address | sed -r "s/^(.{0})(.{8})/\cd:44:5c/") 2021-03-06 13:20:52 OPTS="$OPTS -netdev tap,id=vireth1,script=no,downscript=no,ifname=alpeth" 2021-03-06 13:20:53 OPTS="$OPTS -device virtio-net-pci,netdev=vireth1,mac=$MAC" 2021-03-06 13:21:14 in windows 7/10 guests the bridge is ok, in linux is not 2021-03-06 13:21:54 maybe I'm missing something in QEMU packages ?? 2021-03-06 13:22:15 still amazes me that windows is working 2021-03-06 13:27:05 Wyk72: guest bridge? what about host bridge? 2021-03-06 13:28:07 host bridge is defined in /etc/network/interfaces as 2021-03-06 13:28:10 auto br0 2021-03-06 13:28:11 iface br0 inet static 2021-03-06 13:28:11         address 192.168.111.126 2021-03-06 13:28:11         netmask 255.255.255.0 2021-03-06 13:28:12         gateway 192.168.111.1 2021-03-06 13:28:12         bridge_ports eth0 2021-03-06 13:28:13         bridge_stp 0 2021-03-06 13:28:35 and is working 100% with windows guests (using it right now) 2021-03-06 13:29:02 I add the tap device into the bridge, and i pass it into qemu 2021-03-06 13:29:19 root@imm-prototype:~# brctl show 2021-03-06 13:29:20 bridge name     bridge id               STP enabled     interfaces 2021-03-06 13:29:20 br0             8000.2cf05d5d3108       no              eth0 2021-03-06 13:29:21                                                         alpeth 2021-03-06 13:29:21                                                         w10eth 2021-03-06 13:29:22                                                         w7eth 2021-03-06 13:30:05 Wyk72: for more than 4-5 lines please use some text mode paste service, tpaste.us or ix.io for example 2021-03-06 13:30:15 ok sorry 2021-03-06 13:30:55 hmm, have no idea, everything looks ok 2021-03-06 13:31:11 me too :) 2021-03-06 13:31:24 used to work 2021-03-06 13:31:38 some obscure parameter in the kernel ? 2021-03-06 13:31:56 one thing is different from before 2021-03-06 13:32:10 hardware platform is AMD and not Intel as it used to be 2021-03-06 13:32:14 never had this problem and using alpine with tap and bridge for more than 3 years on hosts and guests 2021-03-06 13:32:40 I'll try changing emulated cpu 2021-03-06 13:33:02 I use on intel and arm64 guests without problem 2021-03-06 13:33:04 when using with host bridge then need "echo" those 3 things: https://wiki.alpinelinux.org/wiki/Bridge#Little_script_to_allow_dhcp_over_iptables 2021-03-06 13:33:53 bridge-nf-call-* 2021-03-06 13:33:53 this shouldn't be reason to interface is not up 2021-03-06 13:34:08 s/to/why/ 2021-03-06 13:47:15 if I define it with -nic bridge,model=virtio-net-pci 2021-03-06 13:47:17 it works! 2021-03-06 13:47:46 it makes no sense to me 2021-03-06 13:57:37 qemu bug I guess 2021-03-06 14:02:05 hello 2021-03-06 14:02:09 did anyone miss me 2021-03-06 14:02:21 :D 2021-03-06 14:02:38 ikke: mps: 2021-03-06 14:02:57 oneinsect: hey, welcome back 2021-03-06 14:03:07 been a long time 2021-03-06 14:03:24 hope all are safe and sound 2021-03-06 14:03:27 thanks 2021-03-06 14:09:58 i don't think there's such thing as -nic model=virtio-net-pci 2021-03-06 14:14:58 I have it into qemu 5.2.0, and it's working 2021-03-06 14:15:24 qemu-system-x86_64 -nic model=? 2021-03-06 14:15:27 reports is 2021-03-06 14:15:37 it 2021-03-06 14:38:17 hm. i guess they added it 2021-03-06 14:53:26 i'm sorry 2021-03-06 14:56:05 i'm trying to get weston working on a minimum alpine installation on a raspberry pi 3b+, but seems like weston does not have the permission to read /dev/input events, with ls it says that only root has the permission. On internet i have found that on systemd systems the user have to be in input group. 2021-03-06 14:56:43 What is the equivalent for openrc systems? Or maybe, have i miss to install something related to input devices? 2021-03-06 14:56:44 need udev 2021-03-06 14:56:54 apk add udev && setup-udev 2021-03-06 14:57:24 oh, already installed udev, but i miss the setu-udev, i try it now 2021-03-06 14:57:51 maybe need to reboot after or udevadm trigger or something 2021-03-06 14:57:54 reboot is probably easier 2021-03-06 14:58:01 if it doens't work right away 2021-03-06 14:58:07 you also need to add to input group 2021-03-06 14:58:25 or install elogind, then you don't need input group 2021-03-06 14:58:27 same on systemd 2021-03-06 15:02:34 Hello71: you was absolutely right, seems that i had installed libudev, not udev. Installed it, runned setup-udev and rebooted. Now works perfectly, thanks! 👍️ 2021-03-06 17:44:41 Hello. First, thank you for making this great distro. Second, does anyone know how to start OpenRC services from an eudev rule? 2021-03-06 17:45:48 I tried RUN+="/sbin/rc-service transmission-daemon start". It works when I hot plug the usb drive, but fails to start when the drive is already connected on boot. 2021-03-06 17:47:16 The udev rule also mounts the drive, that works correctly even on boot. 2021-03-06 17:57:12 This is the udev rule https://pastebin.com/GHXqU8NH . I think the problem is that it doesn`t have networking yet, so the service crashes. How should I make it wait until network is up? 2021-03-06 18:02:09 oleo: maybe you can't specify it your service description, check depend() and tyr adding 'after net' 2021-03-06 18:02:26 ops, maybe you can* 2021-03-06 18:06:31 donoban: I changed ´need net´ to ´after net´ and rebooted but it didn´t work. 2021-03-06 18:29:56 uh oh 2021-03-06 18:30:31 if i do a google seach containing "alpine linux," i get results with "arch linux" bolded as if the are synonyms 2021-03-06 18:31:12 lol 2021-03-06 18:31:21 anyway, i was wondering if alpine ships zfs binaries? 2021-03-06 18:31:31 why don't add the service to default? 2021-03-06 18:31:49 seems that some distros are more careful than others about legal issues 2021-03-06 18:32:05 ubuntu thinks its safe to ship zfs binaries, debian doesn't 2021-03-06 18:32:09 does alpine have a position? 2021-03-06 18:32:15 IDK 2021-03-06 18:32:22 bleb: Yes, I´m using it right now. I think they´re in main. 2021-03-06 18:32:22 but it includes them 2021-03-06 18:32:56 oleo: is there something wrong adding your service to default? 2021-03-06 18:33:01 donoban: Because I wouldn´t like to start the service when the drive isn´t connected. 2021-03-06 18:33:20 cool 2021-03-06 18:33:29 maybe you can add some check the disk is mounted 2021-03-06 18:33:47 then when you connect the disk the service will be restarted with your rule 2021-03-06 18:33:55 debian for example has zfs-dkms which builds the kernel module locally 2021-03-06 18:34:18 i dont see any dkms stuff in the alpine zfs package so i guess it's an actual binary package 2021-03-06 18:34:18 so you have it started just by default on start, it the hard disk is ready nice, if not when you connect, it will be restarted again 2021-03-06 18:35:07 You´re talking about the service definition? Should I check it from there? 2021-03-06 18:35:20 bleb: some thing is the kernel module and other the tools binaries 2021-03-06 18:35:40 not, I'm talking about enabling the service "rc-update add your-service default" 2021-03-06 18:35:48 so it will be started automically on boot 2021-03-06 18:36:10 ah you mean if the hard disk is mounted 2021-03-06 18:37:13 donoban: do you know which package has the kernel module? 2021-03-06 18:37:46 So if I understand correctly, I can add to start(): if [ ! -e /media/usbdisk ]; then exit; fi 2021-03-06 18:38:14 localhost:/lib/modules$ find | grep zfs 2021-03-06 18:38:15 ./5.10.16-0-virt/extra/zfs 2021-03-06 18:38:17 ./5.10.16-0-virt/extra/zfs/zfs.ko 2021-03-06 18:38:26 I just installed, 'apk add zfs' 2021-03-06 18:38:42 it includes some depencies 2021-03-06 18:39:12 oleo: maybe there is some open-rc way but if it works fine :) 2021-03-06 19:01:11 donoban: `return 1` did the trick. Thank you very much! 2021-03-06 19:30:17 oh great :) 2021-03-06 19:58:20 im encountering issues booting 5.11 on a rockchip rk3399 2021-03-06 19:58:25 pmos boots fine though 2021-03-06 20:01:39 bl4ckb0ne: linux-lts or linux-edge, or custom build kernel 2021-03-06 20:01:53 linux-edge 2021-03-06 20:02:06 5.11.3 freshly installed 2021-03-06 20:02:11 though linux-lts 5.11 doesn't exist in alpine 2021-03-06 20:02:12 it hangs on "starting kernel" 2021-03-06 20:02:59 i copy pasted the pmos extlinux entry and replace the names with edge 2021-03-06 20:03:02 bl4ckb0ne: can you remove quiet and add 'loglevel=15' to boot loader cli params 2021-03-06 20:03:36 and btw, what board you use 2021-03-06 20:03:42 helios64 2021-03-06 20:03:59 everything should be in 5.11 2021-03-06 20:04:42 https://paste.sr.ht/~bl4ckb0ne/ac61437cee3e2ccc26c79797ac9d44a145cc3078 still hanging 2021-03-06 20:05:00 hmm, I didn't noticed anything new about this when I upgrade linux-edge to 5.11.0 2021-03-06 20:06:24 https://paste.sr.ht/~bl4ckb0ne/4d468913836a8f4f7a44715ae961bd81d5f464a4 2021-03-06 20:07:56 I see, that part looks ok 2021-03-06 20:08:15 dtb stolen from armbian 2021-03-06 20:08:21 ah 2021-03-06 20:08:34 we don't have dtb for this board? 2021-03-06 20:08:42 i dont think so 2021-03-06 20:08:58 oh yes there is 2021-03-06 20:09:10 I'll check in about hour when I come back to home 2021-03-06 20:09:21 then I think the one in the /boot folder is the good one, and the .dtb.old is the armbian one 2021-03-06 20:09:27 ah, good, don't have to hurry :) 2021-03-06 20:09:28 ill continue to poke around 2021-03-06 20:09:51 dtbs are in dtbs-edge 2021-03-06 20:09:55 dir 2021-03-06 20:10:01 yup its the one from dtbs-edge 2021-03-06 20:11:03 you copied it manually to / 2021-03-06 20:11:06 ? 2021-03-06 20:11:14 yep 2021-03-06 20:11:33 should I make it point to dtbs-edge in the extlinux.conf? 2021-03-06 20:12:44 I usually do 'FDTDIR /dtbs-edge' and u-boot find proper dtb and load it 2021-03-06 20:13:35 I will be afk for some time from now 2021-03-06 20:13:59 ack 2021-03-06 20:25:43 ayy, i'm trying to make a debian container on alpinelinux / in lxc, can't connect to the container 2021-03-06 20:27:24 Guest__: can you ellaborate? 2021-03-06 20:29:06 ERROR utils - utils.c:lxc_rm_rf:1813 - No such file or directory - Failed to open dir "/sys/fs/cgroup/openrc//lxc.payload.id-container" 2021-03-06 20:34:37 there's something edge does that pmos doesnt 2021-03-06 20:34:46 `Moving Image from 0x2080000 to 0x2200000, end=33d0000` 2021-03-06 20:35:55 how do i run lxc debian container ? 2021-03-06 20:37:40 ikke ^^ 2021-03-06 20:39:21 unpriviliged containers, and i think i have problems with systemd, it tries to run openrc 2021-03-06 20:41:35 mps: `fdtdir /dtbs-edge/rockchip/` gives this 2021-03-06 20:41:39 Retrieving file: /dtbs-edge/rockchip/rockchip/rk3399-kobol-helios64.dtb 2021-03-06 20:41:42 Failed to load '/dtbs-edge/rockchip/rockchip/rk3399-kobol-helios64.dtb' 2021-03-06 20:41:44 Skipping EDGE for failure retrieving fdt 2021-03-06 20:41:49 oh wait nvm 2021-03-06 20:42:54 still hanging at "starting kernel" 2021-03-06 20:47:24 trying to start a debian container but it fails, look above for ERROR log 2021-03-06 21:06:08 i assume the problem is diskspace, i run tmpfs 2021-03-06 21:10:24 bl4ckb0ne: `fdtdir /dtbs-edge/` not `fdtdir /dtbs-edge/rockchip/` 2021-03-06 21:10:48 yup i noticed my mistake 2021-03-06 21:12:30 bl4ckb0ne: looks like we are missing serial driver for this SBC or it is built as module 2021-03-06 21:13:29 have to find diff with pmos kernel and look what should be enabled on linux-edge 2021-03-06 21:13:45 i ran the diff with the one from armbian 2021-03-06 21:14:12 hm, maybe this can help also 2021-03-06 21:14:14 https://paste.sr.ht/~bl4ckb0ne/e960c378cc8c41f0970f3456d3a7c025d89570d4 2021-03-06 21:16:41 hmm, this is armbian 5.10 2021-03-06 21:16:56 could you make same with pmos kernel 2021-03-06 21:17:09 https://paste.sr.ht/~bl4ckb0ne/05831ef50289cad89c6a8ab5102eb8277e643431 2021-03-06 21:17:30 the pmos kernel is old though 2021-03-06 21:17:33 5.6.0 #1-postmarketos-rockchip 2021-03-06 21:18:00 ah 2021-03-06 21:18:12 hmm 2021-03-06 21:18:59 could you paste full config file from armbian to https::/tpaste.us 2021-03-06 21:19:05 sure 2021-03-06 21:20:14 https://tpaste.us/ra1L 2021-03-06 21:20:21 that paste tool is neat 2021-03-06 21:21:01 yes, and alpine official :) 2021-03-06 21:21:12 nice 2021-03-06 21:30:28 bl4ckb0ne: when you boot with armbian can you look what driver is used for serial console interface 2021-03-06 21:31:10 sure 2021-03-06 21:31:33 dmesg and look for 'Serial: 8250/16550 driver, 4 ports, IRQ sharing disabled' and probably after this few lines 2021-03-06 21:32:05 looks line we didn't enabled serial driver for this board, but not sure 2021-03-06 21:35:56 2.564073] Serial: 8250/16550 driver, 8 ports, IRQ sharing disabled 2021-03-06 21:35:59 [ 2.567463] ff180000.serial: ttyS0 at MMIO 0xff180000 (irq = 44, base_baud = 1500000) is a 16550A 2021-03-06 21:36:01 [ 2.568513] ff1a0000.serial: ttyS2 at MMIO 0xff1a0000 (irq = 45, base_baud = 1500000) is a 16550A 2021-03-06 21:36:03 [ 2.568795] printk: console [ttyS2] enabled 2021-03-06 21:36:05 [ 2.570235] Serial: AMBA driver 2021-03-06 21:36:07 that? 2021-03-06 21:36:29 bl4ckb0ne: also, maybe try to change console in u-boot append cli params to 'console=ttyAMA0,115200n8' 2021-03-06 21:36:53 ahm 2021-03-06 21:37:02 looks ok 2021-03-06 21:38:30 maybe then 'onsole=ttyS0,115200n8' 2021-03-06 21:39:15 but this depends on which physical port console it 'routed' on this SBC 2021-03-06 21:40:16 bl4ckb0ne: there is no name of driver in next few lines in dmesg output? 2021-03-06 21:42:38 though it looks it is 'old good' 16550A which is strange 2021-03-06 21:53:24 dumb question - what does this mean? why is apk unable to select packages? https://gist.github.com/LAC-Tech/74e6eea674db556775ed9a2c8d8e56b4 2021-03-06 21:56:11 LACampbell: that there is no package in the repositories that provides so:libdruntime-ldc-shared.so.90 2021-03-06 21:57:09 that makes sense. so it's an issue with the package I am trying to install, in that asks for something that doesn't exist in repos? 2021-03-06 21:58:20 yes 2021-03-06 21:58:32 it might be that it needs to be rebuilt 2021-03-06 21:59:58 (I've been very careful this time round to not use any weird repos, just main and community. got myself into trouble with that first time round with alpine) 2021-03-06 22:00:07 ok I'll send the package maintainer a polite message 2021-03-06 22:42:50 anyone else running alpine with an amd rx 5700xt? 2021-03-06 22:44:53 mps: https://tpaste.us/6Pml 2021-03-06 22:46:06 I have `console=ttyS2,1500000` already 2021-03-06 22:58:06 bl4ckb0ne: this boot msg can be useful to look, but I'm too tired and going to bed now, will look tomorrow if I can conclude what must be enabled for linux-edge 2021-03-06 23:36:31 ty for your help mps, let me know if I can give you any help 2021-03-06 23:57:18 bl4ckb0ne: what is your issue again? 2021-03-07 01:43:16 Hello71: linux-edge doesnt boot on my rk3399 2021-03-07 01:43:21 pmos boots fine 2021-03-07 02:07:07 Alright, I'm sober tonight and I'm giving this another shot.   Hopefully I can get alpine to boot from my flash drive. 2021-03-07 02:13:31 If anyone is here, I'm using Rufus to write the iso to usb.  Should I use the "iso" mode or the "dd" mode? 2021-03-07 02:21:59 Far as I can get is a standard isolinux boot prompt and the system just hangs.  nothing happens.  Tried multiple drives,  multiple write methods,  uEFI vs BIOS,  x86 vs x64.  Same story.  Either does nothing or locks up at the isolinux boot prompt. 2021-03-07 02:22:04 idk what to do. 2021-03-07 02:45:03 dd, uefi 2021-03-07 02:45:29 also are you using 3.13 standard 2021-03-07 02:45:38 3.13.2, yes 2021-03-07 02:46:42 My device REQUIRES  secureboot.  I add bootia32.efi to the allowed list.  When isolinux eventually pops up, I get a newline then nothing.   The usb drive falls asleep and the machine is locked up. 2021-03-07 02:48:41 I'm trying "etcher" now instead of rufus as I have no clue what else to try. 2021-03-07 02:53:49 Are alpine mirror updates non-atomic? 2021-03-07 02:54:45 Bwriting with etcher, my EFI lets me add the bootia32.efi to the allow list, then when booting, it claims the file doesn't exist or is not a valid EFI file. 2021-03-07 02:54:51 lolwut 2021-03-07 03:08:40 Trying it via legacy/bios and it tells me "no bootable device".  lol    I've never had this much trouble booting anything before. 2021-03-07 03:21:53 I managed, finally, to boot. What a trip. 2021-03-07 03:35:16 Ok, it's installed. When running setup-interfaces and typing in wlan0(the ONLY interface) I get two big error lines about "linux-firmware-mellanox-xxxxx" and "linux-firmware-liquidio-xxxx" files missing and it simply repeats the question.  Is there an easy fix for this? 2021-03-07 04:09:09 I guess I have to reinstall. blah 2021-03-07 06:10:05 ~ 2021-03-07 06:32:46 ikke: morning. 2021-03-07 06:33:21 are there any updates to the lvm-scanning problem? I'd like to go to 3.13 with the kernel, but this is as of yet not possible. 2021-03-07 06:50:37 hey guys 2021-03-07 06:50:40 can someone point me to some link/docs that shows the boot process for alpine? What scripts/how does it set things up with the lbu and diskless and stuff like that? 2021-03-07 07:42:36 DaRock: https://gitlab.alpinelinux.org/alpine/mkinitfs/-/blob/master/initramfs-init.in starts things up 2021-03-07 09:30:47 artok: thx or that 2021-03-07 09:31:00 now I see why I couldn't find it :-) 2021-03-07 09:48:04 Hello, 2021-03-07 09:48:04 I'm trying to implement "data disk" installation with /var mounted as zfs volume. 2021-03-07 09:48:05 I'm using https://wiki.alpinelinux.org/wiki/Root_on_ZFS_with_native_encryption for zfspool creation and it works, but after reboot I got this "filesystem cannot be mounted unable to open dataset" and empty zpool list. Looks like  it is a bita late to mount zfs volume 2021-03-07 09:48:05 Zfs was added to fstab like this: 2021-03-07 09:48:06 echo "rpool/ROOT/alpine-var /var zfs defaults 0 0" >> /etc/fstab 2021-03-07 09:48:06 Can I ask for help with configuring /var as encrypted zfs (I can hardcode password into initfs as whole memory-based Alpine is on removable storage). 2021-03-07 09:48:07 Thanks 2021-03-07 12:23:38 moin, I'm using an alpine rescue disk (ramdisk) to sort out a zfs boot problem. is it possible to load zfs stuff via modprobe, from the ramdisk? 2021-03-07 12:23:52 I could be missing something very obvious, this is what I've done so far - https://dpaste.org/3zwv 2021-03-07 12:23:58 - add a mirror 2021-03-07 12:24:05 - add the zfs bits 2021-03-07 12:24:51 - try modprobe, get `modprobe: module zfs not found in modules.dep` error 2021-03-07 12:28:04 I'm guessing this is something to do with readonly / but tbh I have no idea 2021-03-07 12:29:00 dch: zfs is not part of the standard kernel modules 2021-03-07 12:30:37 ikke: right, so when I `apk add util-linux udev zfs` that's not sufficient? 2021-03-07 12:32:52 dch: you need zfs-lts as well: https://pkgs.alpinelinux.org/package/edge/main/x86_64/zfs-lts 2021-03-07 12:33:05 but on a run-from-ram system this is a bit more involved 2021-03-07 12:34:13 indeed 2021-03-07 12:34:31 it might be easier to nuke and pave (i have backups) 2021-03-07 12:35:07 doesn't alpine-extended have zfs? 2021-03-07 12:35:29 https://gitlab.alpinelinux.org/alpine/aports/-/blob/master/scripts/mkimg.standard.sh#L33 2021-03-07 12:35:35 unfortunately I don't get to choose the rescue image 2021-03-07 12:36:18 Hello71: but is it included in the modloop? 2021-03-07 12:36:20 you need to build the exact right version of zfs-lts then 2021-03-07 12:36:41 ikke: i mean i assume it works properly 2021-03-07 12:36:56 since a number of people use zfs on alpine 2021-03-07 12:37:07 zfs on root on alpine 2021-03-07 12:37:15 right, kernel-addons 2021-03-07 12:42:06 mmm sounds like a nuke&pave is easiest. thanks ikke for helping me understand it a bit better 2021-03-07 12:44:54 you could also kexec to alpine-extended but that's probably more complicated 2021-03-07 13:20:30 Hello71: good idea, I will take a look at that next time rougn 2021-03-07 13:20:32 *round 2021-03-07 13:21:30 I think it's easier to just build zfs-lts 2021-03-07 14:04:26 Hello again, Now I can mount zfs dataset as var, but there is still problem with encryption. Got error on boot: 2021-03-07 14:04:26 filesystem 'rpool/ROOT/alpine-var' can not be mounted: Permission denied 2021-03-07 14:04:27 # zfs get keylocation 2021-03-07 14:04:27 NAME PROPERTY VALUE SOURCE 2021-03-07 14:04:28 rpool keylocation file:///etc/zfs/key.dat local 2021-03-07 14:04:28 rpool/ROOT keylocation none default 2021-03-07 14:04:29 rpool/ROOT/alpine-var keylocation none default 2021-03-07 14:05:00 rpool/ROOT/alpine-var mounted using fstab 2021-03-07 14:05:47 I will be grateful for hint how to debug mysterious "Permission denied" 2021-03-07 14:39:42 upd looks like I need to execute zfs load-key -a on boot before fstab is parsed, 2021-03-07 14:39:43 what will be the best way to do it? 2021-03-07 16:28:05 Hello folks, was looking into apk-tools (apk) documentation trying to find where packages are downloaded and only saw --cache-dir location (default to /var/cache/apk). Trying to see if I can reuse downloads (index + pkgs) between builds (CI). Any advice? Thank you! 2021-03-07 16:29:00 luislavena: apk --policy pkgname 2021-03-07 16:33:18 @mps, thank you, is that command right? `--policy` doesn't work and `apk policy pkgname` only outputs if the package is installed and the URL where to get it, but not where the package is downloaded before installation. Perhaps apk just uses `/tmp` and doesn't reuse already downloaded pkgs (outside of `apk cache`) 2021-03-07 16:33:30 What is the best / simplest service monitoring for openrc? 2021-03-07 16:34:29 Say you mostly run docker and you don't want to miss anything that typical systemd usage would have helped people catch (considering that I hear openrc isn't "as good for" service monitoring - but I bet it can achieve the same reliability with the right engineering). 2021-03-07 16:38:07 luislavena: apk doesn't keep such info, though it will show in case of multiple sources from which pkg is installed 2021-03-07 16:38:28 and yes, 'apk policy ...' without -- 2021-03-07 16:40:19 mps, thank you. was trying to see if I can reduce docker build time by not only caching /var/cache/apk indices but also downloaded .apk files. Will continue to eat bandwidth :D. Thank you for the promptly response! 2021-03-07 19:18:31 message to me form the future: to configure automount with decryption zfs mount with "-l" arg should be used 2021-03-07 19:18:32 sed -i 's/MOUNT_EXTRA_OPTIONS=""/MOUNT_EXTRA_OPTIONS="-l"/' /etc/conf.d/zfs #load key on mount 2021-03-07 23:27:36 What is this about ? apk add --no-cache --virtual .build-deps 2021-03-07 23:27:46 .build-deps ? 2021-03-08 01:59:00 mps: no luck booting the 5.11.4 2021-03-08 02:58:09 Ok, I managed to get sober for a little bit  again and I got it booted and installed.  Now, how do I ensure that wpa_supplicant and wireless-tools are installed to the HDD "before" I reboot.?  WiFi is the only interface on this machine, sadly. 2021-03-08 02:59:38 Actually no, the installer screwed up on as few files, but didn't bother trying again.  I guess I need to install the 2 things is missed somehow. 2021-03-08 03:04:24 I need an "apk add --root /mnt " flag.  I'm stuck again. 2021-03-08 03:09:41 I think I figured it. Had to chroot. :D 2021-03-08 03:47:31 Hello, just a quick question. I have been reading about musl and locales, but I also just found out about the workarounds for glibc-dependent programs. So, question: is there a workaround for having a translated system? 2021-03-08 03:48:36 Apologies if this has been asked many times 2021-03-08 06:50:37 Hey, been a while since I was on here (my IRC bouncer died and I haven't been bothered to fix it), would anyone happen to know why trying to bootstrap libssh2 results in it not finding openssl? I did notice this in the config.log: `/usr/lib/gcc/powerpc-alpine-linux-musl/10.2.1/../../../../powerpc-alpine-linux-musl/bin/ld: /usr/lib/libssl.so: error adding symbols: file in wrong format` 2021-03-08 06:50:58 Seems like it's trying to use the host libraries rather than the ones in ~/sysroot-ppc but I have no idea why 2021-03-08 06:54:40 bl4ckb0ne: hm, those options/drivers enabled on upgrade didn't helped then. do you know if there is somewhere guide what should be enabled for this board to boot 2021-03-08 07:00:28 bl4ckb0ne: sorry I forgot what is your board, if you tell me I can try to search around what must be enabled for serial console to work 2021-03-08 14:49:03 <[diablo]> good afternoon #alpine-linux .. guys I'm running Alpine 3.13.2 as a Docker host, and have noticed that when I delete a container it's not removing the IPtables rules for the port mapping. It creates them without issue, but doesn't delete.. any ideas please? #docker said it should, and referred me here 2021-03-08 15:22:04 mps: https://wiki.kobol.io/helios64/hardware/ 2021-03-08 15:24:08 https://github.com/search?p=1&q=org%3Aarmbian+helios64&type=Code there's a bunch of patch for the board in armbian repos though 2021-03-08 15:24:46 https://elixir.bootlin.com/linux/v5.11.4/source/arch/arm64/boot/dts/rockchip/rk3399-kobol-helios64.dts dts is in mainline 2021-03-08 15:26:34 https://wiki.kobol.io/helios64/troubleshoot/#serial-console-does-not-appear 2021-03-08 15:26:46 fdti driver 2021-03-08 15:28:21 will ask on #linux-rockchip if someone knows what is needed in kernel for serial console to work 2021-03-08 15:31:59 is the hang at "starting kernel" related to the serial console? 2021-03-08 15:32:30 bl4ckb0ne: I think so, though not sure 2021-03-08 15:33:37 your u-boot config looks ok, it loads dtb and kernel without errors 2021-03-08 15:33:38 i tried putting the log level to 7 without success 2021-03-08 15:33:52 i also tried earlyprintk without success 2021-03-08 15:34:10 fwiw the uboot config is not upstream, they have a bunch of patch 2021-03-08 15:34:29 if the driver is not active then kernel will not show anything 2021-03-08 15:34:54 I think your u-boot is ok 2021-03-08 15:35:30 'Starting kernel' is last message from u-boot and after that all is 'on' kernel 2021-03-08 15:38:42 i could give another try to earlyprintk 2021-03-08 15:41:56 bl4ckb0ne: going to lunch, will ask later on #linux-rockchip about this 2021-03-08 15:42:17 ty 2021-03-08 17:18:38 [diablo]: alpine doesn't manage iptables by default 2021-03-08 17:18:49 check your docker logs 2021-03-08 17:21:00 rodper45: musl doesn't really support C locales that well at the moment. for messages there is gettext-tiny 2021-03-08 17:21:13 but it doesn't handle LC_NUMERIC, LC_PAPER, etc 2021-03-08 17:21:19 which are arguably misdesigned in the first place 2021-03-08 17:21:53 thanks 2021-03-08 17:23:13 i'm not sure if gettext-tiny requires recompiling the programs 2021-03-08 18:37:09 bl4ckb0ne: I got answer what should be enabled for serial console to work on helios64 2021-03-08 18:37:56 are you ready to try boot kernel not from official alpine repo but one which I can upload to my dev.alpinelinux.org dir 2021-03-08 18:39:10 if it works I will upgrade official also 2021-03-08 18:39:53 builders are currently loaded 2021-03-08 19:01:29 https://bpa.st/ITSA Building wine in a 32-bit chroot, I get this segfault. How do I debug a file that isn't a compiled program, or at least get more information about why this is failing? 2021-03-08 20:29:20 mps: sure 2021-03-08 20:33:58 bl4ckb0ne: ok, I will build it and put on dev.a.o and inform you, I think build will finish in about one hour 2021-03-08 20:35:27 we don't have CONFIG_SERIAL_8250_DW=y in config, which is needed 2021-03-08 20:37:59 i can also build it on my side if you want 2021-03-08 20:40:19 ah, no problem and even better if you can 2021-03-08 20:40:58 you mean build linux-edge apk? 2021-03-08 20:41:35 yeah 2021-03-08 20:41:46 go ahead :) 2021-03-08 20:42:22 is the serial config the only one rquired? 2021-03-08 20:42:31 set CONFIG_SERIAL_8250_DWLIB=y and CONFIG_SERIAL_8250_DW=y 2021-03-08 20:42:59 that is what rockchip developer told me 2021-03-08 20:43:19 and that makes sense 2021-03-08 20:46:14 can i just edit the file by hand or its better to run make menuconfig? 2021-03-08 20:46:57 whatever you prefer 2021-03-08 20:48:07 though I usually use make (some)config to check if something else depends or advised when changing options 2021-03-08 20:49:00 should I edit both config (regular and virt) ? 2021-03-08 20:49:37 only regular, virt use AMBA driver for serial 2021-03-08 20:54:22 abuild -r started 2021-03-08 20:55:03 gah perms on /var/cache/distfiles 2021-03-08 20:56:56 started for real this time 2021-03-08 20:57:06 i'll ping you when its built/tested 2021-03-08 21:43:53 hello guys. 2021-03-08 21:45:55 and gals 2021-03-08 21:46:11 where read guidelines to commit a update? 2021-03-08 21:46:26 hi ikke 2021-03-08 21:46:39 good to see you. 2021-03-08 21:52:01 https://wiki.alpinelinux.org/wiki/Creating_patches#Creating_a_merge_request 2021-03-08 22:38:26 bl4ckb0ne: on my side kernel is built, but I'm going to bed. I can upload apk tomorrow if you got problems with build 2021-03-08 23:30:13 has anyone the isse that Alpine doesn't load intalled PHP modules ? 2021-03-08 23:30:37 gah timed out 2021-03-08 23:30:47 i was building through ssh 2021-03-08 23:31:15 i think ill take your apk then pls 2021-03-08 23:33:55 bovis: gdb --args perl winebuild ... 2021-03-08 23:36:14 Lutin: haven't had that issue 2021-03-08 23:36:26 ikke I have and I cannot figure out why 2021-03-08 23:36:38 How are you running it? 2021-03-08 23:37:40 huh weird the helios shut down 2021-03-08 23:39:48 ikke in dockr, jst apk add whatever module I need 2021-03-08 23:39:52 *docker 2021-03-08 23:43:23 ... did you edit the php.ini? 2021-03-09 00:04:00 Modules use drop-in files 2021-03-09 00:49:12 Hello71: Unsure exactly what you're looking for, but running "winebuild --def -E kernel32.spec" gives me this http://sprunge.us/ZaXkcS 2021-03-09 00:49:26 No errors 2021-03-09 01:03:41 Running "gdb --args perl winebuild -w --def -o libkernel32.def --export kernel32.spec" gives me the error: Can't open perl script "winebuild" 2021-03-09 03:01:46 Hello71: Potentially also helpful, though I don't know how much https://bpa.st/ZGKQ 2021-03-09 03:26:33 Ok now I'm seriously worried about the bootstrap script, I wasn't sure if my setup had managed to break so I redid my VM on latest edge, tried to bootstrap again and now GCC doesn't want to compile 2021-03-09 03:26:47 `/usr/lib/gcc/powerpc-alpine-linux-musl/10.2.1/../../../../powerpc-alpine-linux-musl/bin/ld: /usr/lib/gcc/powerpc-alpine-linux-musl/10.2.1/adalib/libgnat.a: error adding symbols: file in wrong format` 2021-03-09 03:34:17 I think it might have something to do with this commit, maybe how it does things breaks cross compilation somehow https://gitlab.alpinelinux.org/alpine/aports/-/commit/9809eecad1ab4bbc48d2a367ac8d785bb34b4b43 2021-03-09 04:10:31 Alright gcc compiled after reverting the commit, not sure what the right approach would be from there, presumably recompiling it on a native host 2021-03-09 04:29:53 do appimages work on alpine linux? 2021-03-09 04:30:00 if they dont, how can i make them work? 2021-03-09 04:34:17 Most of them depend on glibc which means they don't work because Alpine uses musl 2021-03-09 04:57:24 thats sad 2021-03-09 06:22:53 Alright cleaning everything up and reverting that commit seems to have fixed the bootstrap, it's up to building the kernel now 2021-03-09 06:22:55 That's annoying 2021-03-09 06:25:01 Segfault0: are you cross-compiling? 2021-03-09 06:25:27 Yes this is redoing the bootstrap for ppc to get everything up to date so I can have another go 2021-03-09 06:26:32 aports needs a few changes, mostly to the bootstrap script itself, to be able to cross-build for PPC, although this issue with gcc doesn't look specific to ppc so it might be something to look at 2021-03-09 07:50:17 bl4ckb0ne: https://dev.alpinelinux.org/~mps/linux-edge-5.11.4-r0.apk 2021-03-09 09:36:55 Hi. Is it possible to install Alpine 3.11.3 without updating anything? I want that version of the OS, exactly. 2021-03-09 09:37:07 Thus far when I’ve tried it updates to 3.11.8. 2021-03-09 10:06:44 nm … the issue I’m chasing happens on 3.11.8. Thank you ! 2021-03-09 10:32:02 Does anyone know how I get abuild to compile linux-virt? 2021-03-09 10:34:02 aaprts/main/linux-lts 2021-03-09 10:36:03 Yes but that's only compiling linux-lts, it's not also compiling the linux-virt sub package for some reason 2021-03-09 10:37:42 it does 2021-03-09 10:38:00 Is there something in the APKBUILD that determines whether it does or not based on architecture? 2021-03-09 10:38:09 Because that's the only reason I can think of that it's not building linux-virt for me 2021-03-09 10:39:01 arch="xxxx" 2021-03-09 10:39:37 Segfault0: on which arch you are trying to build linux-virt 2021-03-09 10:39:42 ppc 2021-03-09 10:40:11 The bootstrap script is up to a point where it happily goes all the way through 2021-03-09 10:40:15 ppc64le? 2021-03-09 10:40:19 No, just ppc 2021-03-09 10:40:33 Except no linux-virt package build is ever attempted, it does build linux-lts 2021-03-09 10:40:39 you have to create config file for linux-virt.ppc 2021-03-09 10:40:42 I have 2021-03-09 10:42:37 That's why I'm confused about it not trying to build linux-virt 2021-03-09 10:42:39 hmm, 'ppc*) _carch="powerpc" ;;' 2021-03-09 10:43:33 not sure how it works in that case, but APKBUILD is shell script, you can look at it to find what is missing for ppc to build -virt 2021-03-09 10:44:02 btw, isn't 'ppc' too generic name for arch 2021-03-09 10:44:03 I wonder if somehow the bootstrap script is messing with things 2021-03-09 10:47:56 I looked through it and tried building the package separately from the bootstrap script and I can't see a reason why it's not building linux-virt 2021-03-09 10:49:07 Ah, I think I see now, virt is a different flavor, not a sub package 2021-03-09 10:49:13 So it needs an extra argument in abuild 2021-03-09 11:06:15 bl4ckb0ne: are you around? did you tried kernel which I posted on link above 2021-03-09 11:07:09 bl4ckb0ne: I'm preparing 5.11.5 upgrade and would like to add needed options if they work for you 2021-03-09 11:27:15 Oh, Segfault0 left 2021-03-09 11:27:19 It's not an argument 2021-03-09 11:32:01 afaik Segfault is 11 and not 0 :) 2021-03-09 11:37:20 :) 2021-03-09 11:54:05 Is the apk-autoupdate package supposed to update the system automatically, or do I have to run it myself? I have it installed, but I haven't seen it update the system by itself. 2021-03-09 11:54:27 And if it is supposed to update automatically, how can I set it up? 2021-03-09 11:54:46 TheUnknown67: You need to run it in a cron-job 2021-03-09 11:56:07 Do I have to manually set it up using crontab, or by making a separate file? 2021-03-09 11:56:57 either 2021-03-09 11:57:12 /etc/periodic/daily/* 2021-03-09 11:57:13 for example 2021-03-09 11:58:45 "The tool is intended to be executed automatically and regularly by cron or similar tool." 2021-03-09 11:58:50 https://github.com/jirutka/apk-autoupdate/ 2021-03-09 12:02:55 Ok. Thank you for your support. I appreciate it. 2021-03-09 12:35:01 <[diablo]> hey good afternoon #alpine-linux ... I've found a small(ish) issue with docker on Alpine... when I do apk info -a docker-engine it shows it has a dep on iptables... but in actual fact it also has one on ip6tables 2021-03-09 12:35:15 <[diablo]> otherwise when you destroy a container, it does not remove the iptables rules... 2021-03-09 12:35:28 <[diablo]> I also found when debugging docker daemon the following message 2021-03-09 12:35:40 <[diablo]> WARN[0000] Failed to find ip6tables: exec: "ip6tables": executable file not found in $PATH 2021-03-09 12:36:07 <[diablo]> once I added ip6tables , and then stop a container, the iptable rules do get removed ... 2021-03-09 14:49:28 mmm, i was wondering: can Alpine be runned on x86 with nvidia proprietary driver? Or more specific, can Alpine run on a optimus nvidia laptop with proprietary drivers? 2021-03-09 14:53:33 since buying new stuff is not not a possibility due to crazy prices, i was wondering if i can resurrect an old nvidia optimus laptop. I know that nvidia, and especially optimus, was a very bad choice, but at the time i was not entered in the linux world yet ^^ 2021-03-09 15:06:04 Uniformbuffer[m]▸ I believe the proprietary nvidia driver userspace is glibc only 2021-03-09 15:18:54 Hey, so I'm trying to compile something and it reports missing libs related to: libm, libpthread, librt, libssp_nonshared ... I don't see obvious packages to install in apk search... any hints? 2021-03-09 15:20:18 hm one package needed appears to be musl-dev, ok 2021-03-09 15:28:53 programmerq: yeah, i fear that answer 😢. Thanks anyway 2021-03-09 15:30:02 Uniformbuffer[m]: is nouveau enough? 2021-03-09 15:33:37 EF: unfortunately no. That laptop has a maxwell gpu, so no reclocking. Even if i belive Nuoveau developers are doing a great job with the "resources" that they got, using 10% of the power of the gpu seems a bit wasted for the paid price. Obviosly i will never buy anything from nvidia, but at the time i didn't know the implications. At the time i still was a window guy ^^ 2021-03-09 15:34:19 * EF: unfortunately no. That laptop has a maxwell gpu, so no reclocking. Even if i belive Nuoveau developers are doing a great job with the "resources" that they got, using 10% of the power of the gpu seems a bit wasted for the paid price. Obviously i will never buy anything from nvidia, but at the time i didn't know the implications. At the time i still was a window guy ^^ 2021-03-09 15:45:06 Uniformbuffer[m]: I tried hard few years ago (because I have two of these notebooks) but didn't succeed. problem is closed source parts from nvidia 2021-03-09 15:46:19 after that I simply disabled nvidia gpu and use intel, which is quite fine for my use cases 2021-03-09 15:47:45 less power is needed and temperature is a lot lower, notebooks are not noisy too much 2021-03-09 16:00:27 mps: it works :D 2021-03-09 16:01:07 seems to hang after though 2021-03-09 16:01:53 i think itsa config issue on my side though 2021-03-09 16:05:29 your kernel boots then it tries to load the pmos initramfs 2021-03-09 16:12:42 mps: mount: mounting configfs on /config failed: No such device 2021-03-09 16:18:28 black_ant: set it to load initramfs-edge, not pmos 2021-03-09 16:18:56 black_ant: sorry 2021-03-09 16:19:02 bl4ckb0ne: ^ 2021-03-09 16:19:18 yeah https://paste.sr.ht/~bl4ckb0ne/373879fd882bd70453e83d6daddf4a4a5f51ce7a 2021-03-09 16:19:26 seems to be a conflict with the pmos mkinitfs 2021-03-09 16:20:48 pmos is conflict by itself :) 2021-03-09 16:20:56 yeah 2021-03-09 16:21:17 dunno why but linux-edge depends on pmos mkinitfs 2021-03-09 16:21:19 World updated, but the following packages are not removed due to: 2021-03-09 16:21:21 postmarketos-mkinitfs: postmarketos-base device-pine64-rockpro64 linux-edge .makedepends-linux-edg 2021-03-09 16:22:38 could you post kernel boot log 2021-03-09 16:23:49 sure 2021-03-09 16:24:12 https://paste.sr.ht/~bl4ckb0ne/ef9081f010a6219a89aab9b516dd3638eb0a4e3b 2021-03-09 16:24:50 this dependency problem is probably because postmarketos-mkinitfs defines self as 'mkinitfs' package 2021-03-09 16:24:55 yep 2021-03-09 16:25:08 im working on cleaning the pmos crap 2021-03-09 16:26:51 bl4ckb0ne: new linux-edge 5.11.5 is probably already on mirrors 2021-03-09 16:27:01 nice 2021-03-09 16:27:10 with serial driver fixed 2021-03-09 16:27:27 i think i locked myself out by purging postmarketos-base fuck 2021-03-09 16:27:38 it removed sudo and i cant find my root pwd 2021-03-09 16:27:46 huh 2021-03-09 16:27:57 ughhhhhhhhhh 2021-03-09 16:28:19 ouch :) 2021-03-09 16:28:41 ACTION runs pmboostrap install 2021-03-09 16:32:57 bl4ckb0ne: why you don't install alpine 2021-03-09 16:33:04 sup ppl. i'm trying to use a gcc6 on alpine in a docker environment but i couldn't make it. i've seen that I can't do it using pkg, so I tried to download from ftp and build it, but `configure` give's me syntax errors when checking with gcc9 (which is the current installed by apk). I'm missing something? I can't compile gcc6 with gcc9 (I'm supposing), so how to retrieve gcc6 to run within an alpine environment? 2021-03-09 16:33:37 pmos makes it easier to make a proper sdcard 2021-03-09 16:33:44 ercdude: apk add gcc6 2021-03-09 16:34:07 wat 2021-03-09 16:34:11 I though I've tried that 2021-03-09 16:34:51 and it had gave me an error about dependencies... but it worked now anw, huge thanks mps 2021-03-09 16:35:05 I will try a dry run 2021-03-09 16:35:14 bl4ckb0ne: easier than that https://arvanta.net/alpine/install-alpine-on-acer-r13-chromebook/ 2021-03-09 16:38:03 this helios64 manufacturer should send one board to me to make alpine easy install on their board 2021-03-09 16:38:26 that would be awesome 2021-03-09 16:38:36 not being stuck with armbian 2021-03-09 16:40:12 i still need to put alpine on my espressobin 2021-03-09 16:40:13 lol 2021-03-09 16:40:19 i need a backup router 2021-03-09 17:11:48 mps: is your script meant to be run from aarch64 2021-03-09 17:24:54 bl4ckb0ne: it can't be run from any arch (parameter --arch aarch64) or even from other distro if you use apk.static 2021-03-09 17:26:03 ofc, pkgs and some other things must be changed to desired target 2021-03-09 17:42:19 mps: so, gcc6 works but what I meant was g++6 (sorry). When I try to apk add it, it indicates that it breaks with unsatisfiable constraints. Not sure why, but it's related to a dependency with gcc9 I guess 2021-03-09 17:42:52 ERROR: unsatisfiable constraints: 2021-03-09 17:42:52 libstdc++-9.3.0-r2: 2021-03-09 17:42:52 breaks: g++6-6.4.0-r11[libstdc++=6.4.0-r11] 2021-03-09 17:42:52 satisfies: gcc-9.3.0-r2[so:libstdc++.so.6] binutils-2.34-r1[so:libstdc++.so.6] 2021-03-09 17:42:55 gcc-9.3.0-r2: 2021-03-09 17:42:58 breaks: g++6-6.4.0-r11[gcc=6.4.0-r11] 2021-03-09 17:43:08 sry for flooding 2021-03-09 17:44:51 ercdude: only gcc6 exists in repo, not g++6 2021-03-09 17:45:17 oh, sorry 2021-03-09 17:45:35 it exists. I never used it so forgot for it 2021-03-09 17:46:22 ercdude: you could try to uninstall all -dev pkgs and then try to install g++6 2021-03-09 17:53:52 mps: just tried with a clean image, still breaks. :/ 2021-03-09 17:59:54 I don't have any installation right now where I could check problem, sorry 2021-03-09 18:05:07 mps: it's ok, thanks for the help! 2021-03-09 18:09:22 ercdude: maybe you can try in VM, lxc or qemu with clean and minimal install 2021-03-09 18:29:50 mps: I think I found the problem. g++6 depends on gcc6 and libstd++6 (among with others). gcc6 depends on binutils, which depends on libstd++ and gcc... I think that, since binutils is solved first, it gets the latest version of gcc and libstd++ (which is 10) and then conflicts with the dependency of g++6. At least this is what `apk dot` shows me. Do you think that it's possible that its recipe is broken? 2021-03-09 18:31:12 I will try in a vm 2021-03-09 18:31:53 ercdude: yes, looks like you are right 2021-03-09 19:25:00 mps: i see. I suppose it is not possible to make only nvidia diver use glibc, while the rest of the system musl right? ^^ 2021-03-09 19:33:33 Userland applications, yes. 2021-03-09 19:33:37 Kernel modules, no. 2021-03-09 19:33:58 Write to nvidia and ask for musl support. 2021-03-09 19:35:24 Uniformbuffer[m]: right 2021-03-09 19:40:39 i'm sure nvidia would be more than happy to do nothing 2021-03-09 19:40:51 mps: where does the vmlinux.kpart comes from? 2021-03-09 19:40:52 :) 2021-03-09 19:41:23 ok, thanks all. Anyway, fys there is no chance nvidia will support musl themself unfortunately. optimus was bugged as hell for YEARS on linux and only recentely they fixed the vsync problem. They simply don't care about linux.... 2021-03-09 19:41:32 Oh I know. 2021-03-09 19:41:40 really not all that surpising 2021-03-09 19:41:41 But that's literally your only option. 2021-03-09 19:42:14 I've learned to pick my hardware selectively for the task and OS I will be using. 2021-03-09 19:42:21 That is in no way a criticism of you. 2021-03-09 19:42:21 indeed 2021-03-09 19:42:29 it's just how I've learned to cope. 2021-03-09 19:42:33 lol 2021-03-09 19:42:37 learned my lesson the first nvidia card I bought 2021-03-09 19:43:10 My linux desktop is an intel nuc and my gaming pc is now a plex server running ubuntu 2021-03-09 19:43:16 do i like ubuntu? no. no i do not. 2021-03-09 19:43:28 but plex and nvidia support it best 2021-03-09 19:43:45 but it's offloaded away from my desktop 2021-03-09 19:43:49 so i dont care at the end of the day 2021-03-09 19:43:52 lol 2021-03-09 19:44:04 i only see it through a kodi/alpine box i built 2021-03-09 19:44:27 c705: that was the first, and the last hardware with an nvidia gpu. Next time i will go with amd (or intel if the make a good gaming gpu 🤞) 2021-03-09 19:44:51 i've had relatively few problems with amdgpu 2021-03-09 19:45:35 I only care about gpu performance on linux for media server transcoding and nvenc is the best game in town 2021-03-09 19:45:35 * c705: that was the first, and the last hardware with an nvidia gpu. Next time i will go with amd (or intel if they make a good gaming gpu 🤞) 2021-03-09 19:46:09 it does, but for me, dealing with the bullshit isn't worth it 2021-03-09 19:46:18 apparently some amd gpus work in plex but none are supported and they wont even given a list of what might work 2021-03-09 19:46:27 hello there 2021-03-09 19:46:38 got some problem installing alpine last version into a qemu test env. 2021-03-09 19:46:53 the qcow2 disk is able to grow up to 4GB 2021-03-09 19:46:58 bl4ckb0ne: I hope you didn't installed linux-elm, it is for specific chromebooks and not general kernel 2021-03-09 19:47:11 but alpine stop at 41 mb 2021-03-09 19:47:50 error on standard and virt 2021-03-09 20:16:13 still tweaking the script to make it run from arch 2021-03-09 20:18:10 bl4ckb0ne: this notes are just hints how to install alpine on different devices, not a guide 2021-03-09 20:18:37 yeah im reading everything to try to grok what to do 2021-03-09 20:18:49 though yes, with running it alpine will be installed and ready on elm chromebooks 2021-03-09 20:20:40 and replacing linux-elm with linux-gru it will boot on gru kevin chromebook 2021-03-09 20:21:44 bl4ckb0ne: you need u-boot for helios64 board 2021-03-09 20:24:47 i think they have a uboot only image 2021-03-09 20:25:21 'they'? 2021-03-09 20:25:26 'they' are? 2021-03-09 20:25:27 kobol 2021-03-09 20:25:34 https://wiki.kobol.io/helios64/files/u-boot/helios64_sdcard_u-boot-only.img.xz 2021-03-09 20:25:42 good then 2021-03-09 20:28:32 Does anyone else have iptables problems with docker on alpine? When I run `docker run -d -p 0.0.0.0:80:80 --rm --name nginx nginx` then `docker stop nginx && docker run -d nginx && docker run -p 0.0.0.0:80:80 --rm nginx nginx` I have stale iptables rules from when I first created the nginx container. When the container stops or gets removed the rule stays. Is anyone else experiencing this? It 2021-03-09 20:28:34 seems like the only fix is to restart docker 2021-03-09 20:32:55 where can I find the apk recipe for g++? 2021-03-09 20:33:34 ercdude: aports/community/gcc6 2021-03-09 20:33:58 it's the same of gcc? 2021-03-09 20:34:19 it is practically stalled gcc 2021-03-09 20:34:32 oh 2021-03-09 20:35:45 g++ is built from same APKBUILD 2021-03-09 20:36:09 oh, ok 2021-03-09 20:36:14 thanks :) 2021-03-09 20:36:20 <[diablo]> jemershaw 2021-03-09 20:36:23 <[diablo]> jemershaw yes ! 2021-03-09 20:36:32 <[diablo]> I wrote earlier about exactly this problem... 2021-03-09 20:37:03 <[diablo]> you must install ip6tables 2021-03-09 20:37:52 <[diablo]> https://paste.centos.org/view/aad86de2 2021-03-09 20:38:38 <[diablo]> jemershaw so install it, then restart your docker daemon ... 2021-03-09 20:46:09 [diablo]: thanks so much! This was a huge pain! 2021-03-09 20:46:18 <[diablo]> yup 2021-03-09 20:46:19 <[diablo]> I know 2021-03-09 20:46:20 <[diablo]> :D 2021-03-09 20:47:11 <[diablo]> jemershaw it's working correctly now I assume 2021-03-09 20:48:14 indeed! 2021-03-09 20:48:31 this should def be a dependency for docker package 2021-03-09 20:48:40 <[diablo]> yeah 2021-03-09 20:48:42 <[diablo]> voila 2021-03-09 20:48:48 <[diablo]> odd thing is this tho... 2021-03-09 20:48:55 <[diablo]> so I put the deamon in debug mode... 2021-03-09 20:49:14 <[diablo]> and I spotted the line you see in the pastebin... a warning about no ip6tables. ... but thought nothing of it.. 2021-03-09 20:49:41 <[diablo]> I could see when a container came up, it creating the rules... but when stopping the container, nothing about tearing down the rules... 2021-03-09 20:49:55 <[diablo]> then I thought, oh well, lets add the ip6tables .... 2021-03-09 20:50:18 <[diablo]> and, to my surprise , it worked... 2021-03-09 20:50:36 <[diablo]> I then saw the rules being torn down when stopping the container ... 2021-03-09 20:51:00 <[diablo]> what was odd, that there was no warning nor error in the debug logs saying that it couldn't / wouldn't do the removal of the rules.. 2021-03-09 20:55:31 <[diablo]> My very first real interaction with Docker was to create for a friend a new mail server (to replace his old one), and as we'd set him up TrueNAS I said let's do a Byhve VM to run Docker... normally I'd have chosen Ubuntu LTS, but as Alpine was so light, I took a chance on it... and it's been amazing. He's been running it as a Docker host now, hmmm 2021-03-09 20:55:31 <[diablo]> maybe 8 months, no issues - except when he noticed the IPtables problem, and that is why I looked into it. 2021-03-09 20:58:31 I'm proud of you [diablo] and I love you 2021-03-09 20:58:43 <[diablo]> PMSL 2021-03-09 21:00:49 <[diablo]> I really do like Alpine a lot. I work day in day out with RHEL, and there's something, well... real clean about Alpine. Just no bloat 2021-03-09 21:02:01 *cough*nosystemdisntbadeither*cough* 2021-03-09 21:02:33 <[diablo]> hahaha, so so true ! 2021-03-09 21:02:34 I've never used alpine as a distro, just in docker 2021-03-09 21:02:39 is the distro nice? 2021-03-09 21:02:47 i use it for lots of stuff 2021-03-09 21:02:50 so id say yes 2021-03-09 21:02:57 <[diablo]> fowl for Docker host - awesome from my limited experience with it 2021-03-09 21:02:58 its my kodi media box, my router, etc 2021-03-09 21:03:13 alpine is extremely diverse 2021-03-09 21:04:19 <[diablo]> my mate who should actually be in this channel but isn't , runs a quite a few containers ... Plex Server, Ubiquiti Controller, OpenSMTPD, Dovecot, Fetchmail, Thelounge, Townky, and a few others... 2021-03-09 21:04:50 <[diablo]> and it's all running on a 2GB of RAM Alpine VM on TrueNAS 2021-03-09 21:05:26 Plex on Alpine seems ... silly. 2021-03-09 21:05:37 <[diablo]> yeah I agree 2021-03-09 21:05:44 <[diablo]> but he likes it that way 2021-03-09 21:05:57 I love Alpine, but you know what I love more? nvenc. 2021-03-09 21:05:59 haha 2021-03-09 21:06:05 <[diablo]> dunno that 2021-03-09 21:06:15 its nvidia's hardware transcoding stuff 2021-03-09 21:06:16 <[diablo]> oh 2021-03-09 21:06:24 <[diablo]> haha, ok gotcha after a quick Google 2021-03-09 21:06:34 its the ONLY reason i use ubuntu for my plex box 2021-03-09 21:06:39 <[diablo]> well I run Plex on a FreeBSD jail... 2021-03-09 21:06:53 <[diablo]> and I use direct play on my ATV 4K 2021-03-09 21:06:59 <[diablo]> so no transcoding 2021-03-09 21:07:13 i host movies for my friends who are on the road and watch with their phones 2021-03-09 21:07:17 so transcoding is important 2021-03-09 21:08:01 <[diablo]> fys shouldn't your nick be yify ? 2021-03-09 21:08:03 <[diablo]> ;-) 2021-03-09 21:08:18 oy 2021-03-09 21:08:22 <[diablo]> hehe 2021-03-09 21:08:22 that reminds me 2021-03-09 21:08:31 i need to find out when Resident Alien airs 2021-03-09 21:09:04 <[diablo]> ACTION will be relaxing with Mrs [diablo] to finish the Harry and Meghan with Oprah crap 2021-03-09 21:09:08 :) 2021-03-09 21:10:02 <[diablo]> I must actually file a bug report somewhere about this ip6tables thing. I dunno perhaps they're already aware of it, I really don't know 2021-03-09 21:10:18 Better safe than let it go unnoticed. 2021-03-09 21:12:14 <[diablo]> https://gitlab.alpinelinux.org/groups/alpine/-/issues 2021-03-09 21:12:18 <[diablo]> here? 2021-03-09 21:12:30 you got me, i dont know tbqh 2021-03-09 21:12:34 never needed to file a report 2021-03-09 21:13:15 <[diablo]> me neither 2021-03-09 21:38:39 > [ 8.141793] Mounting boot media: failed. 2021-03-09 21:38:43 got further away mps 2021-03-09 21:41:12 bl4ckb0ne: missing mmc driver, add modules in u-boot append line with appropriate mmc driver 2021-03-09 21:42:33 example for another board: APPEND modules=sd-mod,usb-storage,ext4,f2fs,sunxi-mmc root=/dev/mmcblk0p2 rw rootwait console=....... 2021-03-09 21:42:59 example for another board: APPEND modules=sd-mod,usb-storage,ext4,f2fs,sunxi-mmc root=/dev/mmcblk0p2 rw rootwait console=ttyS0,115200 2021-03-09 21:56:11 fails to mount mmcblk0p2 2021-03-09 21:59:19 is the mmcblk detected during boot 2021-03-09 22:00:01 theres no mention of it but the kernel boots 2021-03-09 22:01:05 which driver you added for mmc to u-boot 2021-03-09 22:01:29 sunxi-mmc i think 2021-03-09 22:01:52 sunxi is not for rk3399 2021-03-09 22:02:02 it was just example 2021-03-09 22:02:14 oh then i think thats the one i got wrong 2021-03-09 22:02:41 could you look which driver is used on armbian for mmc 2021-03-09 22:11:10 could it be dwmmc? 2021-03-09 22:11:38 could be 2021-03-09 22:16:43 armbian puts root to `/dev/mmcblk0p1` 2021-03-09 22:17:25 dont see any modules 2021-03-09 22:17:38 where you put root FS this must be parameter for 'root' option 2021-03-09 22:18:50 yeah it's on the second partition of my sd 2021-03-09 22:19:33 it boot from the uuid directly, no modules in the cmdline 2021-03-09 22:20:12 they build driver in-kernel not as modules 2021-03-09 22:21:37 if you can find what driver is used for mmc with dmesg maybe we can find which module should be used 2021-03-09 22:24:10 manufacturer didn't wrote (or I couldn't find) what options/drivers are needed in kernel for this board 2021-03-09 22:25:21 rk3399 use MMC_SDHCI_OF_ARASAN for e/mmc I think 2021-03-09 22:25:32 https://tpaste.us/lEla 2021-03-09 22:27:18 same as on my rk3399 chromebook 2021-03-09 22:28:55 I can tomorrow build new kernel with mmc driver in kernel for you to test, if you want to test ofc 2021-03-09 22:29:35 yes please 2021-03-09 22:30:38 looks like we are in distant TZs 2021-03-09 22:31:14 im on UTC-5 2021-03-09 22:31:44 I'm in CET 2021-03-09 22:32:46 https://gitlab.com/postmarketOS/pmaports/-/blob/master/main/linux-postmarketos-rockchip/config-postmarketos-rockchip.aarch64#L6483 i think you have the good config 2021-03-09 22:34:31 yes, but they also have it in-kernel, and we have it as module 2021-03-09 22:35:29 so if I find the right name i dont need a new kernel upgrade> 2021-03-09 22:35:31 ?* 2021-03-09 22:38:06 I think so 2021-03-09 22:39:18 try to add to module list this 'sdhci-of-arasan' 2021-03-09 22:40:41 tar tvf packages/testing/aarch64/linux-edge-5.11.4-r0.apk | grep 'mmc.*host' 2021-03-09 22:40:56 will show all mmc modules 2021-03-09 22:41:23 problem could be if it is not added to initramfs 2021-03-09 22:42:10 but that could be solved by manually creating initramfs 2021-03-09 22:43:24 bl4ckb0ne: but I have to go to bed, we can continue tomorrow when I build kernel with drivers in-kernel. good night 2021-03-09 22:44:02 whoever did this, don't be a dick pls: https://teddit.net/r/AlpineLinux/comments/lxbmwx/what_exactly_are_subpackages/ 2021-03-09 22:45:05 what is.. teddit? 2021-03-09 22:51:16 Teddit.net is a no-JS front-end to Reddit 2021-03-09 22:51:31 or rather, a lightweight front-end that can work without JS 2021-03-09 23:34:24 bovis: my mistake, winebuild isn't a perl script (anymore?) 2021-03-09 23:34:50 perhaps your confusion is that make doesn't print the name of the command that segfaulted, it prints the name of the target 2021-03-09 23:52:39 mps: ty, good night! 2021-03-10 00:16:51 Hello71: Right, I don't know what's causing the action. As far as I know, it's just Make-ing. How do I get to the real issue? 2021-03-10 00:17:24 https://bpa.st/C6QQ is all that looks useful before my error. I tried building with and without "textrels" as a build option, and the result is the same. 2021-03-10 00:18:56 Can make be forced to output more information, like the name of the command that's segfaulting? 2021-03-10 00:20:24 it already does 2021-03-10 00:21:16 tools/winebuild/winebuild is the failing command 2021-03-10 00:21:44 dlls/kernel32/libkernel32.def is the name of the file to be created. make is telling you that on line 67511 of Makefile there is a job to create dlls/kernel32/libkernel32.def and the command failed 2021-03-10 00:32:57 Hello71: In that case, running the command here gives the output here from gdb (https://bpa.st/E7VQ) 2021-03-10 00:34:12 how are you trying to cross compile 2021-03-10 00:34:18 and why do you have wine-6.0/src/wine-6.3 2021-03-10 00:37:15 I originally compiled 6.0, got the error, compiled 6.3, found the same error, and didn't change the directory name 2021-03-10 00:38:12 Hello71: Compilation is taking place in a 32-bit chroot. I run 64-bit alpine normally. The chroot is inside that running system (what I'm using for this chat). 2021-03-10 01:16:47 i installed intel-ucode, but i can't find it enabled in dmesg 2021-03-10 01:16:53 already using grub 2021-03-10 02:14:05 anyone know? ucode please 2021-03-10 02:24:22 did you read the arch instructions 2021-03-10 02:24:41 also there's no point in asking again if you're just going to say "anyone know" 2021-03-10 02:24:59 the only reason to ask again is if people joined after you asked, but "anyone know" doesn't give them any information 2021-03-10 02:25:19 hey, stop using logic 2021-03-10 02:26:26 otherwise all that asking again does is moderately annoy people and make them less likely to answer 2021-03-10 02:26:36 i did read arch. same precess works in arch but not in alpine 2021-03-10 02:31:18 mps: booted with `modules=dw_mmc-rockchip,sdhci-of-arasan,sdhci-arasan console=ttyS2,1500000n8 rw root=UUID=5ee82f7d-00bb-42ec-9ce5-0438fd4ca0c9 rootwait rootfstype=ext4 loglevel=7` and I still ended up in the initramfs recovery shell 2021-03-10 02:33:42 I have this init script, how can I add a second argument to it 2021-03-10 02:33:42 #!/sbin/openrc-run 2021-03-10 02:33:43 depends() { 2021-03-10 02:33:43     need localmount net 2021-03-10 02:33:44     after firewall 2021-03-10 02:33:44 } 2021-03-10 02:33:45 description="Mumble VoIP Server" 2021-03-10 02:33:45 command="/usr/bin/murmurd" 2021-03-10 02:33:46 command_args="-ini ${INI_CONFIG:-/etc/murmur.ini}" 2021-03-10 02:33:46 pidfile="/var/run/murmur/murmur.pid" 2021-03-10 02:33:47 start_pre() { 2021-03-10 02:33:47     checkpath --owner murmur:murmur --directory ${pidfile%/*} 2021-03-10 02:33:48 } 2021-03-10 02:33:49 # uses the openrc templates for start()/stop() 2021-03-10 02:34:12 If I add it to the end of command_args, it doesn;t work 2021-03-10 02:34:36 RedditLover35211: what second argument do you want to add and how are you adding it? 2021-03-10 02:34:37 first step is to avoid pasting that many lines into IRC, instead using termbin, pastebin, or other such services 2021-03-10 02:34:44 tehcloud: lol 2021-03-10 02:34:51 RedditLover35211: tehcloud is definitely not wrong 2021-03-10 02:34:59 Sorry 2021-03-10 02:35:47 I tried to add it like this 2021-03-10 02:35:48 https://pastebin.com/qsgLJ3kS 2021-03-10 02:36:28 https://pastebin.com/raw/qsgLJ3kS ← raw link for those that won't open pastebin.com links (usually includes me) 2021-03-10 02:42:47 Any ideas hg 2021-03-10 02:43:04 RedditLover35211: what happens when you use that? 2021-03-10 02:44:02 Sorry password works, it doesn't work when I use Josh2006 2021-03-10 02:44:19 … 2021-03-10 02:44:21 what? 2021-03-10 02:44:46 bruh 2021-03-10 02:45:11 Here ill show you 2021-03-10 02:48:04 With password Josh2006 https://pastebin.com/raw/a5p3NDs1 2021-03-10 02:49:33 actually they both don't work 2021-03-10 02:50:38 so, probably it's not to do with the password 2021-03-10 02:50:45 especially since it gets past setting the password 2021-03-10 02:50:53 so, there's probably something else wrong unrelated 2021-03-10 02:51:01 does it work at all without setting the password at all? 2021-03-10 02:51:11 (i.e., without the extra arguments) 2021-03-10 02:51:34 yes it will work, without setting a password 2021-03-10 02:51:42 meaning if I don't have -supw 2021-03-10 02:52:10 can you show more logs? 2021-03-10 02:53:35 I can murmurd logs, but thats nothing special in it, they only output is from service 2021-03-10 02:54:56 they don't indicate a crash or anything? 2021-03-10 02:55:08 it seems like they would 2021-03-10 02:55:29 https://termbin.com/uqal 2021-03-10 02:55:43 The logs in there are from when it was without -supw 2021-03-10 02:55:50 So I think its a problem in the init script 2021-03-10 02:58:17 can you show me the log with the -supw argument? 2021-03-10 02:58:25 so that I can see what the failure looks like in-log? 2021-03-10 02:59:28 That was the log, it ones big log file 2021-03-10 02:59:51 can you point me to one of the ones with -supw passed? 2021-03-10 02:59:58 just a timestamp would be fine 2021-03-10 03:01:34 When I start murmurd with service murmur start, it doesn't record anything in the log 2021-03-10 03:02:03 The last timestamp is 13:37:50.685 2021-03-10 03:02:06 RedditLover35211: even if -supw is not passed? 2021-03-10 03:02:18 I will run service murmurd start, and see if the timestamp changes 2021-03-10 03:02:29 If -supw is not passed, then it start fine and records to the log 2021-03-10 03:02:55 The timestamp in the log file didn't change the last entry was 13:37:50.685 2021-03-10 03:03:09 so I believe its initd complaining about the formatting or something 2021-03-10 03:03:43 perhaps it's intercepting the logging events and shoving them into syslog 2021-03-10 03:03:46 do you see them in the system log? 2021-03-10 03:03:56 Maybe I could create a script that records that args it gets, and set the application to run as the script 2021-03-10 03:05:59 https://pastebin.com/raw/tZPqrjxh 2021-03-10 03:06:03 here is syslog 2021-03-10 03:06:10 a snippet 2021-03-10 03:06:54 hmm 2021-03-10 03:06:58 I'm unsure 2021-03-10 03:07:20 it's clear that something is amiss, but I'm not sure what it would be 2021-03-10 03:07:40 RedditLover35211: you could install the debug symbols and try to get a traceback 2021-03-10 03:07:43 that'd be helpful 2021-03-10 03:08:22 hmm, no debug package for murmur 2021-03-10 03:08:34 Hmm okay 2021-03-10 03:09:23 Let me create a python script, that outputs the args it gets to a file 2021-03-10 03:15:12 Alright this is what it wrote to the file hg 2021-03-10 03:15:14 ['/root/test.py', '-ini', '/etc/murmur.ini', '-supw', 'password'] 2021-03-10 03:15:29 well, that's good 2021-03-10 03:16:07 So it's not initd 2021-03-10 03:16:15 well 2021-03-10 03:16:17 it must be murmurd? 2021-03-10 03:16:29 it's, at least, not something wrong with passing the arguments 2021-03-10 03:17:12 yes 2021-03-10 03:22:08 I got things bootstrapping for ppc again although now I'm running into issues compiling on a PPC host, I gave up on the VM because it was just refusing to start but now on a real host I'm getting the same "libfakeroot, error when sending message: Invalid argument" as I was getting before on qemu usermode emulation 2021-03-10 03:23:54 Its really weird without it works. 2021-03-10 03:28:25 nice to dev. finally patches for qt5 works almost fine, stop render crashes on qutebrowser. 2021-03-10 05:05:43 I've tried messing with patches and haven't been able to gain any useful information on fakeroot having issues, I think I'll just compile it statically against glibc and come back to that later 2021-03-10 05:40:10 Well apparently it might've been the kernel lol, that's annoying 2021-03-10 05:40:19 Just recompiling that now to test 2021-03-10 06:09:35 Nevermind, that still didn't fix it 2021-03-10 07:51:23 bl4ckb0ne: https://dev.alpinelinux.org/~mps/linux-edge-5.11.5-r0.apk 2021-03-10 07:52:58 bl4ckb0ne: this have mmc drivers compiled in kernel so don't need to put them in module list on the u-boot APPEND 2021-03-10 07:53:28 also drivers for ext4/3/2 and f2fs are in-kernel 2021-03-10 08:39:43 anyone know for good DPI (dots per inch) calculator, or formulae 2021-03-10 08:49:05 Hi everyone :) how to know when duplicity and s6-overlay packages will be upgraded into 3.13 ? 2021-03-10 11:42:24 mps: not sure what the issue is, just take dots divided by inches? 2021-03-10 11:51:52 Hello71: already done with bc 2021-03-10 11:55:49 Well I think I found the issue with fakeroot 2021-03-10 11:55:57 aha 2021-03-10 11:55:58 https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=971928 2021-03-10 11:56:38 I'll see if I can implement a fix tomorrow 2021-03-10 15:31:49 mps: giving it a try atm 2021-03-10 15:36:55 mps: still not mounting 2021-03-10 15:39:27 https://paste.sr.ht/~bl4ckb0ne/139f069f2cdbdaa3d13e0c27df610eca84e12d12 everything seems good on the mmc driver side 2021-03-10 15:52:33 bl4ckb0ne: which kernel you use? from alpine mirror or from my dev.a.o url? 2021-03-10 15:54:24 the one from your url 2021-03-10 15:56:21 can you see /dev/mmcblk devices from emergency shell 2021-03-10 15:56:53 i dont see them 2021-03-10 15:56:57 i dont even see my hard drives 2021-03-10 15:58:15 sata driver is not loaded, so this part is ok 2021-03-10 16:02:09 could you try with 'root=/dev/mmcblk1p2' if root fs is partition 2 2021-03-10 16:02:24 i have the uuid in root 2021-03-10 16:02:51 or mmcblk0p2 (I forgot exact order of rk3399 mmc cards) 2021-03-10 16:02:55 ill try 2021-03-10 16:05:01 hi all, I'm try to figure out how to replay to an ~alpine/users thread... ther's a guide that help to setup sourcehut client and explain how to send replaies? 2021-03-10 16:06:32 no luck either 2021-03-10 16:06:51 for ubuntu there is the cubic package but also pinguybuilder to take a snapshot of current install and make that into iso - is there something similar for x86 alpine linux? 2021-03-10 16:07:49 that is I install alpine linux x86 in a vm, install and configure it how I like and then make that into an iso so I can then use that iso to either burn on a cdr or put it in a usb drive and boot from using isolinux/syslinux 2021-03-10 16:07:50 slusenti: https://git-send-email.io 2021-03-10 16:08:54 bl4ckb0ne: sorry for stupid question, but are you 100% sure you use kernel from dev.a.o/~mps/ 2021-03-10 16:09:00 5.11.5 2021-03-10 16:09:23 yes 2021-03-10 16:09:33 > Linux (none) 5.11.5-0-edge #1-Alpine SMP PREEMPT Tue, 09 Mar 2021 23:01:16 +0000 aarch64 Linux 2021-03-10 16:09:33 bl4ckb0ne: thanks 2021-03-10 16:09:34 sorry yes 2021-03-10 16:09:45 > [ 0.000000] Linux version 5.11.5-0-edge (mps@mps-edge-aarch64) (gcc (Alpine 10.2.1_pre1) 10.2.1 20201203, GNU ld (GNU Binutils) 2.35.1) #1-Alpine SMP PREEMPT Tue, 09 Mar 2021 23:01:16 +0000 2021-03-10 16:09:53 `inux version 5.11.5-0-edge (mps@mps-edge-aarch64 2021-03-10 16:10:10 mps ^ 2021-03-10 16:11:56 bl4ckb0ne: I will try later to boot my rk3399 and look at dmesg to see what more/else is needed 2021-03-10 16:12:33 ty 2021-03-10 16:12:37 ill continue to poke around 2021-03-10 16:12:47 now I have to finish some queues in my $day_job 2021-03-10 16:13:30 but we made some progress 2021-03-10 16:13:49 yep thanks a lot! 2021-03-10 16:13:56 np 2021-03-10 16:14:31 i dont find much data about no devices showing in the initramfs recovery shell 2021-03-10 16:15:08 even fdisk -l lists nothing 2021-03-10 16:16:34 ls /dev 2021-03-10 16:17:09 https://paste.sr.ht/~bl4ckb0ne/1016efb038beffeebbda45c6738ad3b10fc5efc4 2021-03-10 16:18:24 uh, no mmcblk, so something still is missing in kernel 2021-03-10 16:19:56 got soem stuff in /proc/devices tho 2021-03-10 16:25:05 I will try to 'copy' drivers settings from linux-gru 2021-03-10 16:29:53 hmm i dont see mmc_core 2021-03-10 16:33:19 i have mmc_test and mmc_hsq but no mmc_block nor mmc_core 2021-03-10 16:34:22 in a 10 minutes I'll upload new kernel to dev.a.o 2021-03-10 16:35:42 neat 2021-03-10 16:40:48 bl4ckb0ne: I think it is there, same file name and same version but upload date is now 2021-03-10 16:42:06 md5sum `c7e48a35d00e13fd91a96aab0f030ce2 linux-edge-5.11.5-r0.apk` 2021-03-10 16:43:54 c7e48a35d00e13fd91a96aab0f030ce2 2021-03-10 16:43:58 yes 2021-03-10 16:44:28 same issue 2021-03-10 16:44:46 huh 2021-03-10 16:45:04 try to skip initramfs 2021-03-10 16:45:29 i.e. remove initram from u-boot config 2021-03-10 16:46:11 what FS type is your root partition 2021-03-10 16:46:17 ext4 iirc 2021-03-10 16:46:27 ok, ext4 is in-kernel 2021-03-10 16:48:54 hangs on `[ 0.266390] Waiting for root device /dev/mmcblk1p2...` without initrd in extlinux 2021-03-10 16:50:12 could you post me pmos kernel config link again 2021-03-10 16:50:41 https://gitlab.com/postmarketOS/pmaports/-/blob/master/main/linux-postmarketos-rockchip/config-postmarketos-rockchip.aarch64 2021-03-10 16:50:52 and btw, what device shows as root fs when you booted pmos 2021-03-10 16:51:13 /dev/mmcblk1p2 2021-03-10 16:51:16 boot on /dev/mmcblk1p1 2021-03-10 16:53:20 ok 2021-03-10 16:54:06 I'll try later to add all these mmc options and drivers in kernel and build new one, but it will take few hours 2021-03-10 16:54:22 ill continue to poke around with this kernel 2021-03-10 16:55:02 but my rk3399 boot fine with less drivers and options enabled than this last kernel I built 2021-03-10 16:55:30 bbl 2021-03-10 16:57:08 eh my kid is up, afk 2021-03-10 17:37:43 alpine has strange behavior hosting docker containers 2021-03-10 17:37:50 network issue 2021-03-10 17:38:09 metbsd: We use docker ourselves on alpine, we experienced no issues 2021-03-10 17:38:22 have to reboot the pc to make network to work 2021-03-10 17:38:35 DNS? 2021-03-10 17:39:00 i can ping, but the port does not open 2021-03-10 17:39:08 That's the only thing I noticed that sometimes, at boot dns would not work under certain circumstanced. Restarting docker would fix that 2021-03-10 17:39:11 i can telnet 127.0.0.1 2021-03-10 17:39:28 port does not open? 2021-03-10 17:39:56 77199b8f4bc1 crazemaze:ease "node app.js -f 103.…" 28 hours ago Up 10 minutes 8081/tcp, 0.0.0.0:80->8080/tcp 2021-03-10 17:40:24 i can telnet localhost 80, but remote pc can't telnet this ip:80 2021-03-10 17:41:03 restarting container don't help. but reboot does 2021-03-10 17:42:10 after reboot and restart all containers, everything becomes normal 2021-03-10 17:53:00 Did you try to restart docker itself? 2021-03-10 17:53:11 Not just the containers 2021-03-10 17:58:04 didn't try. my bad 2021-03-10 17:58:11 did you? 2021-03-10 17:58:24 oh you did 2021-03-10 18:52:40 <[diablo]> metbsd I think you'll find if you add ip6tables package, and restart docker daemon, it'll work 2021-03-10 18:52:59 <[diablo]> ikke I wrote here yesterday about how I found this issue. I (badly) attempted to file a bug report 2021-03-10 18:53:38 <[diablo]> https://paste.centos.org/view/aad86de2 2021-03-10 18:53:49 <[diablo]> metbsd please try that , see if it solves your issue 2021-03-10 18:54:00 <[diablo]> so far, 3 x people have found it fixed it 2021-03-10 18:59:28 [diablo]: oh right, I didn't connect the 2 (no pun intended) 2021-03-10 19:00:09 <[diablo]> ikke basically IMO there needs to be a dep on the ip6tables package... 2021-03-10 19:00:17 <[diablo]> soon as that's added , everything works a dream 2021-03-10 19:00:18 right 2021-03-10 19:00:45 <[diablo]> odd thing was - when you stop a container, it gave no warning or error that it was not or unable to remove the IPtables rules... 2021-03-10 19:03:23 <[diablo]> ah ha, phalanx welcome - ikke how I found this problem was due to phalanx having the issue, and I found I had the same issue on mine. Then yesterday another user was here, same problem, he tried the fix I proposed, and it worked 2021-03-10 19:05:56 [diablo]: aha, I see, it's docker-engine that would need the ip6tables dependency 2021-03-10 19:06:04 bl4ckb0ne: https://dev.alpinelinux.org/~mps/linux-edge-5.11.5-r1.apk 2021-03-10 19:06:05 as it's also the package that depends on iptables 2021-03-10 19:06:27 (it's a subpackage of docker) 2021-03-10 19:06:31 bl4ckb0ne: if that doesn't work I'm out of ideas 2021-03-10 19:07:20 bl4ckb0ne: and I checked my rk3399, external mmc is mmcblk1, emmc is mmcblk2 2021-03-10 19:09:55 [diablo]: https://gitlab.alpinelinux.org/alpine/aports/-/merge_requests/19178 2021-03-10 19:22:58 <[diablo]> ikke yeah exactly, the engine needs it added 2021-03-10 19:23:53 <[diablo]> nice one for the merge 2021-03-10 19:35:36 metbsd: The MR was merged 2021-03-10 19:39:18 hi guys, where should add "iwconfig wlan0 power on" in /etc/rc.conf leave a constant loop 2021-03-10 19:39:58 i mean power off from wlan0 from rpi 2021-03-10 19:40:47 /etc/modprobe.d? 2021-03-10 19:48:29 I'm looking for a VNC client I tried Tiger VNC and such but it seems that `apk` doesn't have this one? 2021-03-10 20:22:20 DylanVanAssche_: gvncviewer is available 2021-03-10 20:24:42 yZ5vlALg86lP: Ah thanks! Weird that I couldn't find it before... 2021-03-10 20:49:37 yZ5vlALg86lP: you should check if u did not change password for nickname by mistake :/ 2021-03-10 20:51:35 or maybe it's a hidden base64 message? 2021-03-10 20:58:03 mps: installing r1 atm 2021-03-10 20:58:14 same, mmc is mmcblk1, emmc is mmcblk2 2021-03-10 20:58:45 brnfmac as no param to disable power_save right? 2021-03-10 21:02:04 > [ 0.000000] Linux version 5.11.5-1-edge (mps@mps-edge-aarch64) (gcc (Alpine 10.2.1_pre1) 10.2.1 20201203, GNU ld (GNU Binutils) 2.35.1) #2-Alpine SMP PREEMPT Wed, 10 Mar 2021 17:39:35 +0000 2021-03-10 21:02:08 still stuck in recovery 2021-03-10 21:03:40 Hey everyone 2021-03-10 21:03:53 Got a slight problem while trying to start salt-minion: * supervise-daemon: /usr/bin/salt-minion does not exist 2021-03-10 21:03:57 Is this packaged incorrectly? 2021-03-10 21:04:09 hmm 2021-03-10 21:04:25 /usr/bin/salt-minion 2021-03-10 21:04:29 that should exist 2021-03-10 21:04:37 No such file :( 2021-03-10 21:04:56 and you installed salt-minion? 2021-03-10 21:05:00 not just salt? 2021-03-10 21:05:05 or salt-minion-openrc 2021-03-10 21:05:08 Yep 2021-03-10 21:05:21 apk info -L salt-minion 2021-03-10 21:05:24 pulled in dependencies, managed to get it enabled with rc-update then hit this roadblock 2021-03-10 21:05:38 that doesn't return any output 2021-03-10 21:05:48 mps: do you have the kernel config you used somewhere? 2021-03-10 21:05:51 https://paste.sr.ht/~martijnbraam/dfdeb17f59310a71fa48995c59dae5f24db14ebb 2021-03-10 21:05:55 oh with -openrc: https://0x0.st/-Zas.txt 2021-03-10 21:05:56 sy: apk version salt-minion 2021-03-10 21:06:18 Did salt-minion-openrc instead: salt-minion-openrc-3000.3-r0 = 3000.3-r0 2021-03-10 21:06:37 You should not install salt-minion-openrc direct 2021-03-10 21:06:40 directly 2021-03-10 21:06:42 ive been told that this config works on the pine book pro, which is also a rk3399 2021-03-10 21:06:51 install just `salt-minion` 2021-03-10 21:07:19 oh wow didn't see that in search 2021-03-10 21:07:45 got it, thanks ikke! back to configuring... 2021-03-10 21:07:54 hmm 2021-03-10 21:07:59 salt is kind of broken on 3.12 2021-03-10 21:08:19 is there a static version of abuild to build apks on other distros? 2021-03-10 21:08:22 ikke: what's wrong on 3.12? 2021-03-10 21:08:46 ah shit I just upgraded to 3.12 2021-03-10 21:09:00 https://gitlab.alpinelinux.org/alpine/aports/-/issues/11614 2021-03-10 21:09:33 ah the patch in edge was never backported to 3.12 2021-03-10 21:09:35 I decided to upgrade it to 3001, but when I tested it, it almost crashed my pc due to the amount of memory it uses, so I was kind of discouraged 2021-03-10 21:09:50 wsinatra: There was no simple patch for 3001 2021-03-10 21:10:05 I did try to manually figure out what fixed it, but never managed to find it 2021-03-10 21:10:19 right, I think it was an internal bug which was fixed between versions 2021-03-10 21:10:42 yeah, I did try to look through the commits, but their commit hygene is poor 2021-03-10 21:11:05 agreed entirely there, it's kind of frustrating 2021-03-10 21:12:16 sy: if I build a package that upgrades it to 3001 in 3.12, can you test it? (download the apks from our CI and install it) 2021-03-10 21:12:28 sure! 2021-03-10 21:13:54 Since the package is broken, could we push 3001? it's backwards compatible with 3000 2021-03-10 21:14:13 wsinatra_: yes, that was my plan 2021-03-10 21:15:19 okay, that's not too big an issue then, it would be better to push the upgrade than let the broken package fester. At least that's my personal take 2021-03-10 21:16:06 wsinatra_: agreed, that ticket has been bothering me for quite some time 2021-03-10 21:17:15 I don't blame you, it's a pretty critical piece of software for the companies/users that use it 2021-03-10 21:17:33 big old headache when it doesn't just do what its supposed to do 2021-03-10 21:18:05 The memory usage has bothered me for some time now 2021-03-10 21:19:11 I have my salt stack dockerized, and using the package from edge right now I see about 700mb with about 50 minions 2021-03-10 21:19:37 not great, but it's not system killing either. So at very least 3001 is a little better performant 2021-03-10 21:20:11 (I take that back, that's 3002) 2021-03-10 21:29:03 wsinatra_: are you interested in taking over maintainership for salt? 2021-03-10 21:31:19 sy: https://gitlab.alpinelinux.org/alpine/aports/-/jobs/343199/artifacts/browse/packages/community/x86_64/ 2021-03-10 21:31:41 bl4ckb0ne: try 5.11.5-r1 without initramfs and remove 'modules=...' part from APEND line 2021-03-10 21:32:24 sy: if it works, I can merge that 2021-03-10 21:36:40 awesome I'm testing the master right 2021-03-10 21:36:46 and installing em all 2021-03-10 21:37:22 i am building a kernel on my side with a few mmc config tweaks 2021-03-10 21:38:44 bl4ckb0ne: ah, that is best method to get boot not well supported boards 2021-03-10 21:40:21 alright ill give it a try 2021-03-10 21:44:14 stuck on `[ 0.268793] Waiting for root device /dev/mmcblk1p2...` 2021-03-10 21:44:36 did the mmcblk1 detected 2021-03-10 21:44:52 i dont think so 2021-03-10 21:45:08 could you post log? 2021-03-10 21:45:13 sure 2021-03-10 21:45:27 https://paste.sr.ht/~bl4ckb0ne/43a5a9230e0262a5abd72739a3cd381a79957cc7 2021-03-10 21:47:30 does this board have PCI interface? 2021-03-10 21:47:37 bridge* 2021-03-10 21:49:01 hello 2021-03-10 21:49:25 i dont think so 2021-03-10 21:49:41 it has a sata controller tho 2021-03-10 21:49:45 https://wiki.kobol.io/helios64/hardware/ 2021-03-10 21:50:01 bl4ckb0ne: yes, I know for sata 2021-03-10 21:50:16 troubles with creating debian container on alpinelinux in LXC, error: systemd/openrc related. logfile=ERROR utils - utils.c:lxc_rm_rf:1813 - No such file or directory - Failed to open dir "/sys/fs/cgroup/openrc//lxc.payload.debian" 2021-03-10 21:50:24 > The SATA Controller is connected to RK3399 SoC via 2x lanes PCIe 2.0 offering 10GT/s (8Gbps) of bandwidth. 2021-03-10 21:50:42 ah, then it have PCIe 2021-03-10 21:50:46 yes 2021-03-10 21:51:10 probably this is missing in-kernel and not as module 2021-03-10 21:51:19 https://wiki.kobol.io/helios64/img/hardware/helios64_block_diagram.png 2021-03-10 21:51:53 bl4ckb0ne: I missed that part 2021-03-10 21:52:49 but mmc is ove SDIO as I see on diagram 2021-03-10 21:52:58 tes 2021-03-10 21:53:00 yes* 2021-03-10 21:54:51 how do i run debian container in lxc? 2021-03-10 21:55:04 i run into problem with openrc . i want it to use systemd instead 2021-03-10 21:56:02 bl4ckb0ne: hmm, I enabled SDIO in kernel 2021-03-10 21:56:25 wait im finishing editing the config, ill show you the diff 2021-03-10 21:57:00 bl4ckb0ne: ok 2021-03-10 21:57:22 I didn't enabled this one CONFIG_SDIO_UART 2021-03-10 21:59:09 bl4ckb0ne: ehm, I'm finishing 3rd glass of red wine, so not sure I can continue to work anything useful. if you manage to boot please inform me and if not please inform me of progress 2021-03-10 21:59:22 sure 2021-03-10 21:59:24 what type of red wine? 2021-03-10 21:59:34 ill give another try to my kernel build 2021-03-10 21:59:43 i have a chroot of my mmcblk1 inside armbian 2021-03-10 21:59:54 dry one, I prefer dry wines 2021-03-10 21:59:57 https://tpaste.us/Yolw 2021-03-10 22:00:06 i took what armbian did https://github.com/armbian/build/blob/e6e5438ff9a1d3d64abe6d5c6524a9125b4d4059/config/kernel/linux-rk3399-legacy.config 2021-03-10 22:00:24 and what my friend sent me https://paste.sr.ht/~martijnbraam/dfdeb17f59310a71fa48995c59dae5f24db14ebb#working-config.txt-L5946 2021-03-10 22:00:41 I can continue tomorrow 2021-03-10 22:00:59 sounds good, ill keep you updated here 2021-03-10 22:01:33 enjoy your night! 2021-03-10 22:01:38 yes, for some time I'm thinking to add more options and in-kernel drivers for alpine kernels for arm32 and arm64 2021-03-10 22:02:10 we can also make a linux-helios or linux-rr3399 if that breaks too much things 2021-03-10 22:02:10 I think it makes some sense to have mmc built-in :) 2021-03-10 22:02:22 a lot of people have problems to find correct drivers for their mmc and consoles 2021-03-10 22:03:04 MartijnBraam: I agree but as you know our policy is small (though it sounds strange in current state of alpine) 2021-03-10 22:03:12 hi there MartijnBraam 2021-03-10 22:03:16 o/ 2021-03-10 22:03:41 it's not like there's that many mmc driver, the other way would be having them all in initramfs to make it work which doesn't make things smaller and simpler 2021-03-10 22:04:02 I guess the other hardware loads from sata/nvme instead? 2021-03-10 22:04:33 yes, and I already set ext4 and f2fs and some other drivers in-kernel and not as modules 2021-03-10 22:04:50 ah fuck i think i dont have the ext4 in kernel in my branch 2021-03-10 22:05:06 oof 2021-03-10 22:05:27 yeah i have it 2021-03-10 22:05:30 heck 2021-03-10 22:05:32 I have even uas in-kernel for linux-gru to boot over usb sata disk 2021-03-10 22:05:33 i killed my build for nothing 2021-03-10 22:05:56 ACTION relaunched abuild -r 2021-03-10 22:05:57 so uas is in-kernel but eMMC drivers aren't 2021-03-10 22:06:37 MartijnBraam: :) this is only for linux-gru (gru-kevin chromebook) 2021-03-10 22:06:49 ah 2021-03-10 22:07:00 and yes, there is also mmc driver in-kernel 2021-03-10 22:07:38 i get this error: Failed to open dir "/sys/fs/cgroup/openrc//lxc.payload.debian-1 2021-03-10 22:07:40 interesting, lets see what happens with linux-gru 2021-03-10 22:08:33 MartijnBraam: it is tailored only for this device, didn't tested on any other 2021-03-10 22:09:44 can abuild do incrementals ? 2021-03-10 22:09:56 maybe you will also interested on linux-elm for pmos, tailored for elm familly of chromebooks 2021-03-10 22:10:18 what are the elm chromebooks? 2021-03-10 22:10:33 oh mediatek? 2021-03-10 22:11:00 mediatek, huh 817... forgot after so much wine 2021-03-10 22:11:10 can i post about 8 lines log_ 2021-03-10 22:11:11 ? 2021-03-10 22:11:43 yeah I see it contains only mediatek dtbs 2021-03-10 22:11:46 bl4ckb0ne: yes if you don't change APKBUILD 2021-03-10 22:11:52 help please 2021-03-10 22:11:54 mediatek is not good for my sanity 2021-03-10 22:12:11 good to know if my helios crash again during build 2021-03-10 22:12:20 MartijnBraam: elm chroembooks are quite good devices 2021-03-10 22:12:23 i have a dmesg -w on a ssh connection to monitor it 2021-03-10 22:12:50 for chromebooks maybe, for phones it's a guarantee there's no mainline support :( 2021-03-10 22:13:01 true 2021-03-10 22:13:14 also found a beautiful mediatek device that does powermanagement by hotplugging cpu cores :D 2021-03-10 22:13:23 but PBP is not phone afaik :) 2021-03-10 22:13:38 no it isn't, that's why it's not well supported in postmarketOS 2021-03-10 22:18:16 eeehm where's the kernel image in linux-gru? 2021-03-10 22:18:40 alpine testing repo 2021-03-10 22:19:03 no where's Image.gz 2021-03-10 22:19:08 or similar 2021-03-10 22:20:30 I think on mirrors 2021-03-10 22:21:10 I have no trouble getting the .apk, it doesn't contain a kernel 2021-03-10 22:21:21 here is notes how to install linux-elm on elm chromebook https://arvanta.net/alpine/install-alpine-on-acer-r13-chromebook/ 2021-03-10 22:22:00 no, it contains vmlinux.kpart which is packed kernel for chromebooks 2021-03-10 22:22:53 look at above url, as hints how to install on gru-kevin 2021-03-10 22:23:22 replace linux-elm with linux-gru 2021-03-10 22:23:49 i need help with lxc debian on alpine linux 2021-03-10 22:24:16 except I don't have a chromeOS bootloader on a pinebook pro 2021-03-10 22:24:24 and also not on the helios64 2021-03-10 22:25:49 I wrote these two kernels are tailored for chromebooks 2021-03-10 22:26:38 but, too late is now, good night to all 2021-03-10 22:26:39 yeah I see, I expected to be able to use it for other rk3399 devices 2021-03-10 22:26:47 I guess I'll just steal the config :P 2021-03-10 22:27:02 sure, do that 2021-03-10 22:37:46 is there a way to add cpu temperature sensor in afc? 2021-03-10 22:38:19 i mean acf 2021-03-10 22:49:44 Hi. I would like to install Alpine Linux on VPS with kvm. What image is better to download? Standard, Virtual or something other? 2021-03-10 22:53:21 I'd use standard, you could always just install linux-virt later on if you need it 2021-03-10 22:53:31 But the default kernel should come with everything you need 2021-03-10 23:00:28 You're using Standard on VPS server? 2021-03-10 23:51:34 Does anyone know how to get abuild to ignore bad certificates while fetching package sources? Perl seems to have let theirs expire (or maybe they haven't and something is wrong with my install?) 2021-03-10 23:51:58 i'd be careful with the perl thing, there was news on that a month ago 2021-03-10 23:52:10 forget what the dtails were, but i think they lost control of a domain 2021-03-10 23:52:46 It's trying to download from cpan.org 2021-03-10 23:52:59 Maybe there's a mirror somewhere I can use instead? 2021-03-10 23:53:03 ah 2021-03-10 23:53:09 the cert looks Ok to me 2021-03-10 23:53:22 probably an option to skip cert checking, but I would advise against that 2021-03-10 23:53:48 Well I can't build the package otherwise and from what I can tell everything is otherwise OK 2021-03-10 23:53:56 It can download things just fine from github for instance 2021-03-10 23:56:05 Oh that's weird 2021-03-10 23:56:13 It wgets just fine on my desktop machine also running alpine 2021-03-10 23:56:51 This alpine install isn't done through the normal methods, it was bootstrapped manually, maybe I missed something involving CA certs? 2021-03-10 23:57:33 err..i've heard of this not long ago 2021-03-11 00:01:55 Also I couldn't find any option in abuild to skip cert checking 2021-03-11 00:02:05 Not sure if I just missed it or there actually isn't one 2021-03-11 00:32:48 Segfault0: date is wrong on the machine where it doesn't work? that can cause cert verification failures 2021-03-11 00:38:25 Well the time is saying my local time but it's also saying UTC which means it's actually 10 or 11 hours ahead, could that be enough to cause problems? 2021-03-11 00:46:21 0.0.0.0:8088->8088/tcp it happened again! 2021-03-11 00:50:23 gotto stop all containers and restart dockerd to get it to work 2021-03-11 01:06:28 Segfault0: possibly? 2021-03-11 01:07:19 I'll do a few things and then try fixing the time and see if that fixes it 2021-03-11 01:09:23 Well what do you know, that did actually fix it 2021-03-11 01:09:40 i think curl prints a reasonable error message in this case 2021-03-11 01:09:48 i bet busybox wget prints something stupid 2021-03-11 01:10:16 busybox just told me the cert expiry check failed or something along those lines 2021-03-11 01:30:23 what did i miss 2021-03-11 02:30:41 hello ladies 2021-03-11 02:30:59 I m little bit lost with intel gpu crap for thinkpad t420 2021-03-11 02:31:21 not sure if have to still adding that kms module bc. somehow it disappeared from wiki 2021-03-11 02:31:52 I have hope, that for a time I did not use alpine, somebody already deleted whole useless wiki :) 2021-03-11 02:32:22 not sure what kind of libva crap should I install for sure 2021-03-11 02:32:55 so if anybody using some sandy bridge cpu with igpu I m listening :) 2021-03-11 02:39:25 only wat I got from wiki was, 'modern intel gpu should work out of box...' while my is not so modern ;) 2021-03-11 09:39:44 Hmm, so my next major problem package for PPC is looking like it's going to be perl 2021-03-11 09:40:02 Seems to fail a few of the tests at the end which means it refuses to finish putting everything together 2021-03-11 11:35:22 yo 2021-03-11 13:29:54 hello, does lualatex work for anyone? I tried alpine:3.13 and alpine:edge docker images, ran apk add texlive-full inside. pdflatex works fine, lualatex starts with "kpathsea: Running mktexfmt lualatex.fmt", dumps the entire output of that and finishes with "(Fatal format file error; I'm stymied)" 2021-03-11 13:51:48 `ln -s /usr/bin/luahbtex /usr/local/bin/lualatex` fixes the issue. eg. archlinux also symlinks lualatex to luahbtex. 2021-03-11 13:54:53 sammko: I had issues with lualatex the other day but eventually gave up and installed debian in a chroot 2021-03-11 13:55:00 I'll have a look and thanks for the hint 2021-03-11 16:44:55 mps: no luck with my kernel 2021-03-11 16:45:06 still no mmcblk1 showing in /dev 2021-03-11 16:46:12 bl4ckb0ne: we should ask on #linux-rockchip, probably someone there could help 2021-03-11 16:47:14 but I have mild headache all day and I'm not in good mood for asking now 2021-03-11 16:47:24 i wonder if helios has a chan 2021-03-11 16:47:32 ill take care of it and keep you updated 2021-03-11 16:47:57 also this we can ask on #linux-rockchip 2021-03-11 16:53:00 bl4ckb0ne: btw, kernel 5.11.6 is released few hours ago 2021-03-11 16:53:22 nice 2021-03-11 16:53:37 anything good? 2021-03-11 16:54:00 though i have more hopes in 5.12-rc3 which will be released on monday 2021-03-11 16:54:52 5.11.6 have one commit about quirks in sdhci, so maybe this will help 2021-03-11 16:55:30 yeah i saw that 2021-03-11 17:08:36 bl4ckb0ne: so armbian and pmos kernels boots? 2021-03-11 17:09:26 yes 2021-03-11 17:09:44 and boot are 5.11.x 2021-03-11 17:10:06 hmm no 2021-03-11 17:10:26 armbian is 5.10 with a bunch of patches 2021-03-11 17:10:26 ah, which they are 2021-03-11 17:10:30 pmos is 5.6 iirc 2021-03-11 17:10:42 uh, this is outdated 2021-03-11 17:11:06 https://github.com/armbian/build/tree/master/patch/kernel 2021-03-11 17:11:10 yep 2021-03-11 17:11:56 yes, I looked at armbian, but these patches are already in 5.11 2021-03-11 17:13:10 I have to find armbian kernel config for this board 2021-03-11 17:13:18 https://github.com/armbian/build/blob/e6e5438ff9a1d3d64abe6d5c6524a9125b4d4059/config/kernel/linux-rk3399-legacy.config 2021-03-11 17:13:45 but im not 100% sure its this one 2021-03-11 17:14:16 Linux/arm64 4.4.213 Kernel Configuration 2021-03-11 17:14:32 https://github.com/armbian/build/blob/master/config/kernel/linux-rockchip64-current.config might be this one 2021-03-11 17:15:22 Linux/arm64 5.10.21 Kernel Configuration 2021-03-11 17:15:45 so not sure will it work on helios64 2021-03-11 17:16:29 > Linux helios64 5.10.16-rockchip64 #21.02.2 SMP PREEMPT Sun Feb 14 21:35:01 CET 2021 aarch64 GNU/Linux 2021-03-11 17:16:31 on armbian 2021-03-11 17:16:50 bl4ckb0ne: if I build kernel with defconfig aarch64 are you ready to run it 2021-03-11 17:17:40 or maybe you can do this 2021-03-11 17:18:02 i am giving a run at another batch of config changes 2021-03-11 17:18:07 then we can do defconfig 2021-03-11 17:18:54 ok, will wait for your results 2021-03-11 17:19:39 in meantime I'm going to prepare yet another coffee, maybe I will feel better after it 2021-03-11 17:19:47 just gotta find a way to use the sleepy 4 remaining cores 2021-03-11 17:20:19 -j8 :) 2021-03-11 17:21:09 what's abuild default? 2021-03-11 17:21:25 defined in /etc/abuild.conf 2021-03-11 17:21:56 ah jobs=2 2021-03-11 17:22:23 export JOBS=8 2021-03-11 17:22:39 set it to 6 2021-03-11 17:22:46 only have 6 cores on that shite 2021-03-11 17:23:44 yes, I know, rk3399 have 4 a72 and 2 a57 (from head) 2021-03-11 17:24:53 my time on the last build was ridicule 2021-03-11 17:25:14 real 10h 45m 27s 2021-03-11 17:25:16 user 16h 3m 59s 2021-03-11 17:25:18 sys 5h 24m 29s 2021-03-11 17:26:13 huh 2021-03-11 17:26:31 you use mmc as FS 2021-03-11 17:26:43 YES 2021-03-11 17:26:46 yes* 2021-03-11 17:26:53 i should have a decent one though 2021-03-11 17:27:08 heh, I had these experience when built firefox 2021-03-11 17:27:30 but ssd disk over usb-c port is quite fast 2021-03-11 17:27:47 i have a whacky setup 2021-03-11 17:27:59 i mounted the mmc as a chroot on the emmc 2021-03-11 17:29:23 that's good, you start build, take kids and go on long walk and some other real life things :) 2021-03-11 17:29:49 building inbetween naps :D 2021-03-11 17:30:14 lets hope this build finished before her afternoon nap 2021-03-11 17:30:34 :) 2021-03-11 17:30:48 going to 'build' coffee, bbl 2021-03-11 19:25:33 bl4ckb0ne: hmm, arm64 defconfig doesn't look quite big and I see some interesting options/drivers are not enabled 2021-03-11 19:25:57 anyway, I built it and can upload if you want to try it 2021-03-11 19:51:45 Hi all :-) 2021-03-11 19:52:18 I've got a brand new raspberry pi waiting for it's alpinelinux system 2021-03-11 19:52:56 The default system for raspberry-pi is using lbu 2021-03-11 19:53:40 If I remount rw and add an apk to apks folder, will it be "installed" ? 2021-03-11 19:56:17 I want to preserve mem for runtime usage 2021-03-11 21:09:18 mps: still building mine :D 2021-03-11 21:09:24 but sure, i can give it a try tonight 2021-03-11 21:23:23 bl4ckb0ne: uploading it 2021-03-11 21:23:39 neat 2021-03-11 21:23:40 this one is simple tarball 2021-03-11 21:23:49 i should be done by tonight i belive for my kernel 2021-03-11 21:25:06 you will have to untar it to your root partition, but before that mount boot partition on your root FS boot dir 2021-03-11 21:25:32 and don't use initramfs because there is not one 2021-03-11 21:26:09 bl4ckb0ne: https://dev.alpinelinux.org/~mps/linux-5.11.6.tar.gz 2021-03-11 21:26:15 ty 2021-03-11 21:26:16 this is big kernel 2021-03-11 21:26:30 ill update you later on tonight 2021-03-11 21:27:19 I guess I asked a noob question :-) ... I'm trying to find a way to mix the best of ram and disk usage. I did try mounting a loopback image on / using overlayfs but I must have missed something 2021-03-11 21:27:35 so, 'mount /dev/mmcblk1p2 /mnt' 'mount /dev/mmcblk1p1 /mnt/boot' tar xf linux-5.11.6.tar.gz -C /mnt' 2021-03-11 21:27:40 for example 2021-03-11 21:28:03 ty 2021-03-11 21:28:10 bl4ckb0ne: I will probably be in deep sleep when you start this 2021-03-11 21:28:50 ack 2021-03-11 21:28:56 but I hope you will see what is needed for KERNEL and FDTDIR params in u-boot config 2021-03-11 21:32:10 oh wait im already in my chroot, everything is mounted 2021-03-11 21:32:14 i can already install it 2021-03-11 21:33:29 anything new in the dtfs? 2021-03-11 21:33:36 dtbs* 2021-03-11 21:33:37 no 2021-03-11 21:34:15 just set fdtir to dtbs-5.11.6 2021-03-11 21:34:33 and kernel to vmlinux-5.11.6 2021-03-11 21:34:36 already did :D 2021-03-11 21:34:42 nice :) 2021-03-11 21:35:00 that is made by allconfig, lets hope now 2021-03-11 21:35:32 nice 2021-03-11 21:36:30 alright its installed 2021-03-11 21:36:35 gonna wait for mine and test both 2021-03-11 21:39:22 good, please report what's result, probably i will read it tomorrow 2021-03-11 21:39:34 well get it eventually! 2021-03-11 21:39:46 I hope 2021-03-11 21:40:17 i was honestly short of about 500$ to make a x86 nas 2021-03-11 21:40:53 and how much cost this one 2021-03-11 21:41:18 about 300$USD 2021-03-11 21:41:27 but armbian has a lot of trouble with zfs 2021-03-11 21:41:36 ive waited about 3 months for them to sort it out 2021-03-11 21:41:38 no luck 2021-03-11 21:42:49 afaik zfs needs a lot of memory 2021-03-11 21:43:12 they fixed iiirc 2021-03-11 21:43:39 ah, that's good news, maybe I should reconsider it 2021-03-11 21:43:57 ill tell you how it goes :) 2021-03-11 21:44:10 sure 2021-03-11 21:45:32 but no, I will not use FS driver which is not in mainline, no chances 2021-03-11 21:46:11 why not? 2021-03-11 21:46:58 enough is to fight bugs in mainline drivers, don't want to fight with those outside 2021-03-11 21:48:00 you can get away with a lot less than the recommended amount of memory for zfs if you don't use many of the features. depends on what your needs are 2021-03-11 21:48:28 shy of 4gb on the helios64 2021-03-11 21:49:10 loosh: for me enough would be dm raid 1 (mirror) 2021-03-11 21:49:35 once you start flipping on dedupe, encryption, etc. you should certainly pay at least a little bit attention to the recommended amount 2021-03-11 22:14:00 zfs still very much needs a ridiculous amount of memory, to say the least. 2021-03-11 22:14:23 if you expect it to not suck ass at 4GB you are going to be incredibly disappointed. 2021-03-11 22:35:39 I mean yeah, don't expect performance if you are under the rec amount. but it *will* work :) 2021-03-11 23:05:45 for extremely generous values of 'work' 2021-03-11 23:27:49 I tried with 512 MB and fbsd literally wouldn't boot 2021-03-11 23:28:32 allegedly you can tune it but I took it as a sign that it was not meant to be 2021-03-11 23:28:38 er, wouldn't install 2021-03-11 23:38:23 mps: what about making a kernel for rockchip https://gitlab.com/postmarketOS/pmaports/-/merge_requests/2021/diffs#a67f76f6974841448a6533d9f69067a599d34a65 2021-03-11 23:48:05 mps: both kernels hang at "starting kernel" 2021-03-12 01:02:21 postfix in 3.13 does not seem to support the map type hash anymore. Is there a replacement for that? 2021-03-12 01:02:46 postmap hash:postfix/aliases 2021-03-12 01:02:46 "postmap: fatal: unsupported dictionary type: hash. Is the postfix-hash package installed?" 2021-03-12 01:02:54 That is the error i get 2021-03-12 01:58:59 read release notes 2021-03-12 03:17:11 mps: i fucked up somewhere, your 5.11.6 boots https://paste.sr.ht/~bl4ckb0ne/34dfa158062025ab2cac680dcde96a7f2748ed51 2021-03-12 03:17:55 but my linux-rockchip lands to a shell :D 2021-03-12 03:19:19 https://paste.sr.ht/~bl4ckb0ne/33724fb92907cf9b8e3847308af8047c43e35679 2021-03-12 03:37:46 mps: got your 5.11.6 to finish init too https://paste.sr.ht/~bl4ckb0ne/98ff29256eb50cfc0515ae3ff8eafd7b808723a2 2021-03-12 03:37:55 i just cant get a fucking serial login on both 2021-03-12 03:45:58 I see RK808 and PCIe devices, is that an RK3399 board? 2021-03-12 03:48:05 Oh it is, neat 2021-03-12 03:48:43 rk3399 ye 2021-03-12 03:48:58 https://wiki.kobol.io/helios64/hardware/ 2021-03-12 03:49:10 Yeah I saw kobol in the kernel log 2021-03-12 03:51:05 Wow that thing has a ton of IO for an RK3399, they must really be pushing the PCIe port 2021-03-12 03:51:06 huh no default inittab? 2021-03-12 03:53:04 got 5*4TB plugged in it 2021-03-12 03:55:15 > Linux snorlax 5.11.6 #2 SMP PREEMPT Thu Mar 11 19:18:36 UTC 2021 aarch64 Linux 2021-03-12 03:55:17 we did it mps 2021-03-12 03:56:18 > Linux snorlax 5.11.6-0-rockchip #1-Alpine SMP Thu, 11 Mar 2021 23:53:59 +0000 aarch64 Linux 2021-03-12 03:58:25 weird /sys is empty with your kernel 2021-03-12 03:58:41 and the hdd dont show in mine 2021-03-12 03:58:51 but enough for one day 2021-03-12 04:16:37 I sure do love when random things like perl, m4 and python3 have tests that fail on ppc with no results showing up when I search 2021-03-12 04:16:49 Guess I'll just have to disable checks on those and hope for the best 2021-03-12 04:17:06 Maybe picking a 32 bit big endian architecture for my first port wasn't the best idea :P 2021-03-12 05:34:11 sy: did you manage to test salt yet? 2021-03-12 07:29:12 Hmm, I tried to compile asciidoc and it complains about the hash not matching, which is weird because it's downloading from github which normally doesn't have problems with that sort of thing 2021-03-12 07:30:24 And if I update the hash to use the file it downloads it complains about there being no configure.ac file even though there clearly is one 2021-03-12 08:10:34 bl4ckb0ne: good news. though I wonder why 5.11.5 didn't mount mmc, it have more options/drivers than 5.11.6 2021-03-12 08:12:52 bl4ckb0ne: re: console, I forgot in that 'frenzy' about kernel that inittab should be changed to use a/getty on ttyS2 instead of ttyS0 and /etc/securetty should also contain ttyS2 2021-03-12 08:14:08 actually I think init script should read /proc/console and set these things automagically 2021-03-12 10:55:57 I'm wondering: is there any chance to have firefox with direct alsa support in Alpine? I don't run pulseaudio and at the moment I've to start chromium for video conferences, as firefox is my default browser for most things at the moment 2021-03-12 10:57:03 telmich: it works fine with alsa 2021-03-12 10:57:24 Oh? Not on my system - is there anything specific to configure it? 2021-03-12 10:57:49 no afaik, for me it works out-of-the-box 2021-03-12 10:58:12 I don't have pulseaudio or pipewire daemons 2021-03-12 10:58:38 Interesting. So audio-out works for me, video works, but not the microphone 2021-03-12 10:58:48 Firefox does not detect it 2021-03-12 10:59:47 Just checking https://bbs.archlinux.org/viewtopic.php?id=252378 2021-03-12 11:02:38 mps: you are saying you can use the microphone with alsa, without apulse, without pulseaudio? 2021-03-12 11:05:42 telmich: not sure, didn't tried microphone long time, but rest works 2021-03-12 11:09:25 Firefox with mic on alsa has not worked for a while, audio only (from my experience) 2021-03-12 11:10:59 mps: do you mind giving it a try on https://talk.ungleich.ch/random-url-with-yourself-only ? 2021-03-12 11:11:50 For me, jitsi/firefox says it cannot access the mic. However, chromium next to it can - so I assume it's not a permission issue 2021-03-12 11:12:11 I think this is the bug report that I saw that originally made me find the problem: https://bugzilla.mozilla.org/show_bug.cgi?id=1345661 2021-03-12 11:22:40 telmich: hm, 'Failed to access your microphone' 2021-03-12 11:22:51 so, it doesn't work 2021-03-12 11:23:14 I was fearing that 2021-03-12 15:25:01 metbsd: please create an issue against acf-alpine-baselayout https://gitlab.alpinelinux.org/acf/acf-alpine-baselayout/-/issues if you want cpu temperature added to ACF 2021-03-12 15:25:11 or, feel free to create an MR 2021-03-12 15:26:46 how about dockerd bug 2021-03-12 15:27:21 sometimes need to restart dockred to make container network to work 2021-03-12 16:02:24 Which crond is used in Alpine? Does it have the concept of running a cron job at startup? 2021-03-12 16:06:42 adhawkins: busybox crond 2021-03-12 16:07:03 adhawkins: I don't think it does 2021-03-12 16:08:08 Ok, thanks ikke. I guess I can just start the service in my container entrypoint. 2021-03-12 16:13:49 mps: i think reinstalling openrc generated a /etc/inittab, i managed to login yesterday 2021-03-12 16:16:32 i still have a couple of things that are not properly initialised in the init process, and the sata controller to take care of 2021-03-12 16:50:23 bl4ckb0ne: good. maybe you can write some short notes of your trial to get alpine login on this board 2021-03-12 16:50:55 sure 2021-03-12 16:51:02 alpine wiki? 2021-03-12 16:51:28 however you like 2021-03-12 16:52:45 I searched u-boot for helios64 but don't see it upstreamed so we can't make 'promise' that alpine works on this board 2021-03-12 16:52:51 yeah 2021-03-12 16:53:00 idk if kobol is working to make this upstream 2021-03-12 16:53:18 maybe is better to write somewhere unofficial for alpine 2021-03-12 16:53:27 https://github.com/kobol-io/u-boot/commit/29d63b29550818992e3bcdb1ceb2a0db49d395cc 2021-03-12 16:53:47 i dont have a personal space atm, I can write offline and wait for the patch to be upstream 2021-03-12 16:54:56 whatever you do, please send it to me 2021-03-12 16:55:03 sure 2021-03-12 16:56:11 after this experience I'm thinking about specific kernel packages for specific SOCs, and not big one 'to rule them all' 2021-03-12 16:57:57 yeah i definetly think a linux-rockchip would be a good idea 2021-03-12 16:58:10 maybe keep the linux-edge aarch64 to defconfing 2021-03-12 16:58:24 something like this is already in my head 2021-03-12 16:58:39 i sadly nuked my sdcard before being able to push what i did 2021-03-12 16:58:45 but i mostly took https://gitlab.com/postmarketOS/pmaports/-/merge_requests/2021 to aports 2021-03-12 16:59:02 https://gitlab.com/postmarketOS/pmaports/-/blob/ac44321fc24b758a6d229ee70757b926a5a570ad/main/linux-postmarketos-rockchip/config-postmarketos-rockchip.aarch64 this exactly 2021-03-12 16:59:13 and brought it to 5.11.6 2021-03-12 16:59:27 actually for now is just u-boot part is important, rest we can recreate from head 2021-03-12 16:59:37 yes 2021-03-12 17:51:17 hi all, performance problem with git.alpinelinux.org <- Eq: https://git.alpinelinux.org/aports/tree/main/postfix?h=3.13-stable <- clic on "log" (right APKBUILD) and... wait... wait... 502 bad gateway 2021-03-12 18:08:19 hi! just installed the minimal alpine linux rootfs inside an x86_64 kvm vm, added grub2 and the linux-virt-5.10 kernel. when trying to boot it, it gets stuck in initramfs. /dev/vda* devices are detected, but somehow the ext4 module is not being loaded, i can load it manually though. any ideas how to fix this? 2021-03-12 18:10:23 Hi. I'm doing so test between groupadd/useradd and addgroup/adduser from busybox. I can not find from busybox how I can set a primary group to the user created with adduser 2021-03-12 18:10:31 so/some 2021-03-12 18:27:57 https://dpaste.com/DV94BZQZG really strange, so my user is in the nogroup but at the same time has a primary group 2021-03-12 18:40:34 btw. regarding groups, should be my regular user be in 'video, audio, input' groups using xorg? or not necessary some of them? 2021-03-12 20:29:55 hi there, just a quick question, why is there no linux-edge x86 (32 bit) flavour? is there no one needing it and therefore it weren't build? 2021-03-12 20:30:27 weiti: there is 2021-03-12 20:30:55 http://dl-cdn.alpinelinux.org/alpine/edge/releases/x86/ 2021-03-12 20:32:13 ikke: ok, but with which kernel? shouldn't i see a linux-edge package with x86 architecture on pkgs.alpinelinux.org? 2021-03-12 20:32:53 ah, I misunderstood you 2021-03-12 20:32:55 the linux-edge kernel 2021-03-12 20:33:18 mps would know 2021-03-12 20:33:33 yes, sorry meant only the kernel (5.11.x) ... 2021-03-12 20:43:58 ikke: weiti: I don't have real hardware to test it and don't want to keep something which I can't test, that is the reason 2021-03-12 20:45:23 mps: ok, understood. 2021-03-12 20:46:45 mps: would it be ok to send an merge-request for activate Segment Routing on the linux-edge flavours (especially x86_64 flavours)? 2021-03-12 20:47:41 is there a way to search package with some keywords in description? 2021-03-12 20:48:20 Had an issue created for the linux-lts packages as well (#12490) and would love to see it also in the edge kernels as 5.11.x has some nice extensions i need in some lab/testings 2021-03-12 20:48:44 weiti: I already intended to enable it but forgot, sorry 2021-03-12 20:49:11 mps: ah ok, no problem. 2021-03-12 20:49:26 (need to start writing notes around) 2021-03-12 20:53:54 That's why I bought a reMarkable2 2021-03-12 20:54:01 Has really helped 2021-03-12 20:54:13 (Semi relevant because it's based on Linux.) 2021-03-12 20:54:15 :) 2021-03-12 20:57:40 where would the cu command be in alpine? for call up 2021-03-12 20:58:16 https://pkgs.alpinelinux.org/contents?file=cu&path=&name=&branch=edge 2021-03-12 20:59:04 not sure if that's the same that you mean 2021-03-12 20:59:41 I think so but now I am trying to figure out how you made the web page do that 2021-03-12 21:00:31 https://pkgs.alpinelinux.org/contents 2021-03-12 21:46:20 Hi guys, quick question 2021-03-12 21:46:34 How do I pass environment variables to an openrc service? 2021-03-12 21:47:00 bitblt: /etc/conf.d/ 2021-03-12 21:47:12 anything you export there is available to the service 2021-03-12 21:47:50 I see that the openrc-run service file that i'm using has this line: 2021-03-12 21:48:23 command_args="${K3S_EXEC} ${K3S_OPTS} >>${K3S_LOGFILE} 2>&1" 2021-03-12 21:48:38 You can set those in the same file 2021-03-12 21:48:44 Do I need to export or just set them in the relevant /etc/conf.d/k3s 2021-03-12 21:48:49 Just set it 2021-03-12 21:49:12 that file is sourced 2021-03-12 21:49:36 I saw that the relevant /etc/conf.d/k3s file, exports PATH but just sets K3S_EXEC and K3S_OPTS 2021-03-12 21:49:42 What is the difference? 2021-03-12 21:51:08 just setting it makes it only availabe in the init.d/file 2021-03-12 21:51:20 exporting it will make it available to the service 2021-03-12 22:18:38 seems to be working thanks! 2021-03-12 22:48:15 ikke: a couple of months ago I was asking about bulk building packages and you suggested aports-build, is there a guide on how to use that anywhere? 2021-03-12 22:52:52 I tried just running the command and it didn't really repond to any flags, also it errors out on line 12, complaining about being unable to create "/var/run/mqtt-exec.aports-build/aports-build.pid" because the directory doesn't exist 2021-03-12 23:02:21 hello 2021-03-12 23:53:44 So, testing nftables, I have written a custom.nft file which has been put in /etc/nftables.d/myfile.nft, but now I wonder if it's well written. Can not find any information about how the include work 2021-03-12 23:54:25 do I need the #!/usr/sbin/nft -f line at the beginning ? 2021-03-12 23:54:52 yes 2021-03-12 23:55:08 do I need to write again chain input { type filter hook input priority 0; policy drop; or chain input { is enough 2021-03-12 23:55:15 no idea 2021-03-12 23:55:30 i dont even know what nft is but i know that is how interperters work 2021-03-12 23:55:41 you need to specify the interperter in the bang 2021-03-13 00:04:15 halala, why open-source is so closed to the end-user 2021-03-13 00:04:26 so badly explained 2021-03-13 00:13:39 ok, so no need to write again the type filter hook input priority 0; policy drop; 2021-03-13 00:16:43 I don't understand why on the redhat documentation they are saying to chmod u+x the script when nft -f myfile works as expected 2021-03-13 00:16:53 without that 2021-03-13 06:34:35 interesting questions in backlog (: 2021-03-13 07:19:50 these is old saying 'Unix is a user friendly, though it (he) is picky about its (his) friends' 2021-03-13 07:20:04 s/these/there/ 2021-03-13 08:58:56 sooo, i had this problem yesterday where booting an alpine with linux-virt kernel got stuck in initramfs because the ext4 module didnt get loaded for some reason 2021-03-13 09:00:11 i fixed it myself now with a band-aid solution, insmod'ing the ext4 module and its dependencies at the start of /usr/share/mkinitfs/initramfs-init 2021-03-13 09:01:44 however i assume this will be overridden whenever mkinitfs gets updated, so i still would prefer a "proper" solution 2021-03-13 09:01:53 any ideas? 2021-03-13 09:06:28 features="ata base ide scsi usb virtio ext4" in /etc/mkinitfs/mkinitfs.conf 2021-03-13 09:06:58 and run mkinitfs to recreate initramfs 2021-03-13 09:07:45 mps: i have done this, doesn't automatically load ext4, however. 2021-03-13 09:08:07 then it must be bug somewhere 2021-03-13 09:09:11 maybe add 'rootfstype=ext4' kernel cmdline in bootloader 2021-03-13 09:09:29 probably 2021-03-13 09:09:48 i could give that a try 2021-03-13 09:09:51 but, it should work without this, though didn't tested 2021-03-13 09:10:30 but shouldnt be features="base ext4 virtio" be enough to indicate that i want virtio and ext4 drivers? 2021-03-13 09:11:34 depends on you setup, I just pasted from one of my VM with -virt kernel 2021-03-13 09:11:54 during my debugging i figured out that i additionally need to load some dependencies: crc16, mbcache and jbd2 2021-03-13 09:12:29 mps: i don't have control over that, its a cloud vm based on kvm 2021-03-13 09:12:29 modprobe ext4 should do that automatically 2021-03-13 09:12:58 yes, thats what i did to make it work in the initramfs emergency shell 2021-03-13 09:13:14 modprobe ext4, then mount /dev/vda1 /sysroot, then exit 2021-03-13 09:13:15 ah, cloud provider, heh 2021-03-13 09:13:44 they like to pretend smartness :) 2021-03-13 09:14:14 yeah not really smart, just cheap AF and i wanted to make use of it 2021-03-13 09:14:25 alpine linux seemed to be a good fit :) 2021-03-13 09:14:36 linode is one of the best for qemu VM (and linode is our sponsor) 2021-03-13 09:15:35 but, virt alpine works quite fine on a lot of fog^Wcloud providers 2021-03-13 09:16:18 it does work fine on mine too now, after fixing the initramfs issue 2021-03-13 09:16:40 well, tell them to fix it 2021-03-13 09:17:06 I wonder if this is an issue with nlplug-findfs 2021-03-13 09:17:26 https://gitlab.alpinelinux.org/alpine/aports/-/issues/12325 2021-03-13 09:18:02 ikke: zocker told that it managed to set it manually 2021-03-13 09:18:25 ikke: block devices are found in my case, ext4 module is present in the image as well, its just not getting loaded automatically 2021-03-13 09:19:30 zocker: anyway try with 'rootfstype=ext4' 2021-03-13 09:20:18 mps: will do. give me half an hour or so and i will report if this works :) 2021-03-13 09:22:06 no worries, I give you even full day :) 2021-03-13 09:30:31 btw: i did the installation via chroot with the minimal rootfs (which is *really* minimal by the way :D) 2021-03-13 09:30:56 so i added grub and linux-virt via apk inside the chroot 2021-03-13 09:31:10 i was thinking that i might be missing some dependency as well? 2021-03-13 09:41:35 mps: just added "rootfstype=ext4" to grub cmdline and removed my insmod hack 2021-03-13 09:41:41 seems to work :) 2021-03-13 10:10:50 that basically solves the issue for me, but i still consider this a bug 2021-03-13 10:11:35 shouldn't features="base ext4 virtio" in mkinitfs.conf imply that i can actually boot from a ext4 rootfs on a virtio block device? 2021-03-13 10:12:18 zocker: the only thing that 'feature' does is make sure that the modules are included in the initramfs 2021-03-13 10:12:49 0https://gitlab.alpinelinux.org/alpine/mkinitfs/-/blob/master/features.d/ext4.modules 2021-03-13 10:13:49 i see 2021-03-13 10:14:04 so specifying rootfstype in the bootloader is actually mandatory then? 2021-03-13 10:14:24 https://gitlab.alpinelinux.org/alpine/mkinitfs/-/blob/master/initramfs-init.in#L460 2021-03-13 10:14:27 yes 2021-03-13 10:14:31 i probably should have a look at the regular alpine installer and see what it does 2021-03-13 10:14:44 You see it modprobes the rootfstype there 2021-03-13 10:15:54 alright 2021-03-13 10:16:21 i just need to remember that for my next alpine installation then 2021-03-13 10:16:23 thanks! 2021-03-13 10:19:22 zocker: initramfs use busybox init and mount which probably don't know to load automatically proper module for root FS 2021-03-13 10:19:59 huh, no 2021-03-13 10:20:21 I have f2fs as root fs and initramfs mounts it quite fine 2021-03-13 10:20:46 mps: did you specify rootfstype in the bootloader? 2021-03-13 10:21:23 no 2021-03-13 10:21:55 that is, on real machine 2021-03-13 10:25:13 hmm 2021-03-13 10:25:40 i think i'll play around with it inside a local qemu 2021-03-13 10:35:20 zocker: miniroot is intended for docker and similar, not good idea to start from it for 'real' machine (though I did few times :) ) 2021-03-13 10:37:00 sure, but it was the best option. the installer iso from the provider was outdated and also choked on the fact that the vm has only 1GB disk space :D 2021-03-13 10:40:47 root 2021-03-13 10:41:04 lol, wrong window :D 2021-03-13 10:48:23 so, just tried the current virt iso inside qemu. it indeed adds rootfstype=ext4 to the bootloader. 2021-03-13 11:20:54 The "docs" link on the website 404s.  Is there a mirror somewhere?  https://docs.alpinelinux.org/ 2021-03-13 11:28:37 piousminion: working for me just 2021-03-13 11:29:02 fine 2021-03-13 11:29:41 It is now.  odd 2021-03-13 11:30:47 Is there an apk option or apk util that lets you search all packages for a specific file they provide? 2021-03-13 11:32:53 apk-file does that 2021-03-13 11:33:57 awsome, thank you. 2021-03-13 12:01:00 I usually use "useradd", but requires "shadow".  Are there any cons to using shadow? 2021-03-13 12:04:09 I figure there might be a good reason it's not the default. I have no idea though 2021-03-13 12:10:20 piousminion: 'keep it small simple' 2021-03-13 12:11:29 adduser is quite enough 2021-03-13 12:13:35 Is does the job.  My question is more about what shadow really does. 2021-03-13 12:14:45 it contains some extra tools, chsh comes to mind 2021-03-13 12:22:42 su is also a useful thing in shadow 2021-03-13 12:23:27 I'm following the docs and it says to install "alpine-desktop" to get the default DM/DE, but that package doesn't exist. 2021-03-13 12:23:39 just do an `apk info -L shadow` to see what it brings. 2021-03-13 12:24:39 yZ5vlALg86lP: busybox-suid have su 2021-03-13 12:26:41 I'm guessing "alpine-desktop" is a future project or a canceled one. 2021-03-13 12:27:39 more like a nonsense 2021-03-13 12:28:24 lol 2021-03-13 12:28:48 piousminion: That package has been removed, just install gnome, kde, xfce etc. if you want a desktop 2021-03-13 12:29:25 Working on that now. :) 2021-03-13 12:29:55 desktop is too generic term which can means a lot, or nothing 2021-03-13 12:30:26 but on microsoft or apple desktop makes sense 2021-03-13 12:37:28 wth is "lbu"? 2021-03-13 12:37:49 local back-up 2021-03-13 12:38:05 it's a tool that is used to persist changes on a run-from-ram system 2021-03-13 12:39:11 interesting. 2021-03-13 12:39:54 Sounds like something I could throw onto my ESP partition as a rescue on my desktop. :) 2021-03-13 12:43:32 alpine is very different from my usual Archlinux, but it's growing on me.  Also "musl". 2021-03-13 12:44:34 minimalism is the best 2021-03-13 12:45:08 if you don't need it, don't install it 2021-03-13 12:45:23 it's like when you travel with only a backpack... do not bring shit you don't need 2021-03-13 12:45:33 Especially on an old "cloudbook" with soldered in RAM and soldered in emmc flash at 32gb. 2021-03-13 12:46:00 sounds like a good use case for a tiling window manager 2021-03-13 12:46:31 I have i3 on it.  Setting up xfce4 for "guests". 2021-03-13 12:54:59 I left a tub of yogurt out overnight after getting groceries when I was tired, and thought I lost the $5 2021-03-13 12:55:14 I put it in the fridge in the morning, left it a day, and it's fine 2021-03-13 12:55:27 why do I say this to you here? 2021-03-13 12:55:32 Technically, it's already "bad". :P 2021-03-13 12:55:36 because it was an assumption that was wrong! 2021-03-13 12:55:43 just like your assumption that guests can't use i3 ;) 2021-03-13 12:55:48 lol 2021-03-13 12:55:56 s/can't/won't 2021-03-13 13:16:01 thanks for the help peeps. I need sleeps. 2021-03-13 13:52:26 i'm sorry to bother you, just for curiosity, there is a particular reason why most of the packages builded with meson have `plain` buildtype instead of `release`? 2021-03-13 13:53:05 * i'm sorry to bother you, just for curiosity, is there a particular reason why most of the packages builded with meson have `plain` buildtype instead of `release` on the repository? 2021-03-13 13:53:17 Uniformbuffer[m]: because 'release' buildtypes ignore the flags we specify 2021-03-13 13:53:57 oh, ok, thanks for the answer 👍️ 2021-03-13 18:24:21 Install alsa, card works. Install pulseaudio, no cards found. Remove pulse, repeat.  What am I doing wrong? 2021-03-13 18:24:47 I have a feeling I have somehow created a loop, but no idea what to do to fix it. 2021-03-13 18:27:28 i.e. only dummy output 2021-03-13 18:27:58 u nothing, that's pulsecrap 2021-03-13 18:31:15 Works by default on every other distro. *shrugs* 2021-03-13 18:31:42 helby: +1 :) 2021-03-13 18:31:59 I'm not big on dbus ,systemd, etc, but the features of pulseaudio and it's "handiness" are nice. 2021-03-13 18:34:09 I'll gladly NOT use it if you can show me a way to get the same results without it. 2021-03-13 18:37:31 having pulseaudio in distro and call that distro secure doesn't make much sense 2021-03-13 18:39:00 I would say, to choose alpine and even check if pulse is in repo is weird 2021-03-13 18:41:58 I need lightweight (cloud book), but I still need features.  If I could attain them otherwise, I would.   Imagine if I told you that installing alpine and looking for a browser is weird.  You need what you need. 2021-03-13 18:42:52 I coudl get by if it was just me.  I loan this one and people would lose their mind. 2021-03-13 18:48:32 what is cloud book? 2021-03-13 18:52:54 Acer CloudBook 14.  Soldered in ram, cpu, and emmc flash of 32gb.  Slow as dirt. 2021-03-13 18:53:15 alpine is speedy on it 2021-03-13 18:53:30 probably something in line 'you pay book but you don't own your copy' 2021-03-13 18:54:40 It's more of a Chromebook, but hey look, stripped down Windows 10 with no room for any other apps. 2021-03-13 18:55:04 Made to sell MS office and one drive space. 2021-03-13 18:55:33 I used for some time pixelbook, that was relatively good machine 2021-03-13 18:55:34 I didn't buy this.  Someone gave it to me and I'm poor. So, yeah., 2021-03-13 18:55:36 how much RAM it have? 2021-03-13 18:55:42 This? 2gb 2021-03-13 18:55:52 not much 2021-03-13 18:55:52 btw. why u need pulsecrap? 2021-03-13 18:56:00 winners don't use pulseaudio 2021-03-13 18:56:24 pulse allows easy config new audio devices for the noobs that will use it from time to time. 2021-03-13 18:56:32 I have acer R13 chromebook, it is quite fast and good, about 5 years old though 2021-03-13 18:56:32 like 50% of my dev time is spent on audio stuff, I can tell you pulse sucks 2021-03-13 18:57:37 I "KNOW" pulse sucks, but given my ignorance of a suitable replacement and the needs of the less technically inclined that will use it...  I'm kind of stuck. 2021-03-13 18:57:51 i run pulseaudio as a client behind jack2 which has exclusive access to the audio interface 2021-03-13 18:57:54 zero issues 2021-03-13 18:58:00 u can't just plug in jack? 2021-03-13 18:58:01 Pulse sucks? 2021-03-13 18:58:19 what do you mean by new audio devices? 2021-03-13 18:58:22 alsa doesn't include a fancy GUI config for whatever random audio device the user will connect. 2021-03-13 18:58:40 usb and/or 3.5mm  headsets 2021-03-13 18:58:49 why do those need a config? 2021-03-13 18:58:52 auto-switching to the default. 2021-03-13 18:59:06 no idea what to config on sound from chromebook 2021-03-13 18:59:40 helby: I have 3 different chromebooks, alsa works quite fine on them 2021-03-13 19:00:08 I need the user to be able to connect whatever headset (or none) to the machine and have it JustWork(tm) as the default audio device.  I have no idea how to accomplish that other than pulse. 2021-03-13 19:00:25 oh by bluetooth? 2021-03-13 19:00:26 that just works for me on every laptop I own 2021-03-13 19:00:27 heh 2021-03-13 19:00:39 bluetooth could be an option too. 2021-03-13 19:00:49 not sure what u mean 2021-03-13 19:00:59 u have all devices connection to chromebook? 2021-03-13 19:01:02 piousminion: all that can be done with alsa 2021-03-13 19:01:06 What if the user has headset in, but wants audio to come from speaker instead?  Alsa gui for that? :P 2021-03-13 19:01:07 bc. otherwise when u plug them they are rady 2021-03-13 19:01:20 mps: Do tell. 2021-03-13 19:01:27 that sounds legit ;) 2021-03-13 19:01:50 there are some guis for alsa but I forgot names because I don't use them 2021-03-13 19:01:55 tell him instead of fuckering with gui, that he should take off headset 2021-03-13 19:02:22 It's not one person and the less technically inclined do not listen to reason 2021-03-13 19:02:48 I'm talking about people who can't figure out how to click the gmail app on their phone. 2021-03-13 19:03:10 0 concept of hot to accomplish basic....   human thought. lol 2021-03-13 19:03:14 but they wanna use your chromebook with 3 devices and changing them every hour, understand 2021-03-13 19:03:30 You understand my pain. 2021-03-13 19:03:46 liar! tell the truh, it's only you :) 2021-03-13 19:03:51 lmao 2021-03-13 19:04:28 It's not.  I don't ever use this thing.  I'm on my desktop with a 5930k #HumbleBrag 2021-03-13 19:05:04 did u try 'alsa-tools-gui' ? 2021-03-13 19:05:06 not sure what's that 2021-03-13 19:05:13 I will admit that I have pulse on my archlinux install....  but that's really about laziness. 2021-03-13 19:05:19 not yet 2021-03-13 19:09:47 How do you list files in a package again? 2021-03-13 19:10:20 also what de/vm u use? 2021-03-13 19:10:25 apk info -L pkgname 2021-03-13 19:10:39 The DE for the other users is xfce4 2021-03-13 19:10:51 there something 'pnmixer' 2021-03-13 19:11:12 at least they write it works with alsa, some kind of applet or how they call that crap 2021-03-13 19:12:08 something what u can probably add to your useless panel in your desktop ;) 2021-03-13 19:12:40 lol 2021-03-13 19:13:02 I'm an i3 user myself, but these people would lose it. 2021-03-13 19:13:49 Why is the output of apk-file so jacked up all over the place? 2021-03-13 19:13:50 try spectrwm sometime, it's better than i3 2021-03-13 19:13:58 ACTION googles 2021-03-13 19:14:02 especially for multiple monitor use cases 2021-03-13 19:19:58 pnmixer is just applet 2021-03-13 19:20:35 What do you guys recommend for switching wifi APs? 2021-03-13 19:20:42 setup-interfaces?  lol 2021-03-13 19:21:20 piousminion: iwd+iwgtk 2021-03-13 19:21:34 My card won't work with iwd. :/ 2021-03-13 19:21:53 broadcom or what? 2021-03-13 19:21:58 intel 2021-03-13 19:22:29 kmod "iwlwifi" 2021-03-13 19:23:03 it is strange, upstream iwd devs works for intel 2021-03-13 19:23:18 network and wifi is interesting in linux 2021-03-13 19:23:27 300 utilities for everything 2021-03-13 19:23:39 slowly there will ifraining 2021-03-13 19:23:43 I know.  this particular card/chip doesn't work with it.  idk why. 2021-03-13 19:23:47 piousminion: did you tried to ask on #iwd channel 2021-03-13 19:24:04 I have not, but I do recall reading specifically that it was unsupported. 2021-03-13 19:24:58 hm, I don't have it to test and no one reported bug, afaik 2021-03-13 19:25:13 I think the issue was that it only supported WEXT, which iwd does not. 2021-03-13 19:25:26 ah, could be 2021-03-13 19:25:58 I know this isn't the authority, but a quick search gave me this chart.  https://wiki.archlinux.org/index.php/Network_configuration/Wireless#Utilities 2021-03-13 19:26:09 chart/table 2021-03-13 19:26:59 alternatives to iwgtk? 2021-03-13 19:27:13 kurva like it can't be in one utility :/ 2021-03-13 19:27:49 2020-11-16 04:13...... Net147| are there known issues using iwd ap mode with brcmfmac? it works fine with iwlwifi and rt2800usb but with brcmfmac I am unable to connect 2021-03-13 19:28:17 from #iwd log 2021-03-13 19:28:44 No idea.  I know when I tried it simply refused to function and then I read some and switched to wireless-tools 2021-03-13 19:28:56 iwd wouldn't even see a wlan0 2021-03-13 19:31:05 well, can't help, I don't have this kind of cards 2021-03-13 19:31:25 I appreciate the feedback anyhow. :) 2021-03-13 19:32:31 but my advice is to join #iwd and ask there, maybe someone can help 2021-03-13 19:33:29 I shall as time allows.  I've been awake for 24 hours now. prescription stimulants and steroids will do that I guess. 2021-03-13 19:35:28 listen some music using alsa, that will help ;) 2021-03-13 19:36:00 u want too much from your little chromebook, internet connection and sound at the sime time, come on 2021-03-13 19:37:18 lol 2021-03-13 19:37:54 I could watch youtube videos via tty with mpv's asciiart output.  :P 2021-03-13 19:38:02 libcaca 2021-03-14 13:50:01 Hi. I've setting up Alpine for a very first time on very old Dell Latitude e6400. Why I can use xorg all right as root, but I can't as a normal user? Should I add myself to any special group? Quick check in /etc/group doesn't show anything related. 2021-03-14 13:50:35 (I'm kinda experienced Linux user, just didn't run any desktops (so: also xorg) for last ~10 years). 2021-03-14 13:51:03 ftpd: adduser username video audio input 2021-03-14 13:52:01 Thanks, now it works. 2021-03-14 13:52:16 https://arvanta.net/alpine/install-x/ 2021-03-14 13:52:33 this is for qemu but could help 2021-03-14 13:57:14 So, the other issue (I hope the last one, as everything other seems to work): this laptop has broadcom wi-fi card, so I've followed https://wiki.alpinelinux.org/wiki/Connecting_to_a_wireless_access_point with installing b43 module from aports. Everything went fine, but there is still no wlan0 device available. 2021-03-14 13:57:14 https://wiki.alpinelinux.org/wiki/Connecting_to_a_wireless_access_point#Broadcom_Wi-Fi_Chipset_Users - I followed it and build my pkg with firmware. 2021-03-14 13:57:14 you probably need firmware for card 2021-03-14 13:57:14 (I _never_ tried wifi on Linux - I've switched to Mac in 2008 as my desktop, using Linux just as servers). 2021-03-14 13:57:14 but afaik, firmware is non free so not distributed with alpine 2021-03-14 13:57:15 firmware is built, module loaded. 2021-03-14 13:57:26 I have b43 on this machine and it works 2021-03-14 13:57:55 Network controller: Broadcom Inc. and subsidiaries BCM4331 802.11a/b/g/n (rev 02) 2021-03-14 13:58:07 Hmm. Maybe it's not broadcom... 2021-03-14 13:58:47 lspci will tell 2021-03-14 13:58:49 this whole article needs revamp, it mixes wireless-tools and wpa_supplicant, then claims that only b43 supports broadcom 2021-03-14 13:59:14 wiki should be deleted 2021-03-14 13:59:39 Ok, lspci shows it's the same card. 2021-03-14 13:59:51 same with? 2021-03-14 14:00:16 Network controller: Broadcom Inc. and subsidiaries BCM4331 802.11a/b/g/n (rev 01) 2021-03-14 14:00:24 Same as yours. 2021-03-14 14:00:27 aha 2021-03-14 14:00:46 Oh, it's actually BCM4322. 2021-03-14 14:02:56 I have b43 firmware apk on this box, if you trust me and want I can upload it to dev.alpinelinux.org 2021-03-14 14:03:45 though not sure will it work with your card 2021-03-14 14:04:29 hmm, actually I have to ask first is that allowed 2021-03-14 14:04:35 ikke: ^ 2021-03-14 14:04:46 isn't the whole point of non-free that it is not allowed 2021-03-14 14:05:13 put non free package on alpinelinux.org servers sounds suspicious to me 2021-03-14 14:05:44 ftpd: check dmesg 2021-03-14 14:05:45 Hello71: yes, I come to this just after I wrote offer to upload it 2021-03-14 14:07:39 but debian have non-free firmware for download from some separate repos 2021-03-14 14:07:41 https://files.insomniac.pl/IMG_0280.jpeg - yeah, dmesg shows something. 2021-03-14 14:07:50 sure would be nice if everyone was honest and all software/firmware was MIT licensed 2021-03-14 14:08:30 I was hoping manual build of b43-firmware will do it. 2021-03-14 14:08:55 mps: Actually this is worthless laptop just for me to play, in a separated network, so I will trust your firmware even when you upload it to evilcorp.com ;-) 2021-03-14 14:09:34 ftpd: first time I installed alpine on macbook pro 2010 I downloaded firmware from debian repo and unpacked on alpine, and put in proper dirs 2021-03-14 14:09:39 and it worked 2021-03-14 14:10:59 ftpd: I can paste my APKBUILD.b43-firmware file which I used to build on alpine 2021-03-14 14:11:22 mps: Please do. I took mine from 'aports', whatever it is. 2021-03-14 14:11:24 this is not forbidden I think, yet hmm 2021-03-14 14:11:40 ftpd: it says you are missing firmware 2021-03-14 14:11:51 https://tpaste.us/PKz7 2021-03-14 14:11:58 ftpd: ^ 2021-03-14 14:12:17 also if you have network access it is preferable to use ix.io or similar service 2021-03-14 14:13:14 Hello71: I have my own 'uploader' script to my home NAS serving as https for tmp files - but sure, I can upload to something well-known. 2021-03-14 14:13:40 well the difference is between text and picture 2021-03-14 14:13:52 picture is hard to read especially for blind users 2021-03-14 14:21:06 mps: That build helped, thank you! 2021-03-14 14:21:18 Ok, so I have X working and wifi ready to set up. 2021-03-14 14:54:37 anyone thought of how to replace debian or centos with alpinelinux inside a LXC without access to the host? 2021-03-14 14:55:36 (or anyone know of a cloud provider that supplies cheap alpine linux LXC containers? :) 2021-03-14 17:52:00 ScrumpyJack: there is an article in the wiki on how to do that. 2021-03-14 18:13:23 How are you supposed to set the PATH variable to something sensible? I'd like to have the same $PATH as the root user? 2021-03-14 18:17:17 Copy root's default PATH value? 2021-03-14 18:17:34 johnaj: to where? 2021-03-14 18:17:47 johnaj: .profile isn't sourced? 2021-03-14 18:18:06 ah, .profile is used only for login shells 2021-03-14 18:19:19 That's the only thing ash supports 2021-03-14 18:20:15 apparently busybox sh also reads whatever file specified by the ENV variable (?) 2021-03-14 18:20:58 That might work if you want to launch sh as a non-login shell. 2021-03-14 18:21:08 ENV=~/.profile sh 2021-03-14 18:22:01 so where do I put it? 2021-03-14 18:22:28 Well, you put it in ~/.profile, if you want it to be run when you log in. 2021-03-14 18:22:42 Another caveat is that you must export the PATH variable. 2021-03-14 18:22:46 export PATH=whatever... 2021-03-14 18:22:52 otherwise it won't work 2021-03-14 18:24:11 I've done that, but it clearly doesn't read that file 2021-03-14 18:24:49 Are you sure you're running sh as a login shell? 2021-03-14 18:25:07 If you run sh manually, i.e. already in a shell session, then it doesn't run as a login shell and doesn't read .profile 2021-03-14 18:25:31 thus you can't test .profile updates by just running sh 2021-03-14 18:26:08 I execute `su myuser` 2021-03-14 18:26:31 and even when I run ash again, it doesn't set the PATH 2021-03-14 18:26:33 lord4163: You need to run `su - myuser`, then you get a login shell 2021-03-14 18:27:43 ikke: ah thanks! 2021-03-14 18:57:31 hi everyone, I'm trying to use net-snmp's snmpd from the package manager. It quits every few minutes. If I run "snmpd -V -f -Lo" to try and find out what is going wrong, it segfaults immediately. Any ideas what could be wrong? 2021-03-14 19:03:03 Chewie9999: https://gitlab.alpinelinux.org/alpine/aports/-/issues/12309 2021-03-14 19:03:44 ah, nice one, thanks 2021-03-14 19:03:56 I did try to look into it 2021-03-14 19:04:42 it seems to run on my system for 15mins or so as a service or in the foreground, but as soon as i add "-Lo" it segfaults immediatly 2021-03-14 19:06:08 interesting, commenting out trapsink works with "-Lo" 2021-03-14 19:06:24 @whois ikke 2021-03-14 19:06:49 did u have the same problem too? 2021-03-14 19:07:02 I did not experience it myself (not using net-snmp on alpine) 2021-03-14 19:07:06 ah ok 2021-03-14 19:31:27 How do I generate locales? I can't find any package containing locale-gen. 2021-03-14 19:32:38 musl does not have locales 2021-03-14 19:33:42 https://wiki.musl-libc.org/functional-differences-from-glibc.html#Character-sets-and-locale 2021-03-14 19:34:14 "Up through version 1.1.10, musl provided a purely UTF-8 C locale". 2021-03-14 19:34:21 That's enough, thanks. 2021-03-14 19:37:10 I have my dwm+st configured with NF-patched font, but some chars aren't displayed correctly still. My first guess was non-utf locale, but now I know I have to search for other reason. 2021-03-14 20:11:12 heyhow, does anyone have an aarch64 rootfs at hand? 2021-03-14 20:11:24 I have 2021-03-14 20:11:35 I mean, I have a container running 2021-03-14 20:12:32 may i have it?^^ 2021-03-14 20:12:42 https://dl-cdn.alpinelinux.org/alpine/edge/releases/aarch64/alpine-minirootfs-20210212-aarch64.tar.gz 2021-03-14 20:12:48 do you mean something like that? 2021-03-14 20:12:57 I should give pipewire a shot on my pbp 2021-03-14 20:13:03 wonder if it'd make audio more stable 2021-03-14 20:13:06 that'd be nice 2021-03-14 20:13:57 ikke: exactly, thank you very much!:) 2021-03-14 22:45:38 hey everyone 2021-03-14 22:47:26 anyone managed to run mongodb in edge? I tried the latest official binaries (mongodb-linux-x86_64-v4.0-latest.tgz), but they won't work due to many missing libraries (libresolv.so.2, librt.so.1, libdl.so.2, libm.so.6, etc..). 2021-03-14 22:48:41 sam_son: looks like these binaries are built with gnu libc and not musl 2021-03-14 22:48:50 alpine use muls libc 2021-03-14 22:49:01 ah I see 2021-03-14 22:49:33 no chance of getting mongodb running under alpine then 2021-03-14 22:49:42 option is to rebuild it for yourself 2021-03-14 22:49:51 I see there is an open issue from 2018 (https://jira.mongodb.org/browse/SERVER-36790) 2021-03-14 22:50:50 it is updated few weeks ago in aportm non-free repo 2021-03-14 22:51:21 I mean, APKBUILD is updated 2021-03-14 22:53:27 I'm afraid rebuilding mongodb will unleash pandora's box... and lead to many other problems 2021-03-14 22:54:22 I tested build, it worked, don't know does mongodb works 2021-03-14 22:56:09 is there a guide that explains how to build apk's and what tools are needed? I'd like to give it a try 2021-03-14 23:00:12 wiki.alpinelinux.org 2021-03-14 23:01:41 that link should be prohibited 2021-03-14 23:02:39 https://wiki.alpinelinux.org/wiki/Creating_an_Alpine_package 2021-03-14 23:02:58 helby, that bad? :) 2021-03-14 23:03:32 sam_son: always when need something, just wasting time bc. trying something 'by wiki' 2021-03-14 23:04:34 I'll waste a little just, maybe I'll learn something new 2021-03-14 23:05:08 if things are too messed up, I'll just boot another distro for mongodb 2021-03-14 23:05:46 new? u will learn something very old dear ;) 2021-03-14 23:09:40 well, new to me 2021-03-14 23:10:14 I've been spoiled with rpm packages... haven't compiled a package since several decades... 2021-03-14 23:11:56 so now u decided to glibcrap to musl? good luck :/ 2021-03-14 23:13:04 and I just wanted to run mongodb... oh well 2021-03-14 23:26:18 followed the steps in the wiki 2021-03-14 23:26:29 got the non-free/mongodb things 2021-03-14 23:26:46 and I tried abuild -r 2021-03-14 23:27:00 while it seems to compile... I see errors like: 2021-03-14 23:27:02 internal compiler error: Bus error 2021-03-14 23:31:29 probably not enough memory, I'll up the VM memory to 4gb and retry 2021-03-15 00:42:48 NTS enabled in chrony, nice 2021-03-15 10:41:30 hello 2021-03-15 10:42:01 anyone knows if its possible to tell abuild to use more cpu cores? I've got 6 but it only uses 2 of them. 2021-03-15 10:42:27 Set JOBS to a higher value in /etc/abuild.conf 2021-03-15 10:42:55 aha! thank you 2021-03-15 10:43:45 or, export JOBS=6 2021-03-15 10:44:45 this compilation will never end, I've been stopping it and restarting it for 10 hours now.. fun times 2021-03-15 10:45:01 (compiling mongodb from non-free) 2021-03-15 10:46:11 it has been an interesting experience 2021-03-15 10:46:33 and no, you can't compile mongodb with 512 of ram :) 2021-03-15 10:46:51 and neither can you do it with a 1GB partition :) 2021-03-15 10:47:52 512MB? 2021-03-15 10:48:09 well, its a VM... that I cloned from another VM 2021-03-15 10:48:28 usually 512 is enough for my tests 2021-03-15 10:49:38 in most of them I use 256mb... its enough to boot a clean alpine and run a single application with a memory footprint of 100mb or so 2021-03-15 10:55:49 q: when running abuild with "abuild -r -k -K -c", does it start from where it last stopped? or does it always start fresh (even if I told it to preserve the cache) 2021-03-15 11:36:21 back 2021-03-15 11:37:36 weird compilation completed 2021-03-15 11:37:40 scons: done building targets. 2021-03-15 11:38:00 but fails with 2021-03-15 11:38:03 ERROR: Unable to read database state: No error information 2021-03-15 11:38:06 ERROR: Failed to open apk database: No error information 2021-03-15 11:38:15 >>> ERROR: mongodb*: create_apks failed 2021-03-15 11:38:18 >>> ERROR: mongodb: rootpkg failed 2021-03-15 11:38:22 hmm 2021-03-15 11:40:05 I made a serious effort to get a testing package (sssd) working and failed. I'm fairly sure it's a glibc issue. Does one report this sort of thing or does one assume the packagers and alpine devs already know about it? 2021-03-15 11:41:45 nfc: if there is not a gitlab issue about it, I would not assume we know about it 2021-03-15 11:41:53 yes, alpine is great for glibc things :/ 2021-03-15 11:42:03 good choice boys ;) 2021-03-15 11:44:19 so do I report the issue on the aports issue tracker? 2021-03-15 11:44:44 nfc: yes 2021-03-15 11:45:25 thanks. i'll do that. 2021-03-15 11:46:56 is there something special I should do for booting kernel directly using efistub? Somehow can't get it done. 2021-03-15 11:47:15 helby: no experience with that 2021-03-15 11:47:28 right now I have feeling stupid grub is bigger than whole my system :/ 2021-03-15 11:48:02 ikke: last try got me kernel panic that it can't mount or something like that /boot 2021-03-15 11:48:12 not boot, root I think ;0 2021-03-15 11:48:51 kernel modules missing perhaps? 2021-03-15 11:48:58 maybe some kernel parameters should be there or some shitty mkinitfs crap 2021-03-15 11:49:09 ikke: not sure, I have default in mkinitfs.conf 2021-03-15 11:49:16 if it requires something special 2021-03-15 11:50:04 something is broken with my apk database (unable to read database state), anyone knows where its located? 2021-03-15 11:50:48 sam_son: /lib/apk/db 2021-03-15 11:50:53 thank you 2021-03-15 11:53:37 the files installed/lock/triggers are there, but missing the scripts.tar, "apk fix" returns with error unable to read database state. weird... 2021-03-15 13:00:06 EF: for replacing a running LXC with alpine? 2021-03-15 15:19:28 Today is a happy day, I just wanted to say thanks to everyone for contributing to alpine - it's really a lot of fun using it 2021-03-15 15:24:49 heh, after todays $day_job I wanted to declare this day as 'drunk day' :) 2021-03-15 15:26:52 Uh, I know those days. What's your favorite poison? 2021-03-15 15:32:04 red wine and home made brandy 2021-03-15 15:33:06 but never too much, 'just enough' TM 2021-03-15 15:44:09 hello 2021-03-15 15:44:56 how to let apk only output version ? 2021-03-15 15:45:03 q: after abuild is finished, what file do I need to copy from my dev machine to my app machine, so as "apk add ./custom.apk" does not complain about an unknown/invalid signature? 2021-03-15 15:45:35 key under you $HOME/.abuild 2021-03-15 15:45:36 I want to avoid using "--allow-untrusted" 2021-03-15 15:45:44 copy to /etc/apk/keys 2021-03-15 15:46:03 ah I see there is an rsa pair (private/public keys) 2021-03-15 15:46:13 wener[m], thank you! 2021-03-15 15:51:44 wener[m]: apk version -q I believe 2021-03-15 15:52:15 apk version -q mus output is `musl` 2021-03-15 15:55:27 telmich: and your? (just curios) 2021-03-15 15:56:08 mps: I'm also into red wine and a variety of whiskeys, the latter helped a lot for studying 2021-03-15 15:58:13 cheers :) 2021-03-15 16:05:17 I drink just 10 cups of coffee every day 2021-03-15 16:11:02 uh, my head would explode if I do so much 2021-03-15 16:11:56 don't worry, nobody is perfect :/ 2021-03-15 16:13:02 if I drink or even smell one glass of whiskey I will probably vomit 2021-03-15 16:20:10 wener[m]: I don't see an option to only print the version 2021-03-15 16:21:20 @ikke emmm, seems like a missing function of apk, want to get current version, used to tag other things when ci. 2021-03-15 16:37:52 I just had to `setcap cap_net_raw+ep $(command -v blackbox_exporter)`, because the blackbox exporter is otherwise not able to use icmp (see https://github.com/prometheus/blackbox_exporter for details). Is anything speaking against adding this into the post-install of the blackbox_exporter? 2021-03-15 16:40:47 telmich: not sure how relevant that is, but it would prevent those binaries from running in docker 2021-03-15 16:42:58 telmich: oh, just tested it, this capability is apparently not an issue 2021-03-15 16:47:20 ikke: that's good to hear. Then I think it would be a sane thing to give that permission by default, because icmp is rather the standard/default case for using the blackbox_exporter 2021-03-15 17:03:09 nfc: fwiw sssd almost certainly needs musl-nscd to be useful 2021-03-15 17:17:57 hello! i need some help installing apk in a docker container that does not have apk install. i could get to download apk.static but the database is missing 2021-03-15 17:19:12 magnusmaster: what do you expect to happen? Trying to manage packages on an existing system would most likely result in file conflicts 2021-03-15 17:19:49 i just want to install missing commands 2021-03-15 17:20:42 Does APK have Xorg available so I can run fluxbox or ratpoison? 2021-03-15 17:23:01 dukester: https://pkgs.alpinelinux.org/packages?name=xorg-server&branch=edge 2021-03-15 17:23:18 i am using the .net core 3.1 alpine image. how can i install packages to it? 2021-03-15 17:23:33 magnusmaster: you cannot use apk to just install single packages, it will install all dependencies as well 2021-03-15 17:23:36 ikke: thx 2021-03-15 17:24:12 magnusmaster: you have to ask the author of that image 2021-03-15 18:56:42 Does Alpine Linux have a relocatable install, a la nix, or gentoo? 2021-03-15 19:04:48 Sargun: I don't think so 2021-03-15 19:19:12 I'm trying to do a "sys" mode install on a Raspberry Pi and setup-disk says "ext4 is not supported. Only supported are: vfat" Am I doing something wrong here? 2021-03-15 19:19:37 Running this thing with root on tmpfs is useless on the 256mb model i have. 2021-03-15 19:21:23 AbortRetryFail: please be more descriptive, which RPi flavor 2021-03-15 19:21:43 the original rpi 1 model b 2021-03-15 19:22:24 hmm, I have only RPi zero 2021-03-15 19:22:41 it works fine in diskless mode, but there's not much RAM left to do anything, so i'd rather have rootfs on the sdcard 2021-03-15 19:24:13 yes, I know, I installed it 'manually' in sys mode 2021-03-15 19:25:01 I was working from this wiki page https://wiki.alpinelinux.org/wiki/Raspberry_Pi#Traditional_disk-based_.28sys.29_installation 2021-03-15 19:25:30 but setup-disk complains about the partition I want to use (/dev/mmcblk0p2) which has a EXT4 filesystem 2021-03-15 19:27:21 you are trying to install on same mmc from which you booted it? 2021-03-15 19:27:40 yeah, to another partition 2021-03-15 19:28:10 not sure but I think this doesn't work 2021-03-15 19:29:13 anyway, try 'modprobe ext4' 2021-03-15 19:29:59 says that's not found in modules.dep. I can mount it manually though and the filesystem works though. 2021-03-15 19:32:34 have no idea why, I always install 'manually' (scripts) on small devices 2021-03-15 19:33:49 do you have a link to some documentation about how to do that? I assume manually you mean without using setup-disk or setup-alpine. 2021-03-15 19:34:57 yes for manually, but not for RPi scripts 2021-03-15 19:35:20 though I have some notes about installing alpine on some arm boxes 2021-03-15 19:35:33 https://arvanta.net/alpine/ 2021-03-15 19:35:46 maybe you can find some ideas there 2021-03-15 19:36:19 Thanks 2021-03-15 19:37:53 this one https://arvanta.net/alpine/armhf-setup/ but should be adapted for RPi, i.e. kernel and bootloader configs 2021-03-15 19:38:11 yep, already on it. Seems simple enough. 2021-03-15 19:39:15 yes, but imprecise and buggy, don't run it blindly :) 2021-03-15 19:39:35 especially not the parted ... comands 2021-03-15 19:40:11 Hello folks. I have a question. I need to setup a specific PATH for a user, for (Bash) remote (SSH) command execution. Bash documentation states that .bashrc for the user should be sourced in this case, but it's simply not happening. I looked at my Gentoo, and it sources an envfile via PAM. I tried to replicate this, but it doesn't work. Anyone bumped into an issue like this? 2021-03-15 19:40:52 Are you using Bash or busybox sh, the default shell? 2021-03-15 19:41:08 busybox sh supports only .profile (and that's only for login shells) 2021-03-15 19:41:16 johnaj - it's 100% Bash. 2021-03-15 19:41:58 user shell is Bash, and when I 'ssh user@boxname "ps auxwww | grep bas[h]"', bash -c shows up 2021-03-15 19:42:14 bash -c is not a login shell 2021-03-15 19:42:38 ikke - yeah, of course not. But the docs say that when a command is executed via SSH, it would read .bashrc 2021-03-15 19:43:40 Does it work if you launch the interactive shell via ssh? 2021-03-15 19:43:54 johnaj, yeah, it does 2021-03-15 19:44:10 aha, so it's just when you run a single command. hm 2021-03-15 19:44:24 man bash says it's only executed when an interactive shell is run 2021-03-15 19:44:54 "When an interactive shell that is not a login shell is started, bash reads and executes commands from ~/.bashrc, if that file exists." 2021-03-15 19:44:55 ikke - it says separately about a remote shell 2021-03-15 19:45:30 I think it still has to be interactive 2021-03-15 19:45:32 ikke - Bash attempts to determine when it is being run with its standard input connected to a a network connection, as if by the remote shell daemon, usually rshd, or the secure shell daemon sshd. If bash determines it is being run in this fashion, it reads and executes commands from ~/.bashrc, if that file exists and is readable. It will not do this if invoked as sh. The --norc option may be 2021-03-15 19:45:38 used to inhibit this behavior, and the --rcfile option may be used to force another file to be read, but rshd does not generally invoke the shell with those options or allow them to be specified. 2021-03-15 19:45:53 anyway, this is something bash specific 2021-03-15 19:45:57 nothing to do with alpine itself 2021-03-15 19:46:24 bash -c is not an interactive shell, so it apparently does not source .bashrc 2021-03-15 19:46:28 ikke - it doesn't appear to be the case on either Alpine, or Voidlinux, but on Voidlinux I managed to get the env working via PAM, like my Gentoo 2021-03-15 19:46:30 no matter if it's remote or local 2021-03-15 19:46:53 That's something differen 2021-03-15 19:47:01 The base installation does not use PAM 2021-03-15 19:47:23 There is https://pkgs.alpinelinux.org/packages?name=openssh-server-pam&branch=edge 2021-03-15 19:48:02 Sounds like that bash feature doesn't really work... maybe a bug could be reported to the bash project? 2021-03-15 19:48:06 ikke - yeah, I installed that package 2021-03-15 19:48:17 johnaj: I think it's as intended 2021-03-15 19:48:20 johnaj - yeah, I guess I could 2021-03-15 19:48:33 russoisraeli: you are invoking bash -c, right? 2021-03-15 19:48:35 not according to what I pasted directly from the manual 2021-03-15 19:49:18 no, what I was saying is when I run "ssh user@alpinebox 'ps auxwww | grep bas[h]'", 2021-03-15 19:49:38 it reports that "bash -c 'ps auxww.." is running 2021-03-15 19:49:44 meaning Bash is used 2021-03-15 19:49:46 right, so bash -c, not an interactive hsell 2021-03-15 19:49:49 that's what I was saying 2021-03-15 19:49:56 try to pass -t 2021-03-15 19:50:01 that's what sshd must be forking 2021-03-15 19:50:09 yes, because you pass a command 2021-03-15 19:50:11 directly 2021-03-15 19:50:22 So it does not start an interactive shell 2021-03-15 19:50:59 ikke - it's not, but look at the snippet from Bash manpage that I pasted above 2021-03-15 19:51:04 And because it's not an interactive shell, .bashrc is not sourced. No matter if it's remote or local 2021-03-15 19:51:14 russoisraeli: You ignore the interactive part that was mentioned earlier 2021-03-15 19:51:51 what decides which shell is used when passing a command to ssh? shouldn't that be the user's login shell? 2021-03-15 19:51:59 johnaj: I think so, yes 2021-03-15 19:52:12 then you could manually change the relevant user's login shell 2021-03-15 19:52:20 ikke - I triple-looked over the stanza, and unless I am blind, it says nothing about interactive shells 2021-03-15 19:52:21 but running ssh server 'command' is different from runinng ssh server 2021-03-15 19:52:35 russoisraeli: It's mentioned earlier 2021-03-15 19:52:35 hey, what changed with compilers or permissions between 3.13? 2021-03-15 19:52:44 I've dockerfile working on latest and not working on edge 2021-03-15 19:52:49 JuniorJPDJ: Let me guess 2021-03-15 19:52:58 docker 2021-03-15 19:53:08 ACTION uploaded an image: (32KiB) < https://matrix.org/_matrix/media/r0/download/juniorjpdj.pl/VfvuuhXXSWLBPRDfTnxcrTqU/Screenshot_20210315_205301.png > 2021-03-15 19:53:13 and this is error 2021-03-15 19:53:17 and I'm like 2021-03-15 19:53:19 "hell what?" 2021-03-15 19:53:34 this is on edge, on latest it works >__< 2021-03-15 19:53:34 https://wiki.alpinelinux.org/wiki/Draft_Release_Notes_for_Alpine_3.14.0#faccessat2 2021-03-15 19:53:35 ikke: :< 2021-03-15 19:54:06 ikke - well, it might be misleading... anyhow, any idea how to get anything sourced 2021-03-15 19:54:29 I just need to set the damn PATH :) 2021-03-15 19:54:35 russoisraeli - you could perhaps set the user's login shell to a shell script that sets whatever environment variables you need to set and runs bash 2021-03-15 19:54:49 johnaj - smart, that's a possibility 2021-03-15 19:55:13 I think git-shell is something similar 2021-03-15 19:55:25 I use it for the git user on my server 2021-03-15 19:55:36 'bash -lc "echo test"' 2021-03-15 19:55:47 that works as well 2021-03-15 19:56:02 mps: Well, I figured out why it wasn't working. Somebody broke it on purpose. https://gitlab.alpinelinux.org/alpine/alpine-conf/-/blob/master/setup-disk.in#L218 2021-03-15 19:56:31 JuniorJPDJ: this is a bug in docker 2021-03-15 19:57:13 johnaj: what is the host OS? 2021-03-15 19:57:16 JuniorJPDJ: * 2021-03-15 19:57:32 alpine 2021-03-15 19:57:39 what I don't understand is why PAM doesn't work when I install the openssh-server-pam package. I enabled UsePAM in the sshd_config file, restarted 2021-03-15 19:58:02 might need to run it in debug more 2021-03-15 19:58:03 mode 2021-03-15 19:58:08 sshd 2021-03-15 20:00:25 AbortRetryFail: interesting, good catch 2021-03-15 20:00:43 Hello71: ^ 2021-03-15 20:00:54 ikke: host os is ubuntu 2021-03-15 20:01:20 ohm, no sorry Hello71, it is not your 2021-03-15 20:01:23 JuniorJPDJ: maybe consider opening a bug there 2021-03-15 20:01:25 thanks for link, I'll check it out 2021-03-15 20:04:08 btw. there is also 'PermitUserEnvironment' in sshd config 2021-03-15 20:18:23 ikke: well, host update fixed issue, thanks 2021-03-16 07:52:26 hey. i have some aliases in my .profile but they wont get loaded. is there any other syntax than just normal "alias a=b" ? 2021-03-16 07:52:59 Do you test it with a login shell? 2021-03-16 07:54:11 oh why does it work in login shells? i read online .profile is loaded when an interactive shell is startet 2021-03-16 07:54:29 No, only for login shells 2021-03-16 07:55:17 ahh, now that makes sense. thank you! 2021-03-16 08:21:06 Hi, is there a list of all files inside all alpine apk packages? 2021-03-16 08:21:32 I need to find where a (not installed) file is 2021-03-16 08:24:06 (e.g., the busybox manual page "busybox.1" ) 2021-03-16 08:24:31 apk-file is for filesearch 2021-03-16 08:30:40 g0r3: many thanks! it works (though the output is a bit strange) 2021-03-16 08:57:35 mnhrdt: yeah. on the package search webpage you can search file too 2021-03-16 11:20:13 https://lore.kernel.org/lkml/20210315095710.7140-1-henning.schild@siemens.com/ 'add device drivers for Siemens Industrial PCs' 2021-03-16 11:20:37 these could be interesting use cases for alpine 2021-03-16 11:58:33 where can one find a list of the "Desktop Linux current best practises", eg, for input, just install libinput, supercedes all the gymnastics of yesteryears. For AMD gpus, if you plan on using xorg, install pkg1,pkg2,pkg3 to get acceleration. vaapi is preferred to vdpau.. etc 2021-03-16 11:58:47 is there a place/reference where one can find such information? 2021-03-16 12:01:15 abenz: yes, in your brain after spending some nights with testing all nonsense around 2021-03-16 12:03:24 abenz: but arch wiki gives a lot of good info in this manner 2021-03-16 12:12:38 look what is (and how it is) in mainstream desktops and do opposite 2021-03-16 12:18:02 that helps too ;) 2021-03-16 12:18:19 u can apply this for entire life, not just soft :) 2021-03-16 12:30:14 yes, also I do that mostly 2021-03-16 12:47:24 so now u can delete alpine wiki, bc. nobody still did it :) 2021-03-16 12:52:02 I forgot my credentials there ;) 2021-03-16 12:56:02 it does not matter, get access, remove everything, all backups, that nobody can put that shit back 2021-03-16 12:57:34 instead 'wiki was here' u can make just 'mps was here' 2021-03-16 13:00:19 hehe, or 'helby is here' 2021-03-16 13:01:14 but no, wiki doesn't matter to, I don't look at it. simple 2021-03-16 13:01:39 to me* 2021-03-16 13:02:00 (don't lay on couch and type on IRC) 2021-03-16 13:03:18 pulse, pipewire, gnome, kde ... are better candidates for removal ;) 2021-03-16 13:03:38 well there is some shit in wiki about desktop 2021-03-16 13:03:57 between first line is to install bash and all other gnu utils with some extra crap 2021-03-16 13:04:00 so 2021-03-16 13:12:48 anyway if you mentioned crapware, I have some 'issue' with firefox 2021-03-16 13:13:35 it somehow looks that if using sway/wayland instead of xorg crap I got more cpu/gpu work, at least got 10'C higher temperature of cpu 2021-03-16 13:14:05 not sure what exactly does that shit 2021-03-16 13:15:36 as I set up acceleration for firefox 2021-03-16 13:16:09 but machine is thinkpad t420 so gpu is powerfull like my coding skills 2021-03-16 13:17:10 thanks all for reading me, I love this deadsleeping channel ;) 2021-03-16 13:17:25 xDDDDDD 2021-03-16 13:18:20 in openbsd they usually send you to manuals 2021-03-16 13:18:37 here it's more focused on freedom, so nobody is sending me anywhere :/ 2021-03-16 17:10:50 Morning, is there a way to configure musl to handle dns queries with the nocookie option? 2021-03-16 17:13:17 trebuh: perhaps ask in #musl 2021-03-16 17:15:34 I just did :-) but that would mean for me to recompile musl probably :-) 2021-03-16 17:52:04 trebuh: What problem are you trying to solve with nocookie? 2021-03-16 18:05:03 Hum, the full extent is. I use Alpine as a Docker image on Kubernetes. I try to resolve some names. When using `dig my-private-server` it does not work from an Alpine image but it does from a Debian image. Same version of dig on each side, same network config. So I got into what cool be different. When I try dig +nocookie on the Alpine it does work. I wanted to set that behaviour by default. 2021-03-16 18:05:28 s/cool/could/g 2021-03-16 18:07:08 Interesting. It might be a good idea to take a look at the DNS packets and see what's different. 2021-03-16 18:07:31 You could try running something like unbound and see if your containers can do lookups through that. 2021-03-16 18:08:55 I was going into looking into the packets via tcpdump indeed 2021-03-16 18:09:16 There are several differences between musl and glibc DNS implementations. For example, musl doesn't implement DNS over TCP. However, dig doesn't use the libc implementation of DNS. As a part of the BIND project, it has its own implementation. 2021-03-16 18:11:44 I had a similar problem with machines behind a seriously braindead recursor (windows 2008) where the work-around was to disable edns 2021-03-16 18:24:12 trebuh: Could the difference be caused by different compile options? There are some differences: https://git.alpinelinux.org/aports/tree/main/bind/APKBUILD#n118 https://salsa.debian.org/dns-team/bind9/-/blob/debian/main/debian/rules#L49-73 2021-03-16 18:25:07 possibly, I'll have a look. My goal is still to have it work on Alpine since I want to keep it as my base image 2021-03-16 18:30:56 Is dig the only thing not working or are there other programs having issues? 2021-03-16 18:48:37 nope dig is not the only one having issue. I found the issue on a java program deployed on an Alpine-based image and also while trying to deploy Bank-Vaults 2021-03-16 19:02:20 trebuh: If you want to see the same results that you program will see, you can use `getent ahosts my-private-server`, which uses the standard libc interface. It is intersting that both dig and your program don't work. It's worth looking into whether they have the same problem or different problems. 2021-03-16 19:23:47 Alpine has a minikube package in testing, but it's version 0.35.0 from 2019 while current is 1.18.1; what's the best way to request that it be bumped? 2021-03-16 19:25:10 yjftsjthsd: issue on https://gitlab.alpinelinux.org/alpine/aports, or even a merge request 2021-03-16 19:27:28 xordspar0: oh cool I'll try, thanks xordspar0 2021-03-16 19:36:09 thanks ikke 2021-03-16 22:47:33 is there RTKit on alpine? I can't find it 2021-03-16 23:34:51 does some noob here use 'foot' terminal? 2021-03-16 23:35:41 url mode somehow does not work for me 2021-03-16 23:36:45 ... just saying, don't expect some help :/ 2021-03-17 01:06:57 go it, it's not yet implemented, thanks for your input ;) 2021-03-17 04:14:51 hello 2021-03-17 04:22:28 hi pilanto 2021-03-17 12:02:07 .o/ 2021-03-17 13:19:15 still having issue with docker network, always need to restart dockerd to fix it 2021-03-17 13:19:34 except that, containers running well on it 2021-03-17 14:04:57 metbsd: Sounds like a docker issue 2021-03-17 14:05:09 Or are you using alpine to run docker? 2021-03-17 14:17:20 alpine host 2021-03-17 14:17:40 i know it's not officially supported 2021-03-17 14:17:51 but i like alpine lightweight as a docker host 2021-03-17 14:21:24 ah 2021-03-17 14:21:35 I just use AWS ECS Fargate 2021-03-17 14:23:04 is it free? 2021-03-17 14:50:35 metbsd: is ip6tables installed? 2021-03-17 14:50:43 What files should I pass through into an alpine container to have it use the same timezone as the host it's running on? 2021-03-17 14:52:04 Looks like /etc/timezone and /etc/localtime? 2021-03-17 14:54:08 restart dockerd fix the problem. iptables? 2021-03-17 14:58:06 metbsd: which version ? 2021-03-17 14:58:25 I run docker on my alpine machine and that seems to work 2021-03-17 14:59:03 metb 2021-03-17 14:59:10 latest 2021-03-17 14:59:20 no iptables 2021-03-17 14:59:25 metbsd: apk version ip6tables 2021-03-17 14:59:27 it's default installation without anything 2021-03-17 15:01:35 metbsd: docker depends on iptables 2021-03-17 15:01:44 https://pasteboard.co/JT2TXet.png 2021-03-17 15:03:17 metbsd▸ compare the output of 'iptables-save' before and after restarting dockerd to fix the problem. 2021-03-17 15:03:57 if you have any other component that touches iptables in any way on a docker system, you have the possibility of a conflict. Are you using a firewall manager alongside docker? 2021-03-17 15:04:12 some work fine alongside docker, others need some extra config. 2021-03-17 15:04:25 I also run docker on alpine and don't have issues. 2021-03-17 15:05:37 The dependency on ip6tables was missing, which caused some issues 2021-03-17 15:06:03 So that's why I asked 2021-03-17 15:14:00 metbsd: what does the command I gave earlier return? 2021-03-17 17:10:24 iirc, someone asked in last week/s to enable some options/drivers in linux-edge 2021-03-17 17:10:43 mps: context? 2021-03-17 17:11:03 weiti │ mps: would it be ok to send an merge-request for activate Segment Routing on the linux-edge flavours (especially x86_64 flavours)? 2021-03-17 17:11:11 I will upgrade it this evening so please write if something is needed 2021-03-17 17:11:20 ikke: thanks 2021-03-17 17:11:33 i searched but didn't found it 2021-03-17 17:12:46 and I saw somewhere something about realtek drivers, but I think it is one old bug/issue report 2021-03-17 17:14:15 bl4ckb0ne: anything for aarch64? 2021-03-17 17:22:25 programmerq, i don't have a firewall manager. docker is managing it 2021-03-17 17:27:50 metbsd: Can you please check if ip6tables is installed on your system? 2021-03-17 17:28:00 apk version ip6tables 2021-03-17 17:28:12 ikke, says not available 2021-03-17 17:28:28 What version of alpinelinux are you running? 2021-03-17 17:28:47 Installed: Available: 2021-03-17 17:28:50 that's all output 2021-03-17 17:29:01 hmm i don't even know 2021-03-17 17:29:08 cat /etc/alpine-release 2021-03-17 17:29:30 3.13.2 2021-03-17 17:29:49 ok. apk add ip6tables 2021-03-17 17:29:57 and see if you are still having this issue 2021-03-17 17:30:18 If that fixes it, I will make sure the dependency gets added on 3.13 as well 2021-03-17 17:30:23 eh, #7455 is in staging kernel area, so I think we should not enable it 2021-03-17 17:32:09 ok i'll do that 2021-03-17 17:32:35 need to restart docker? 2021-03-17 17:32:40 after adding 2021-03-17 17:32:57 now ip6tables is empty 2021-03-17 17:33:44 yes, that does not matter, but docker wants the ip6tables command to be available 2021-03-17 17:33:59 Now, of possible, try to reproduce the issue 2021-03-17 17:34:33 i don't know how to reproduce. sometimes when i readd a container it happens 2021-03-17 17:34:47 i'll report here if it happen again 2021-03-17 17:37:15 restart docker and ip6tables still empty 2021-03-17 17:37:31 metbsd: don't worry about the contets 2021-03-17 17:37:34 contents 2021-03-17 17:37:52 If you don't enable ipv6 networks in docker, it won't use it 2021-03-17 17:38:08 but it wants the ip6tables command to be present 2021-03-17 17:38:32 ok didn't know that 2021-03-17 17:38:53 [diablo] mentioned that earlier to you 2021-03-17 17:39:56 does docker require glibc or musi is ok 2021-03-17 17:40:18 musl 2021-03-17 17:40:47 docker runs fine on musl 2021-03-17 17:52:50 docker's somewhat libc agnostic 2021-03-17 17:52:55 composer, less so 2021-03-17 17:53:43 composer? 2021-03-17 17:53:50 composer is basically python 2021-03-17 17:53:52 docker-compose = composer 2021-03-17 17:54:02 We use docker-compose without issue on alpine linux 2021-03-17 17:54:03 (and related) 2021-03-17 17:55:23 ikke: when it deigns to cooperate, which is not always. especially not with more complex setups. believe me, i would know 2021-03-17 17:55:42 We try to avoid complex setups :P 2021-03-17 17:55:56 but at least we manage our gitlab instance with it 2021-03-17 17:56:01 and other things as well 2021-03-17 17:56:19 But I wonder, as metbsd mentioned, docker-compose is python. In what matter is the libc relevant? 2021-03-17 17:58:57 there's some nastiness lurking in a couple of the dependent modules, both libc and just 'that's not the way I want it' 2021-03-17 18:00:15 FWIW, I just back-ported support for apk package manager with CFEngine package promises. https://tracker.mender.io/browse/CFE-3451, will be available in the next round of three different releases (3.12.7 (last 3.12.x LTS release), 3.15.4 and 3.18.0 (next LTS)). Let me know if I can do some work to pull this into one or more releases of alpine. :) 2021-03-17 18:17:41 'try to avoid complex setups' but use docker :) 2021-03-17 18:18:55 helby: docker simplifies things for us 2021-03-17 18:19:35 ikke: 'us' :) how many of you are there? 2021-03-17 18:19:49 2 / 3 2021-03-17 18:20:03 alot docker users 2021-03-17 18:20:32 of course, docker, kubercrap and some other hiphopers are popular 2021-03-17 18:21:12 there is a world of difference between docker + docker-compose and kubernetes 2021-03-17 18:22:12 second that 2021-03-17 18:23:04 but still they have one thing the same :) 2021-03-17 18:33:56 aaaaayup.. compose is very necessary for anything decently complex though. 2021-03-17 18:42:56 "compose is just way to do docker cli commands" ;) 2021-03-17 18:57:03 suuuuuuure, and running systemd-in-systemd is a-okay 2021-03-17 18:57:06 ;) 2021-03-17 19:02:05 ;) 2021-03-17 19:48:00 mps: not yet, im waiting https://gitlab.com/postmarketOS/pmaports/-/merge_requests/2021 to import it to aports 2021-03-17 19:55:19 bl4ckb0ne: I doubt that such pile of patches will be acceptable for alpine 2021-03-17 19:56:47 i want the config 2021-03-17 19:56:50 bl4ckb0ne: in meantime could you test https://dev.alpinelinux.org/~mps/linux-edge-5.11.5-r1.apk but with u-boot parameters you used for https://dev.alpinelinux.org/~mps/linux-5.11.6.tar.gz 2021-03-17 19:56:59 sure 2021-03-17 19:57:47 don't forget 'rootfstype=ext4 rootwait' 2021-03-17 19:58:07 yeah i already had those for the 5.11.6 2021-03-17 19:58:44 yes, I remember 2021-03-17 19:59:39 would be nice to see boot log with these params but for linux-edge-5.11.5-r1.apk 2021-03-17 20:00:00 then I hope we will be closer to final 'fix' 2021-03-17 20:31:11 what's that? an alpinelinux for smartphones? 2021-03-17 20:31:51 postmarketos is that indeed 2021-03-17 20:32:15 it sounds great 2021-03-17 20:32:40 altought I doubt my smartphone supports it 2021-03-17 20:33:26 ouch, it's not supported 2021-03-17 20:34:06 yeah, it's hard to get a broad support for devices 2021-03-17 20:34:28 I suppose that the older the device the more likely to work right? 2021-03-17 20:34:42 I have a Huawei P30 Lite, do you think I could even consider trying? :S 2021-03-17 20:35:03 Drivers are the issue most of the time 2021-03-17 20:35:20 Things being locked down does not help either 2021-03-17 20:36:05 I hope it improves in the future 2021-03-17 20:36:28 It depends on volunteers 2021-03-17 20:36:36 I see xiamo pretty well supported 2021-03-17 20:36:41 xiaomi 2021-03-17 20:37:54 I meant about vendors selling locked hardware/firmwares 2021-03-17 20:38:21 how it is practically usable on supported phones? 2021-03-17 20:41:36 https://www.youtube.com/watch?v=r5AGEujqrVw 2021-03-17 20:44:54 well watched it sooner today already :/ 2021-03-17 20:45:34 but I meant if somebody use it daily, not 10 minutes during shitty youtube stream 2021-03-17 20:46:58 hehe IDK, I just discovered it today 2021-03-18 00:23:04 what command will update initramfs? 2021-03-18 00:27:18 oh, got it mkinitcpio 2021-03-18 00:40:38 FYI: mkinitfs trigger can fail with small (~128M) /boot partitions (due to switch from libLLVM-10.so to libLLVM-11.so) -- https://gitlab.com/postmarketOS/pmbootstrap/-/issues/2022 2021-03-18 00:54:27 :+1: 2021-03-18 02:07:23 I have a problem with one of my shell scripts. It works perfectly on Void Linux, but when I tried to run it on alpine it gives me an error: awk: cmd. line:1: Unexpected token 2021-03-18 02:07:56 probably you have busybox awk, and not proper awk 2021-03-18 02:08:01 I figured that was it. 2021-03-18 02:08:03 that's one way to word it 2021-03-18 02:08:14 another way to word it is 'probably your awk script isn't portable and depends on GNU awk features' 2021-03-18 02:08:17 :) 2021-03-18 02:08:20 Here are the two awk statements: 2021-03-18 02:08:21 can fix it either way 2021-03-18 02:08:26 awk '{ print length, $0 }' $indexfile | sort -nr | head -1 >> "$blockfile" 2021-03-18 02:08:32 unsort=$(awk 'NR == n' n="$g" file.txt) 2021-03-18 02:11:31 Anyone have any idea which bits of that are wrong? 2021-03-18 02:11:47 ChmEarl: wat 2021-03-18 02:12:21 fuckit. apk install coreutils binutils 2021-03-18 02:12:30 neither of those have gawk 2021-03-18 02:12:31 ffs apk add. 2021-03-18 02:12:34 and binutils doesn't even make sense 2021-03-18 02:12:49 I was looking at this page: https://wiki.alpinelinux.org/wiki/How_to_get_regular_stuff_working 2021-03-18 02:12:53 you might as well install gnome, that's more likely to pull in gawk eventually 2021-03-18 02:13:23 yes, gnome pull everything :D 2021-03-18 02:13:45 binutils is a good laugh 2021-03-18 02:13:57 but there is 'gawk' package btw 2021-03-18 02:14:03 for gnufans 2021-03-18 02:14:06 Found that already 2021-03-18 02:14:20 Thanks 2021-03-18 02:14:42 seems like busybox awk is noncompliant and requires an argument to length 2021-03-18 02:14:55 bc. why not use gnu on alpine right 2021-03-18 02:15:25 This is in an LXC container on Proxmox 2021-03-18 02:16:54 Also, I need to use fcron, and frcon isn't in the debian repos for whatever reason, but it is in Alpine 2021-03-18 02:17:20 "Hur dur use monotonic systemd timers instead", is what I imagine the package maintainers said 2021-03-18 02:17:30 anyway yeah, a quick fix is to `apk add gawk` 2021-03-18 02:17:40 (no need for excessive snark here, folks) 2021-03-18 02:18:45 anyways it would be much faster to do awk 'length($0)>n{n=length($0};s=$0}END{print n,s}' 2021-03-18 02:19:05 although i guess that's not exactly the same due to fallback sorting 2021-03-18 02:19:23 length($0)>n&&$0 wait 2021-03-18 02:20:51 I'm trying to pipe the longest line in a file to another file. The first file is an HTML file containing a stupid massive JSON that's all packed into one line, so I seriously doubt any length-counter could mess up my use case. Thanks for the advice. I'm not very experienced with awk. 2021-03-18 02:21:40 also my command mixes up } and ) due to stupid 2021-03-18 02:21:59 By the $0? 2021-03-18 02:22:07 'length($0) > n { n=length($0); s=$0 } END { print n, s }' 2021-03-18 02:22:53 it's not *exactly* the same because i think sort will give you the last line by orthographic sort 2021-03-18 02:23:12 unless you do --stable-sort, in which case it should give you the first line 2021-03-18 02:23:36 I just ran it on a test file, and it worked. Thanks 2021-03-18 02:23:52 highly recommend reading the first few chapters of awk book 2021-03-18 02:24:33 takes like an hour to skim and you will (hopefully) no longer write grep|awk|sed 2021-03-18 02:25:41 I definitely should. Awk just seems like black magic fuckery to me right now. Thanks again. (Also, I didn't realize that "sed -e" and "grep -e" were things that existed until recently, so you can imagine how some of my code looked like before that). 2021-03-18 02:26:15 awk is def one of those weird DSLs that's hard to read until you sorta get it 2021-03-18 02:26:26 like druidism. 2021-03-18 02:26:58 druid-specific language 2021-03-18 02:36:15 awk's not that bad.you just have to NF == 0 && key != "" && nvalues > 0 { printf "%s%s", key, OFS for (i = 2; i < nvalues; i++) { printf "%s%s", value[i],OFS } 2021-03-18 02:37:08 yes, that makes sense ;) 2021-03-18 02:39:02 i assure you, it makes more sense in context with $1 == "'$VALUE_ATTRIBUTE'::" { for (i = 2; i <= NF; i++) { value[i] = $(i) } nvalues = NF b64 = 1 } 2021-03-18 02:41:15 see? easy!@ 2021-03-18 02:42:54 yes, exactly 2021-03-18 02:43:07 I am telling this to my mom every day :/ 2021-03-18 02:47:40 i pulled that out on a java dev who insisted systems engineers were troglodytes when they complained about ldap not working. these days uppity devs get told to figure out msbuild themselves. >:) 2021-03-18 07:24:23 I hate it when people don't expect other peoples professions. just because I don't do it doesn't mean it's not useful... or easy 2021-03-18 07:26:18 I also hate git. so much. linus torvalds should be shot for making it and making us all use it. 2021-03-18 07:47:17 Perl - Practical Extraction and Reporting Language, for text processing instead of .... you name it 2021-03-18 09:46:36 LACampbell: Linus designed a specific tool for his specific use case. Who should be shot are the 'The linux kernel uses it, so it must be good. Let's shoehorn our completely different workflow into it' hipsters. 2021-03-18 10:05:24 <[diablo]> metbsd morning - did you get your Docker IPtables issue resolved by adding the ip6tables package like ikke said? 2021-03-18 13:21:08 Hello! 2021-03-18 13:21:21 I am trying to install patchelf on a new Alpine installation. 2021-03-18 13:21:55 I noticed that it was on edge, so I added edge to /etc/apk/repositories 2021-03-18 13:22:24 And then I ran apk upgrade --update-cache --available and rebooted as suggested by the Wiki. 2021-03-18 13:22:58 Then I noticed that it was in the community repository, so I repeated the procedure 2021-03-18 13:23:07 but for the community edge repository. 2021-03-18 13:23:31 So now I am left with a system using main edge and community edge, but I get two errors when trying to use apk: 2021-03-18 13:23:35 You dit 2021-03-18 13:23:52 libgcc... packaged mentioned in index not found 2021-03-18 13:23:58 Don't need to reboot when you add repositories 2021-03-18 13:24:19 I know, but I was doing it by the book^H^H^H^HWiki 2021-03-18 13:24:45 the other error was for libstdc++ 2021-03-18 13:25:24 Can you show the exact error? 2021-03-18 13:26:31 https://x0.at/RbR 2021-03-18 13:27:18 Can I just ignore the error? I seem to be able to add patchelf now 2021-03-18 13:28:11 Seems that mirror is not up-to-date 2021-03-18 13:28:26 oh, OK - I didn't think of that. I just chose one that seemed close. 2021-03-18 13:29:16 They're a few hours behind 2021-03-18 13:29:27 ah, is that all it is? Thanks! :) 2021-03-18 13:30:30 The index was already synced, but not all the apk files yet 2021-03-18 13:30:47 ah, right... I'll change the mirror... 2021-03-18 13:31:42 yay! Thanks :) 2021-03-18 13:36:10 Great! That's Go patched to use musl now. 2021-03-18 13:36:12 https://www.numbersandreality.com/notes/alpine-wsl/go-lang-on-alpine/ 2021-03-18 13:39:01 Why not install go directly from apk? 2021-03-18 13:39:17 erm... that's an option? :facepalm: 2021-03-18 13:39:24 Yes 2021-03-18 13:39:33 oh well I'm here now :) 2021-03-18 13:42:38 Fun fact: with musl, you can create a completely static go binary 2021-03-18 13:49:04 ooh, interesting. I'll certainly have a look at that 2021-03-18 13:49:25 ikke, I think I had this discussion before in 2021-03-18 13:49:42 #hacking-and-learning:ungleich.ch on matrix. You can do so with glibc, too 2021-03-18 13:50:02 (And probably dietlibc, too, but I haven't checked whether go compiles against it) 2021-03-18 13:50:54 telmich, did you see the talk at FOSDEM 21 about toolchains? 2021-03-18 13:50:56 when I used go on debian I had to compile/link with musl to get fully static exe 2021-03-18 13:51:02 You probably know it already but it's a good talk 2021-03-18 13:51:11 https://fosdem.org/2021/schedule/event/toolchains_choice/ 2021-03-18 13:51:34 Sorry, but I need to be away-from-keyboard now 2021-03-18 13:51:37 see you soon 2021-03-18 14:48:02 I'm still struggling with an issue I posted here in compiling wine on alpine: https://forum.winehq.org/viewtopic.php?f=8&t=35046 Since posting, I've tried adding CFLAGS=-no-pie to my apkbuild, but with no success (I get the same segfault). What else might I be able to try? 2021-03-18 14:49:04 I see success here on Void by setting a _nopie flag (https://github.com/void-linux/void-packages/issues/25644) which is described as "disabling building the package with hardening features (PIE, relro, etc)" here (https://github.com/void-linux/void-packages/blob/master/Manual.md). 2021-03-18 14:49:16 Unsure how I can recreate those build conditions 2021-03-18 16:48:56 https://www.alpinelinux.org/community/ → the "mailing list" points to a 404 :/ 2021-03-18 16:50:08 https://lists.alpinelinux.org/lists/%7Ealpine works for me 2021-03-18 16:51:17 https://lists.alpinelinux.org/lists/%7Ealpine is 404 not found here :/ 2021-03-18 16:51:34 mh, it works with curl 2021-03-18 16:51:38 but not in my firefox 2021-03-18 16:52:37 Why would ~ need to be encoded in a URL anyway? Is firefox choking on that? 2021-03-18 16:52:46 Try it as https://lists.alpinelinux.org/lists/~alpine 2021-03-18 16:52:59 zcrayfish: it came out like that when I copied the link 2021-03-18 16:53:09 in the browser bar it's just ~ for me 2021-03-18 16:53:30 https://imgur.com/VNi4rRt.png looks like this 2021-03-18 16:55:07 Your Firefox is sending it as %7E 2021-03-18 16:56:15 Chromium sends it as ~ regardless of if it's urlencoded. 2021-03-18 20:01:55 Intel Corporation Mobile GM965/GL960 2021-03-18 20:02:00 what firmware name for this? 2021-03-18 20:04:51 i installed linux-firmware-i915 2021-03-18 20:04:54 hope it's right 2021-03-18 20:24:22 metbsd: that should be for all intel, nothing more there so 2021-03-18 21:22:48 modern linux kernels print firmware name in dmesg 2021-03-18 21:26:52 Hello71: like what? 2021-03-18 21:27:31 uh... it could be only if not found 2021-03-18 21:29:51 i think ddevault also added a feature to make it print the paths searched or somesuch 2021-03-18 21:30:48 it shows a lot of crap, so not easy to find something, but regarding gpu also xorg.log can help 2021-03-18 21:30:48 but still if u have some firwmare u don't need much to know that it was loaded, more u need it if u miss something :/ 2021-03-18 21:30:48 xorg log is unrelated to firmware 2021-03-18 21:30:48 what's relatively hard with gpu, as there are 100 drivers and whatever xorg will find, that will load 2021-03-18 21:30:48 Hello71: yes, I had drivers in my head 2021-03-18 21:40:30 modinfo `lsmod | grep -v Module | awk -e '{ print $1}'` | grep firmware 2021-03-18 21:40:33 btw. there were some cool command to find which firmware is loaded 2021-03-18 21:40:44 well something like that ;) 2021-03-18 21:41:06 so I insalled all firmware then checked it and deleted all useless crap obviously 2021-03-18 21:41:12 mps: u should delete wiki too! 2021-03-18 21:43:12 delete everything 2021-03-18 21:43:12 well, nothing is there so not exactly 'everything' 2021-03-18 21:43:12 that's the path to enlightenment 2021-03-18 21:43:12 rm -rf / 2021-03-18 21:43:12 sounds good 2021-03-18 21:43:12 nirvana is samsara, samsara is nirvana 2021-03-18 21:43:24 funny thing is that loading kms disappeared from wiki somehow 2021-03-18 21:43:43 so probably only one line which was okay there is no more there 2021-03-18 21:44:03 but encouraging users to install bash is still there 2021-03-18 21:44:22 somebody whould pay for that ... with life obviously 2021-03-18 22:40:24 I'm having trouble compiling a program on Alpine. It gives me this error: "collect2: fatal error: cannot find 'ld'" (here's the program: https://github.com/JayXon/Leanify) 2021-03-18 22:41:16 I already have gcc installed 2021-03-18 22:43:48 do you have "build-base" installed? 2021-03-18 22:44:22 Ues 2021-03-18 22:44:24 *Yes 2021-03-18 22:57:16 i'd suspect a weird PATH or something 2021-03-18 22:57:46 maybe running the build under 'strace -ff -e trace=execve' will show some useful information 2021-03-18 23:09:47 ajshell1: check that you have binutils and libtool installed? 2021-03-18 23:10:34 ... I did not have libtool installed. Let's try again 2021-03-18 23:11:36 RootWyrm: Nope. That still didn't fix it. 2021-03-18 23:19:37 jn__: https://pastebin.com/15ZzQ93H 2021-03-18 23:21:49 hm, that doesn't show how it searched for ld 2021-03-19 00:30:58 -fuse-ld=gold 2021-03-19 00:32:23 not sure how much point there is to this program considering different tradeoffs for different formats 2021-03-19 00:33:13 seems like most of these "optimizations" are lossy 2021-03-19 01:54:36 Hey does anyone know what populates $CC in APKBUILD files and if ccache is used as a part of it? 2021-03-19 01:55:13 I'm building the linux-lts package and I want to figure out how to speed up build times. 2021-03-19 01:57:01 markpash: you could find out (by using echo in a test APKBUIL), and you could absolutely set it yourself 2021-03-19 01:57:13 (it's just a shell script after all) 2021-03-19 02:10:17 halosghost: So I gave that a go, it just uses gcc. I was wondering how I could use ccache to accelerate repeated builds without really changing the APKBUILD. 2021-03-19 02:11:38 markpash: why would it be an issue adding `CC="ccache "` ? 2021-03-19 02:12:10 abuild has support for using ccache via $USE_CCACHE 2021-03-19 02:12:17 Could I do something like that in abuild.conf? 2021-03-19 02:12:28 maxice8: that sounds very pleasant 2021-03-19 02:12:34 markpash: do that ☺ 2021-03-19 02:12:57 echo "USE_CCACHE=1" >> [/etc/abuild.conf|~/.abuild/abuild.conf] 2021-03-19 02:13:02 yeah so I enabled the line in abuild.conf, but how do I get it to take for the build step for this package? (linux-lts) 2021-03-19 02:13:42 Does abuild export that as a var if I enable it in the config? 2021-03-19 02:14:32 And is there any way I can stuff it into a var that's picked up in either the APKBUILD or the makefile? 2021-03-19 02:15:05 Sorry for all the questions, I just don't know how USE_CCACHE is wired up in abuild 2021-03-19 02:15:25 markpash: never apologize for having lots of questions 2021-03-19 02:15:40 markpash: there's no need ☺ having lots of questions just means there's a lot you're interested in learning 2021-03-19 02:15:46 that's never a bad thing 2021-03-19 02:16:03 Awesome :) 2021-03-19 02:21:00 markpash: have you looked at the wiki regarding ccache? (I haven't; but it's always worth looking) 2021-03-19 02:23:11 There's not much on the wiki about it 2021-03-19 10:06:49 no alpine iso can boot in vmware 2021-03-19 10:17:46 hi, can someone tell me why an "/etc/init.d/chrony start" fails with return value of 1, when then service is already running 2021-03-19 10:17:52 *chronyd 2021-03-19 10:18:59 my ansible playbooks fails to run for this very reason because I have set it to "state: started" just to make sure it's started. but because of the init script return value the playbook fails to run 2021-03-19 10:35:30 did xwayland break for anyone else in the past couple of days? 2021-03-19 10:55:09 I'm gonna try to swtich to alpine-linux as main desktop OS, I would like to use KDE, firejail, some KVM vm's (or maybe Xen :S) , selinux?... Should I be warned about some potential problems? For this kind of setup is sys-install the best way? 2021-03-19 10:56:02 don't 2021-03-19 10:57:00 don't should have problems? 2021-03-19 10:57:11 don't do it 2021-03-19 10:57:20 hehe 2021-03-19 10:57:46 uhM: metbsd | no alpine iso can boot in vmware , really? 2021-03-19 10:57:56 I have alpine vm's running in vmware workstation 2021-03-19 10:58:16 i fixed it 2021-03-19 10:58:24 so do you think that alpine desktop experience is not very good? 2021-03-19 10:58:47 it's very good and very hard 2021-03-19 10:58:56 I'm asking because I tried some years ago and it was pretty painful 2021-03-19 10:59:07 but maybe now it improved a lot 2021-03-19 10:59:15 it's a lot better now than earlier, at least package wise 2021-03-19 10:59:26 maybe depends what os you come from 2021-03-19 10:59:34 should KDE run easily? 2021-03-19 10:59:53 there's a lot of KDE stuff packaged so I don't see why not 2021-03-19 10:59:59 but I've never tried it myself 2021-03-19 11:00:05 I didn't think Alpine was designed with desktop in mind. It's all about making tiny docker images. 2021-03-19 11:00:22 it's not really designed but it works just fine IMO 2021-03-19 11:00:26 there are many kde-optimized distributuion 2021-03-19 11:00:31 https://wiki.alpinelinux.org/wiki/KDE 2021-03-19 11:00:31 yes yes I know it is not its design objective 2021-03-19 11:00:58 alpine is nice because it's simple, which should also extend to desktop I think 2021-03-19 11:01:51 I think that I should just try 2021-03-19 11:01:59 opensuse will give you way better experience running kde linux 2021-03-19 11:02:17 could always try in a VM if you just want to see if you can install everything you need 2021-03-19 11:02:18 also kubuntu 2021-03-19 11:02:21 my other option in mind was fedora 2021-03-19 11:02:47 good choice 2021-03-19 11:03:06 different things for different job 2021-03-19 11:03:19 kubuntu will give you a better package management experience than anything CentOS / RedHat based 2021-03-19 11:03:45 what package do you need 2021-03-19 11:04:16 why? I'm running qubes based on fedora and fedora templates and I don't notice package management problems 2021-03-19 11:04:22 there are snapd, flatpak, appimage 2021-03-19 11:04:30 they are not enough? 2021-03-19 11:04:41 I only use snapd for spotify 2021-03-19 11:04:52 pakcages are all the same 2021-03-19 11:05:01 apt dnf 2021-03-19 11:05:08 the other things is pretty straigforward, firefox, thunderbrid, mumble... 2021-03-19 11:05:29 my problem is that I some kind of ex debian fanboy 2021-03-19 11:05:39 i remember opensuse optimize their kde 2021-03-19 11:05:40 so I'm not very huge fan of debian derivates hehe 2021-03-19 11:05:49 kde must be nice on opensuse 2021-03-19 11:06:28 I feel that I'm trying to swtich my OS in some more community/social aspect than thecnical 2021-03-19 11:07:22 alpine community seems very alive 2021-03-19 11:07:37 opensuse leap for kde, ubuntu for gnome 2021-03-19 11:07:42 but what's best for xfce 2021-03-19 11:08:08 I will test alpine and kde on my other laptop 2021-03-19 11:08:33 metbsd: xubuntu 2021-03-19 11:08:56 that's what i'm using 2021-03-19 11:09:11 always give me glitch though 2021-03-19 11:16:18 Meh 2021-03-19 11:17:38 buntu is actually not bad 2021-03-19 11:17:46 but for kde i'd go for opensuse 2021-03-19 11:42:20 wyoung: lol, just today there is a CVE for rpm https://bugzilla.redhat.com/show_bug.cgi?id=1934125 2021-03-19 11:55:15 markpash: https://gitlab.alpinelinux.org/alpine/abuild/-/blob/master/abuild.in 2021-03-19 11:56:23 however CONFIG_GCC_PLUGINS is incompatible with ccache 2021-03-19 13:12:51 May be a silly question, but where can I find POSIX? I'd like to know the standard, since I want to write a system utilities package and I'd like to know what POSIX wants from such a package. 2021-03-19 13:17:39 https://www.opengroup.org/austin/ 2021-03-19 15:19:49 lol, I'm just testing KDE on alpine, wayland login rebooted sddm 2021-03-19 15:20:19 Is there some apk backend for discover? Would it have sense to write one? 2021-03-19 15:23:27 There's libapk-qt, I think pmOS already uses that 2021-03-19 15:26:31 arkho: shellhaters.org is a very nice index into the POSIX spec. 2021-03-19 15:27:40 it seems that there is only on edge 2021-03-19 15:41:10 xordspar0: Thanks! 2021-03-19 17:06:30 hello everone 2021-03-19 17:07:30 q: is it possible to make the default syslog to redirect logs to a remote server? (possible with rsyslog but alpine uses syslog by default I think) 2021-03-19 17:10:15 sam_son: not surew but the Busybox syslog is very basic, you're better off with either rsyslog or syslog-ng for that, or use something else like filebeat to ship the logs 2021-03-19 17:10:32 ah I see 2021-03-19 17:11:16 I looked at 'apk list --installed' and I couldn't even find syslog there, so I guess you answered my next question, that syslog is part of busybox 2021-03-19 17:12:11 since I'm using rabbitmq already, rsyslog with the rabbitmq plugin would be ideal, I'll give that a try in my test environment and see how it behaves 2021-03-19 17:13:08 sam_son: apk info -W /sbin/syslogd 2021-03-19 17:13:26 target is owned by busybox 2021-03-19 17:13:38 nice, I didn't know about the -W parameter, thanks 2021-03-19 17:14:03 Yeah Busybox syslogd is little on features 2021-03-19 17:14:09 s/little/light/ 2021-03-19 17:14:36 no worries, rsyslog is available 2021-03-19 18:10:41 for some very odd reason centos isn't offering alpine images for lxc anymore 2021-03-19 19:38:03 Hello alpine 2021-03-19 19:38:13 \o 2021-03-19 19:46:49 Well I’m having an issue 2021-03-19 19:46:56 maybe alpine related may be not 2021-03-19 19:47:42 I have the livego container running on alpine 2021-03-19 19:48:00 and I want to secure the website it serves 2021-03-19 19:48:16 it will only provide a http url not the https 2021-03-19 19:48:32 I am a noob so idk how to do any of that 2021-03-19 19:48:42 it was a little easier to do on Debian with Apache 2021-03-19 19:55:40 What part are you strugling with? 2021-03-19 19:56:48 whole life! ;) 2021-03-19 19:57:45 leibniz: check 'uacme' package in alpine for cert, nginx or shitty appache should be the same 2021-03-20 00:07:27 Hey! I just found out my alpine's time is out of sync by a few minutes. This messes with 2fa. How can I get alpine to sync ntp properly? 2021-03-20 00:09:00 Peasant65: u should run ntpd permanently, u can run it also manualy 2021-03-20 00:11:01 I run busybox ntpd and had no issue 2021-03-20 00:11:27 I selected busybox on install, but I don't think it's running (can't find it in htop) 2021-03-20 00:12:37 probably can try again as root setup-ntp 2021-03-20 08:19:52 u 2021-03-20 08:19:59 meh, EWIN 2021-03-20 08:23:52 hey, I'm trying to share a directory from a host to alpine VM using virt-manager, but it doesn't seem to work 2021-03-20 08:24:49 I'm getting "9pnet_virtio: no channels available for device /host_dir" in dmesg, but all advice for that doesn't seem to work 2021-03-20 09:28:49 ignaloidas: did you fix it? I can't test at this moment but probably in near future would be interested on that 2021-03-20 09:29:44 donoban, not yet, tried every option on virt-manager, but that didn't help 2021-03-20 09:31:48 well as a workdaround you could use ssh and/or sftp,ssfhs 2021-03-20 09:31:57 sshfs* 2021-03-20 09:36:41 ignaloidas: could you chec if are this modules are loaded: virtio_input, 9pnet, 9pnet_virtio, virtio_pci 2021-03-20 09:59:30 .o/ 2021-03-20 10:03:17 donoban, yep, all of them are loaded 2021-03-20 10:03:55 I have a question, should be easy to ansnwer, should I use alpine linux? 2021-03-20 10:04:14 lol 2021-03-20 10:04:41 Or should I use FreeBSD? 2021-03-20 10:04:50 wyoung: no, you should be world class football player 2021-03-20 10:05:34 artok: I should use a world class football player? 2021-03-20 10:05:47 I am asking what I should use, not what I should be 2021-03-20 10:06:18 wyoung, your decision, depends on what you want to do 2021-03-20 10:06:33 I want a tidy docker image 2021-03-20 10:06:42 I have some some very small FreeBSD images 2021-03-20 10:06:48 and Alpine is tiny too 2021-03-20 10:07:26 I know I should expect a bias response here, but I am sure people use alpine-linux for other reasons, not just because it is the smallest 2021-03-20 10:10:17 I'm thinking on install alpine as my main desktop, I'm inclined to use luks+btrfs with at least different subvolumes for root and home 2021-03-20 10:10:40 but I'm not sure if would be a good idea to create also subvols for /var or anothers dirs 2021-03-20 10:10:54 I usually just create subvols for /home and /root 2021-03-20 10:12:00 it seems enought, it would be nice to do experiment like installing different versions or OS in other subvolume root 2021-03-20 10:12:15 There's little point in creating more subvolumes unless you want to use them independently (can be handy for /home when you switch roots), want to snapshot them separately (e.g. / before an upgrade) ot want different mount options (e.g. disable CoW for volumes where a DB is 2021-03-20 10:12:40 You can always create new subvolumes later on for additional distros :) 2021-03-20 10:12:53 yes, for a desktop usage it seems pointless 2021-03-20 10:13:02 are you runing alpine as desktop? 2021-03-20 10:31:09 donoban: Yup 2021-03-20 10:37:41 I'm currently using qubes, altough I suppose that an equivalent security isolation is near impossible with a linux based host... I would like to isolate all possible things with firejail, containers (docker/podman), kvm... 2021-03-20 12:38:29 Hello! 2021-03-20 12:39:47 Is there an option to tell default alpine-linux client (udhcpc) to send the current hostname on the dhcp request? 2021-03-20 12:40:53 bitblt: what version of alpine? 2021-03-20 12:40:59 3.12 2021-03-20 12:44:16 hostname $(hostname) 2021-03-20 12:45:56 ikke: Should I put this on some udhcpc specific configuration or on /etc/network/interfaces? 2021-03-20 12:47:43 latter 2021-03-20 12:47:48 note that this changes in 3.13 2021-03-20 12:48:02 ikke: How this changes? 2021-03-20 12:48:27 (Its a vagrant box so this could be updated soon) 2021-03-20 12:49:39 It switches to ifupdown-ng 2021-03-20 12:49:48 Then there is an explicit setting for iut 2021-03-20 12:50:37 Ah I see 2021-03-20 12:51:17 The only problem I have is that /etc/networking/interfaces configuration gets generated by vagrant https://github.com/hashicorp/vagrant/blob/master/templates/guests/alpine/network_dhcp.erb 2021-03-20 12:51:36 So I suppose I must patch -h udhcpc flag elsewhere 2021-03-20 13:11:25 Peasant65: busybox ntpd only runs once on boot 2021-03-20 13:11:33 sorry, ntpdate 2021-03-20 13:11:40 actually, hm 2021-03-20 13:12:12 check rc-service ntpd status 2021-03-20 13:12:51 ignaloidas: i don't think you can put / in device name 2021-03-20 13:13:29 there is a ntpd service 2021-03-20 13:13:34 and ntpd has as -d flag 2021-03-20 13:21:46 Hello71, all the guides I see online use the slash 2021-03-20 13:23:24 oh, I was dumb, nevermind 2021-03-20 13:23:39 mixed up the arguments in virt-manager config 2021-03-20 13:23:48 (though you still use the slash) 2021-03-20 15:40:23 is there any reason for no including btrfs as sys option like lvmsys ? or including an option for full encryption? 2021-03-20 17:12:38 metbsd: Did you have issues with your containers again? 2021-03-20 17:13:15 ikke, nope. not since i add ip6tables 2021-03-20 17:13:36 metbsd: alright, I will backport that fix 2021-03-20 17:13:43 did you? 2021-03-20 17:14:29 metbsd: I only noticed it on a system right after the system was rebooted, and I haven't rebooted that system agian 2021-03-20 17:15:09 same here. it's quite stable 2021-03-20 17:15:12 and fast 2021-03-20 18:16:13 Did anybody build hikari wm in alpine? 2021-03-20 18:44:16 trying to build that shit, by my knowledge about building crap is close to zero 2021-03-20 18:44:32 don't worry, I don't expect any help here, just saying :/ 2021-03-20 18:44:56 :S 2021-03-20 18:45:08 why do you want to build it if you consider it shit? 2021-03-20 18:46:48 it was not meant that way :/ 2021-03-20 18:46:58 try posting your build errors 2021-03-20 18:47:09 by shit I meant, little shitty piece of crap 2021-03-20 18:47:16 https://paste.dismail.de/?9e5f3a99b1c23109#FjVfJdG3ReFEEsbUk1WkZtcHwD9qAzuYccK9DyK1Sj5A 2021-03-20 18:47:27 that I get when try it with bmake as mentioned in hikari doc 2021-03-20 18:48:32 lot of warnings about wlroots don't found in the path 2021-03-20 18:49:05 and finally fails becaouse it don't find wayland-scanner 2021-03-20 18:49:23 it seems that you miss a lot of dependecies 2021-03-20 18:49:49 I don't know if they are included in alpine 2021-03-20 18:49:49 but it blaiming about dependencies I even have installed 2021-03-20 18:49:58 not sure what is pkg-config it find nothing 2021-03-20 18:50:47 what about wayland-scanner? 2021-03-20 18:54:45 did not find it 2021-03-20 18:54:57 not sure what that package is, first time hear about it 2021-03-20 18:55:18 but whatever I installed, error output is the same as pkg-config whatever it is find nothing 2021-03-20 18:56:05 Hello ppl 2021-03-20 18:56:44 I managed to boot a fully working Alpine on a AoE-based SAN 2021-03-20 18:56:44 is there maybe a way to somehow build package with abuild somehow from aur - archlinux or gento 'template' ? 2021-03-20 18:57:02 I wonder if anyone i sinterested on an how-to or wiki page about it 2021-03-20 18:57:27 the process can be modified to work with iscsi of course 2021-03-20 18:57:28 helby: someone created a converter from PKGBUILD to APKBUILD 2021-03-20 18:57:41 it's actually very cool :) 2021-03-20 18:58:14 'AoE-based SAN' is? 2021-03-20 18:58:52 ikke: in that case I just then run that abuild without any make bmake and other crap? 2021-03-20 18:59:35 If the project requires bmake, it still needs to be installed as a makedepend 2021-03-20 18:59:37 ikke: just got confused, bc. did not find anything in alpine docs/wiki or anywhere about 'make' and other similar toosl 2021-03-20 19:00:23 ikke: yes, but I meant, that a building proces of package is using tools like abuild as in wiki and it take care of tools like 'make or bmake' ... ? 2021-03-20 19:00:42 helby: It does what you tell it to do 2021-03-20 19:00:59 You specify the steps that are required to build a package 2021-03-20 19:01:06 which might include calling makje 2021-03-20 19:01:13 or other tools the project uses 2021-03-20 19:01:19 Please take a look at other APKBUILDs 2021-03-20 19:01:24 AoE is ATA-OVER ETHERNET, an ancient protocol, but truly fast, and well-written 2021-03-20 19:01:31 part of Linux since long ago 2021-03-20 19:02:19 it works in layer 2, similar to fibre cjannel 2021-03-20 19:05:41 ikke: yes, thanks, any idea where can I find that converter? 2021-03-20 19:05:49 It's called 'toapk' 2021-03-20 19:10:40 ikke: thanks will try that piece :/ 2021-03-20 19:11:09 fucking hihkari use matrix instead of irc :/ 2021-03-20 19:12:41 what a mess 2021-03-20 19:20:46 Hi, I'm running a wordpress installation with mariadb on alpine, but mariadb occasionally crashes. I've searched through the logs, but haven't found any information at all. rc-service just gives me status: crashed and I have to restart it manually. 2021-03-20 19:20:50 Any ideas? 2021-03-20 19:26:15 Stupid question but does Alpine's ls indicate acls with a '+' sign? 2021-03-20 19:27:22 i guess that depends on which ls you use? 2021-03-20 19:29:27 the default one. Cannot seem to get acls to work 2021-03-20 19:30:34 lsattr 2021-03-20 19:31:13 e2fsprogs-extra 2021-03-20 19:33:55 thanks mps, will look tomorrow. 2021-03-20 20:07:43 johnaj: antything in dmseg? maybe it's a segfault or OOM 2021-03-20 20:12:00 donoban: hm, I'm checking now, not finding anything, but maybe it's been wiped, because I restarted since the last time it happened. 2021-03-20 20:21:47 ok 2021-03-20 20:53:31 lsattr is unrelated to acls 2021-03-20 20:54:28 based on https://git.busybox.net/busybox/tree/coreutils/ls.c i think acl is not supported 2021-03-20 20:54:45 you can install coreutils or getfacl 2021-03-20 20:54:52 getgetfattr 2021-03-20 20:54:56 technically you can also display it with getfattr 2021-03-20 20:55:01 getfattr 2021-03-20 20:56:23 coreutils doesn't hava getfacl 2021-03-20 20:56:59 acl pkg contains it 2021-03-20 20:58:12 when it is appeared on linux I used it for some time but 'ditched' later as not much useful 'thing' 2021-03-20 20:58:41 I use it once in a while 2021-03-20 21:00:22 ACTION was (and still is) early adopter of a lot of this shiny crap 2021-03-20 21:02:08 built kde from source before it was packaged for any distro, same with gnome, same with xfce, same with systemd ..... 2021-03-20 21:06:26 nowadays contemplating buying apple macbook with macos and forget linux, and forget nearly 30 years of promoting and advocating open source. linux becoming distro for interest of big companies 2021-03-20 21:07:33 disappointed? yes 2021-03-20 21:07:36 it will be a bad decision 2021-03-20 21:07:41 I tried it, it didn't work out for me 2021-03-20 21:08:13 it's nice to have a "normal person" OS that still has a working toolchain though 2021-03-20 21:08:19 tehcloud: then windows ;) 2021-03-20 21:08:36 I can't use Windows anymore, I actually am clueless if I sit down in front of it 2021-03-20 21:08:57 it has been over 20 years since I used it regularly 2021-03-20 21:09:17 never used windows, because that I'm contemplating about macos (osx) 2021-03-20 21:10:10 it at least has a working toolchain and open source tools available 2021-03-20 21:10:23 but linux going wrong path for sure 2021-03-20 21:10:24 the transition is not extremely painful, but it's still not quite Linux 2021-03-20 21:10:40 well, most distros are going down the wrong path... anything with systemd has done so 2021-03-20 21:11:11 but at the same time I am seeing more "good" distros popping up nowadays as the older players go down the wrong path 2021-03-20 21:11:17 not only systemd, but all these GUI desktops are real crap 2021-03-20 21:11:28 yeah, I can't stand any of them either, they're all trash 2021-03-20 21:11:46 GTK is trash too 2021-03-20 21:12:24 get yourself a nice tiling window manager and you'll be happy :) 2021-03-20 21:12:26 and these new crap languages which allow clueless programmers to write programs, now maybe even kernel drivers 2021-03-20 21:13:00 tehcloud: you are advising hard core awesome wm user ;) 2021-03-20 21:13:14 don't give up... us UNIX greybeards can keep C alive :p 2021-03-20 21:13:52 uh, even cut my beard two weeks ago, though not hair, still :) 2021-03-20 21:14:17 so that's what happened... you need to get it grown back fast, and stop entertaining these ideas about using proprietary software 2021-03-20 21:15:22 if it will be red-brown-black again I would let it to grow, but grey ... hmm 2021-03-20 21:15:54 mine's been grey my whole life thanks to a genetic quirk where the middle of it never had colour 2021-03-20 21:19:20 tehcloud: no worries, I will not abandon linux in near future, but long time - who knows 2021-03-20 21:22:34 my son 'must' use macos for his $job and looking his complicated environment and his complaints daily about it I think I understand what you wanted to told me 2021-03-20 21:24:09 since Steve Jobs passed, Apple products have been going downhill fast 2021-03-20 21:26:48 hmm, this M1 macbook is good machine, I even booted alpine on it 2021-03-20 21:27:30 I can imagine 2021-03-20 21:28:03 propiertary even air around chip 2021-03-20 21:28:30 and running alpine in qemu on it is really good experience, fast boot and alpine in qemu is faster that on i7 bare metal 2021-03-20 21:30:07 also u did not mentioned where did u steal that machine 2021-03-20 21:33:08 my son bought it for his work, and I 'steal' it sometimes to test alpine on it 2021-03-20 21:33:36 lol, so you're judging it based on how it runs Alpine 2021-03-20 21:33:38 that's awesome 2021-03-20 21:34:04 but if you're running the Apple stuff and have been for many years you can see the degradation of quality 2021-03-20 21:34:23 yes, I don't have experience with macos, except to start terminal 2021-03-20 21:34:55 I used that shit like 5 or so years ago 2021-03-20 21:35:28 I left that crap after so many issues did not get fixed, but every fucking update some emojis were added 2021-03-20 21:35:36 Apple was a pretty good open source citizen for awhile but since Jobs passed they have been doing the proprietary lockdown and adding gimmmicks to hardware 2021-03-20 21:36:22 actually my son gave me his old macbook (i5) but I just booted alpine iso and installed alpine, didn't even tried to run macos 2021-03-20 21:37:25 tehcloud: hmm, I don't think that intel/amd and arm are better on this 2021-03-20 21:37:52 well, when I say hardware gimmicks I am mostly referring to that stupid touchbar that was added right after Jobs died 2021-03-20 21:38:15 and I also don't like that they made their own 3D API instead of getting on board with Vulkan and continuing to maintain a working OpenGL implementation 2021-03-20 21:38:28 they all have some closed bios/firmware and cpus in cpus 2021-03-20 21:40:14 I would like to see computer with 'reset vector' not hidden and locked on some ROMs 2021-03-20 21:40:47 or reset program pointer, to say so 2021-03-20 21:45:59 at least they use some new tech, not ancient ps/2 mouse connectors like I see often on regular motherboards ;) 2021-03-20 21:46:53 that M1 also looks like a nice progress 2021-03-20 21:48:32 keyboard is not good as on previous models, more like on cheap chromebooks 2021-03-20 21:49:27 actually, old (5 years) samsung peach pi chromebook have better keyboard 2021-03-20 21:54:13 what model u speak about? these M1 air/pro models? 2021-03-20 21:55:04 well there is also option mini with all proper external components 2021-03-20 22:11:04 virtualbox fail mi in pipeline cos kernel headers.. i can fin a right build depends for.. any hint? 2021-03-20 22:11:12 https://gitlab.alpinelinux.org/Sodomon/aports/-/jobs/351908 2021-03-20 22:15:23 You need to specify the kernel version of the kernel that is available 2021-03-20 22:15:33 not the one that is running on the host 2021-03-20 22:15:55 linux-lts is 5.10.24 on edge 2021-03-20 22:19:13 helby: macbook pro M1 2021-03-20 22:30:58 ikke sodomon added linux-lts-dev but this is only 3.11+ what similar package is for 3.10 ? 2021-03-20 22:32:19 umm ikke that was a build dinamic module your response dont said anything in fact i already know thtat.. i mean how to added build depends that not are hardcoded to the branch 2021-03-20 22:32:40 linux-vanilla-dev? 2021-03-20 22:32:48 checking 2021-03-20 22:33:53 hey mps APKBUILDs support alternative depends.. like the own great debian such like "linux-vanilla-dev | linux-lts-dev" ? 2021-03-20 22:34:07 no 2021-03-20 22:34:28 i m tiryed of limitation puff.. but thanks .. i must brand again the package.. 2021-03-20 22:37:13 mckaygerhard: maybe this can be solved by some shell trickery in APKBUILD, but I dind't tried 2021-03-20 22:37:54 no.. too much work.. and merge request was made .. sodomon does not have enough network internet and that sh*t of gitlab is heavy.. 2021-03-20 22:38:20 really guys.. is very hard collaboration .. puff 2021-03-20 22:38:50 well i hope this time virtualbox packge builds.. mps.. limitatios was the rule in this course 2021-03-20 22:39:22 virtualbox shold be removed, not built imo 2021-03-20 22:40:57 is resolved now mps.. i have runing in my alpine vmware desktop.. is just that env in home is not clean.. but now i have a detail.. i must pust -j4 to able to build .. why? if not process i dead itselft! why? 2021-03-20 22:42:07 mckaygerhard: /etc/abuild.conf and set JOBS number, iiuc your question 2021-03-20 22:42:33 i mean.. in pipeline.. in my home buils perfect 2021-03-20 22:42:37 or 'export JOBS=4' 2021-03-20 22:42:45 aha 2021-03-20 22:42:52 err if piupeline doe snot have 4 procesors?¡ 2021-03-20 22:43:03 sorry i m writing too hurry .. 2021-03-20 22:43:06 didn't looked how it is set on CI 2021-03-20 22:43:29 so .. how can we detect correct "-j" for apkbuilds ? 2021-03-20 22:44:35 well, APKBUILD is shell script, so what can be done in shell is in 'your hands' 2021-03-20 22:45:21 it seems build mps but that is automatic? maybe parallel is not good for virtualbox build.. as you said is in a config file.. but for pipelining is set where? 2021-03-20 22:46:27 mckaygerhard: really have no idea, never needed so didn't looked, ask ikke tomorrow, he will know I think 2021-03-20 22:48:36 Hello alpine linux users! I am currently trying to install alpine linux in disk mode on my raspberry pi 3 2021-03-20 22:49:04 brandflake11: I answered on #linux-devel, it is bug 2021-03-20 22:49:11 umm mps ikke .. now i adde the linux-lts-dev and still no luck.. 2021-03-20 22:49:20 mps: I'm sorry, my irc client kicked me out, I couldn't see the message 2021-03-20 22:49:38 mps: Is there any way to get past the bug? 2021-03-20 22:49:47 ok, here it is, mckaygerhard: really have no idea, never needed so didn't looked, ask ikke tomorrow, he will know I think 2021-03-20 22:50:02 ah .. also about rasperri pi .. how i can setup wifi if i do not have wifi enought packages in alpine img ? mps.. is a common question in my telgram groups 2021-03-20 22:50:15 meh, I'm confused 2021-03-20 22:50:16 ummm pufff 2021-03-20 22:51:22 Sodomon: is here! ¬_¬ .. tel lme or past the link to photo of virtualbox to mps and ikke.. it build locally but not in ci .. we are very very confused too 2021-03-20 22:51:36 brandflake11: I'm not sure, except if you edit setup-disk before starting it and fix part about vfat only for RPis 2021-03-20 22:52:32 mps: Is setup-disk just a script? Do you mean I can just edit it in a text editor? 2021-03-20 22:52:39 mckaygerhard: what is missing? last few releases worked quite fine for me 2021-03-20 22:53:06 i explain.. : 2021-03-20 22:53:07 brandflake11: /sbin/setup-disk, and yes it is script 2021-03-20 22:53:16 mps: Thanks! I'll look into editing it 2021-03-20 22:53:59 brandflake11: it have part that it only 'use' vfat as FS for rpis, which is bug obviously 2021-03-20 22:54:43 brandflake11: I think addind '| ext4' could fix it 2021-03-20 22:54:49 mps ikke sodomon created the APKBUILD for virtualbox.. and locally it works, obviously it is not a clean environment.. and it has headers for its own kernel uses, but in the alpine CI it always fails, it says it has no headers.. but they are put with the linux-lts-dev package but it seems that it is not correct since these headers do not work for most of the things in the APKBUILDS 2021-03-20 22:54:52 but didn't tested 2021-03-20 22:55:48 mckaygerhard: huh, you want me to look at this MR I noticed and ignored 2021-03-20 22:55:57 whe i said "most of the thinks" i said many APKBUILD we created but hard to upload due those umm.. "estrictions" ? (to do not said limitations) 2021-03-20 22:56:50 and I promised to seld (and some other people) that I *will not* look to MRs anymore 2021-03-20 22:57:01 s/seld/self/ 2021-03-20 22:57:41 no mps for our problem yo dont need to look for MR .. if you already know that packages that depends of kernel headers always fails.. 2021-03-20 22:58:58 mckaygerhard: for normal APKBUILD you don't need linux-$flavor-dev, linux-headers are enough 2021-03-20 22:59:18 ok trying.. with linux-heders 2021-03-20 22:59:40 linux-$flavor-dev is mostly for local build, out of tree drivers usually 2021-03-20 23:00:40 but I have to go to bed, to late is here now, good night all 2021-03-20 23:01:19 yo/ 2021-03-20 23:03:38 nonsense.. "outr of the tree " is external third party modules and is the correct build depend but is not working i m tyred of limitations 2021-03-20 23:06:42 cd aports ; grep -r "\ cd aports ; grep -r "linux-lts-dev" . 2021-03-20 23:40:12 lexit 2021-03-21 00:00:29 hehe 2021-03-21 00:01:00 I just boot a sys install with luks/btrfs, I have a problem with the keyboard layout on luks passphrase 2021-03-21 03:35:56 hi how do I do the equivalent of man -K on alpine 2021-03-21 04:01:53 I'd just install the mandoc package; that version has the -k option. 2021-03-21 04:07:31 what I need is -K 2021-03-21 04:07:42 maybe I need to install man-db instead of mandoc 2021-03-21 04:07:52 I need to search the full text of the formatted text 2021-03-21 04:12:36 ah, I missed the lowercase, but yes, the manpage for the man included in man-db includes that option. 2021-03-21 04:12:55 lowercase-->uppercase.... mmm tasty typos 2021-03-21 12:58:10 Ariadne_: ping for alpineconf - interested in helping out 2021-03-21 12:59:15 telmich: nice, welcome 2021-03-21 13:14:21 have you done alpienconf's befoer? 2021-03-21 13:14:25 before* 2021-03-21 13:17:10 I'm trying to mount btrfs partitions with compress/compress=lzo but it seems ignored 2021-03-21 13:20:42 donoban: this is our first 2021-03-21 13:21:44 hehe nice :) 2021-03-21 14:17:20 trying to curl a service running in an alpine container 2021-03-21 14:17:45 any idea how I can see the attempts and why they're being blocked on the destination machine? 2021-03-21 14:17:49 I see nothing in /var/log 2021-03-21 14:35:43 daft question, are the acl tools in Alpine nfs4 compliant? 2021-03-21 14:40:06 EF: there is http://git.linux-nfs.org/?p=bfields/nfs4-acl-tools.git;a=summary but doesn't appear to be Alpine-packaged 2021-03-21 14:41:43 minimal: thanks. I'm struggling with acls over a nfs4 share and it might be the lack of these that is the cause or (more likely) me being a numpty 2021-03-21 14:42:13 there's general info at https://wiki.linux-nfs.org/wiki/index.php/ACLs 2021-03-21 14:42:47 from that doc: "The kernel nfs client exposes ACLs on NFSv4 filesystems to userspace in the extended attribute named "system.nfs4_acl", which contains the raw xdr data which the client receives from the server as the value of the NFSv4 "acl" attribute." 2021-03-21 14:45:39 seems that if nfs-act-tools is packaged then you'd have nfs4_getfacl and nfs4_setfacl commands to play with 2021-03-21 14:46:25 I'll have a look at packaging that for Alpine, I'm already doing some work on the existing nfs-utils package currently 2021-03-21 14:48:44 That's what Void has. It seems, according to that article the server (Alpine) has POSIX ACLs and the nfs4_* tools on Void can read them correctly but coping data over (from Synology) doesn't respect the inherited ACLs! 2021-03-21 14:49:46 Maybe I'll copy them to Void and see if it removes the ACLs from Synology. Thanks for the pointers minimal 2021-03-21 14:50:42 EF: also saw this quote in a discussion: "and in addition NFSv4 ACLs are overriden by the umask. The umask is often 022 (remove write bit for group and other), which makes it impossible to inherit e. g. group write access. In NFSv3 it works properly." 2021-03-21 14:53:08 So maybe I should go to v3 then. umask is 0022 so that would certainly explain it with remote copying but locally it works. 2021-03-21 14:59:23 Not done much with ACLs in general. Am setting up NFSv4-only here at present, trying to avoid any v3 stuff at all 2021-03-21 15:54:32 I can't get compression enabled on btrfs :\ 2021-03-21 16:01:14 "acl tools" doesn't mean very much 2021-03-21 18:39:16 I started sddm and I can't select any session (I installed kde-applications) 2021-03-21 18:40:26 maybe I'm missing plasma 2021-03-21 18:44:57 uhM, it seems pretty broken in edge 2021-03-21 19:25:59 should I hope to run wayland? 2021-03-21 19:49:42 plasma and kde-applications is separate 2021-03-21 19:57:19 yes and plasma and plasma-desktop too 2021-03-21 19:58:21 but I feel that wayland problem is unrelated 2021-03-21 19:58:25 now I get a black screen 2021-03-21 19:58:38 donoban: what's your wayland problem? 2021-03-21 19:59:06 currently after enter my password on sddm I get a black screen, let me check logs... 2021-03-21 20:00:54 https://tpaste.us/0Eb1 2021-03-21 20:01:11 holy I did not use display manager for years 2021-03-21 20:01:43 and Xorg.. 2021-03-21 20:02:50 https://tpaste.us/ax9q 2021-03-21 20:05:45 https://tpaste.us/qg9w 2021-03-21 20:06:04 pulseaudio complains about dbus 2021-03-21 20:07:25 You'll need dbus, yes 2021-03-21 20:07:41 I enabled it and tried again 2021-03-21 20:07:44 but same black screen 2021-03-21 20:08:23 what about this? [20:07:21.477] (WW) HELPER: Failed to write utmpx: No such file or directory 2021-03-21 20:08:54 Don't worry about that one, it's only a warning and not fatal 2021-03-21 20:10:13 avoiding sddm, should startplasma-wayland just work? 2021-03-21 20:12:23 trying to run startx in the hope it calls startplasma-wayland ends with same error 2021-03-21 20:12:42 Could not start D-Bus. Can you call qdbus? 2021-03-21 20:28:11 omg 2021-03-21 20:28:25 I don't know what I did but it works now 2021-03-21 20:29:27 ahh no 2021-03-21 20:29:31 ;) 2021-03-21 20:29:42 ohh yes 2021-03-21 20:29:43 this is crazy 2021-03-21 20:29:56 I was in reallity logged on x11, but logout again and try with wayland 2021-03-21 20:30:08 and now it works! 2021-03-21 20:30:19 I think that installing firejail maybe added some needed dependency 2021-03-21 20:30:53 helby: is your IRL name “shelby”? 2021-03-21 20:36:40 halosghost: no :) I assume your u don't have name at all, as u are ghost :) 2021-03-21 20:36:55 helby: yes 2021-03-21 20:45:43 should I expect running netflix? 2021-03-21 20:47:36 I think if u have enabled DRM shit in your browser yes 2021-03-21 20:48:01 it complains about some plugin crashing 2021-03-21 20:48:44 maybe it needs to run something privative build for glibc 2021-03-21 20:48:59 but well, I could use a virtual machine for this 2021-03-21 20:49:11 if u just enabled that drm wait, it has install some shitty plugin 2021-03-21 20:49:12 testing youtube I have no sound 2021-03-21 20:49:25 or maybe try another browser 2021-03-21 20:49:58 it installed Widevine 2021-03-21 20:50:11 but crashes when I try to play something 2021-03-21 20:50:45 firefox? 2021-03-21 20:50:48 yes 2021-03-21 20:50:56 do u have also ffmpeg? 2021-03-21 20:51:06 uhM I don't know 2021-03-21 20:51:21 but I don't have sound too 2021-03-21 20:51:31 that ffmpeg can help 2021-03-21 20:52:14 it was already installed 2021-03-21 20:52:34 Does widevine work with musl? 2021-03-21 20:52:52 no 2021-03-21 20:53:04 that was I thinking 2021-03-21 20:54:21 so good news 2021-03-21 20:54:42 also to be honest ... pornhub has much better content 2021-03-21 20:55:00 :P 2021-03-21 20:56:00 helby: never looked there nor netflix but I trust you, because it sounds reasonable 2021-03-21 20:56:04 great! I rebooted and sound works :) 2021-03-21 20:56:29 mps: well u are lucky, u have dog and boyfriend, so 2021-03-21 22:52:33 hehe, hello from alpine-edge :P 2021-03-21 22:53:20 I'm trying to run firefox with podman 2021-03-21 23:04:43 well I will have to wait tomorrow for watch BoJack horseman here 2021-03-21 23:12:37 Greetings! I am turning down part of my business that used Rack/Power/Pipe at a Datacenter after 15years ... 2021-03-21 23:12:37 I see Linode is a AlpineLinux sponsor. I need resources to replace some components that are in the datacenter. 2021-03-21 23:12:37 How can I best let Linode know about new business because of the support of AlpineLinux? Is that valuable to do? 2021-03-21 23:14:25 what? 2021-03-21 23:14:40 (I'm honestly not sure I understand what you're asking) 2021-03-21 23:16:46 I need to create a new host and think it would be good to support Linode as they are an AlpineLinux sponsor. 2021-03-21 23:16:47 I am asking if there is a formal means to let Linode that net new business related to their support of Alpine? 2021-03-21 23:16:54 ipcloud[m]: tell to LBlaboon 2021-03-21 23:17:23 LBlaboon: ^ 2021-03-21 23:17:34 mps: Thanks! 2021-03-21 23:43:58 ahh 2021-03-21 23:44:00 gotcha 2021-03-22 02:29:39 ipcloud: glad to hear that! i'll be sure to pass that news along to our management 2021-03-22 07:00:05 hello 2021-03-22 07:02:18 q: I'm running apache on edge and after running `rc-update add apache2 boot` apache loads twice on boot. I am not sure why this happens, I've set other daemons to start on boot, but for some reason apache has an issue and tries to start itself a second time (and fails because the ports are already in use by the first apache process). Anyone seen this issue before? 2021-03-22 07:03:29 just a guess; any chance it's on boot and default at the same time? 2021-03-22 07:06:01 rc-update -v | grep apache2 2021-03-22 07:06:37 apache2 | boot 2021-03-22 07:06:41 only on boot 2021-03-22 07:07:13 I did a quick find /etc -iname "*apache*" just to make sure there are no stray symlinks or anything, but nothing came up 2021-03-22 07:08:17 The only other idea I have, which is extremely unlikely, but possible, apache2 hasn't been set to start via an @reboot line in crontab by chance? 2021-03-22 07:09:26 crontab? that would be strange, let me check please 2021-03-22 07:10:22 It would be strange, it's just an off the wall idea. :) 2021-03-22 07:10:28 nope, crontabs is empty apart from the default run-parts (15min, hourly, etc) and they are all empty as well 2021-03-22 07:11:26 ah bummer. Guess you'll have to wait and see if someone else can think of something, I'm out of ideas. Good luck 2021-03-22 07:11:27 when I run 'rc-update del apache2 boot' and reboot, it does not load at all 2021-03-22 07:11:43 zcrayfish, no problem, thanks for the ideas 2021-03-22 07:19:16 Hey, if there is any Let's Encrypt certbot users here, how do you manage auto renewing certificates ? Just with cron, or is there a more canonical way of doing that ? 2021-03-22 07:21:09 Personally I am using acme.sh with busybox cron 2021-03-22 07:21:19 my crontab entry looks like this: 2021-03-22 07:21:24 19 0 * * * "/root/.acme.sh"/acme.sh --cron --home "/root/.acme.sh" > /dev/null 2021-03-22 07:22:16 incidentally I don't think I am using the acme.sh apk, so you'll have to change paths likely if you go that route, but it gives you some ideas at least 2021-03-22 07:36:47 hum. Idk what are the advatages of each one, but I just did certbot renew and it messed with the nginx service.. 2021-03-22 08:27:10 So does anyone here use certbot ? 2021-03-22 08:29:36 I just use acme.sh 2021-03-22 08:32:57 certbot can be used in cron also, they recommend using the --noninteractive flag. 2021-03-22 11:46:48 hi all. I'm installing alpine to a Power9 system using the alpine-standard-3.13.2-ppc64le.iso image and have run into a bug when setting up raid1 using the standard 'setup-alpine' installer (by selecting two drives in sequence during the 'setup-disk' script execution). 2021-03-22 11:46:53 the error is: Installing for powerpc-ieee1275 platform. grub-install: error: ofpathname: not found. 2021-03-22 11:47:02 any advice would be appreciated 2021-03-22 11:48:01 ofpathname exists at: /usr/sbin/ofpathname so it's not an issue of the binary not being found 2021-03-22 11:51:20 and ldd /usr/bin/ofpathname? 2021-03-22 11:57:17 prior to running the installer the file is not found and ldd cannot locate the binary. after running setup-alpine but ctrl-c before the disks are chosen it's also not found. setup-disk probably installs the package at some point. I'll look at the source. after the installer completes (despite the error) ofpathname does exist, so there's a sequence problem somewhere. 2021-03-22 12:00:21 if I manually install the powerpc-utils and grub packages, which apk reports as providing ofpathname, then the binary is found in '/usr/sbin/ofpathname' but ldd reports: '/lib/ld-musl-powerpc64le.so.1: cannot load ofpathname: No such file or directory' 2021-03-22 12:01:29 whoops, didn't specify the full path. with full path ldd reports: '/lib/ld-musl-powerpc64le.so.1: /usr/sbin/ofpathname: Not a valid dynamic program' 2021-03-22 12:03:06 Oh, it's a script 2021-03-22 12:03:11 hmm. when running "ofpathname --help" it reports: 'ofpathname: is not supported on the PowerNV Host platform' 2021-03-22 12:03:54 yep, it's a bash script 2021-03-22 12:04:38 copyright IBM, very official ;) 2021-03-22 12:04:55 GPL license 2021-03-22 12:05:31 But I don't know what to do about that 2021-03-22 12:06:02 I think it means I need to use a hardware raid card instead of mdadm raid :| 2021-03-22 12:06:15 doesn't grub provide /usr/sbin/grub-ofpathname ? Same thing? 2021-03-22 12:06:17 problem is PCIe gen4 NVMe RAID cards are $$$ 2021-03-22 12:08:03 zcrayfish: possibly... I'll need to find where the setup script is specifying ofpathname and replace it with grub-ofpathname 2021-03-22 12:34:23 you can also partition manually 2021-03-22 15:35:32 mps: finally got some tiem to run your kernel 2021-03-22 15:35:33 https://paste.sr.ht/~bl4ckb0ne/cec1a4b80bcc3efbf1277738aabd9b90ba73e331 2021-03-22 15:36:12 fwiw the 5.11.2 upgrade was merged into pmos, gonna port it to 5.11.6 and send you linux-rockchip for review 2021-03-22 16:15:28 clandmeter, friendly greetings be unto you 2021-03-22 16:18:37 bl4ckb0ne: thanks for testing 2021-03-22 16:19:28 bl4ckb0ne: now I have something to make diff between working and non working configs 2021-03-22 16:19:43 nice 2021-03-22 16:20:36 I'll try to prepare linux-edge-5.11.8 later, and upload it 2021-03-22 16:20:50 will inform you when I finish 2021-03-22 16:23:01 thanks 2021-03-22 16:23:05 oh 5.11 is out 2021-03-22 16:23:08 .8* 2021-03-22 16:23:18 do you still want the linux-rockchip package though? 2021-03-22 16:25:39 bl4ckb0ne: no, I will require too much time to maintain separate flavors for separate SBCs 2021-03-22 16:25:45 alrighty then 2021-03-22 16:26:41 I'm enabling SBCs for users who are willing and ready to test (like you) and for those I can test 2021-03-22 16:26:52 same is fir armv7 kernels 2021-03-22 16:27:14 s/fir/for/ 2021-03-22 16:48:11 I'm experiencing a weird failure with name resolution, ping and wget both fail with 'bad address:' on sr.ht, but work for other stuff, even like google.ht 2021-03-22 16:48:40 haven't found any other domains that fail like that, so seemingly it's the only one failing 2021-03-22 16:48:57 (besides it's subdomains) 2021-03-22 16:56:39 ignaloidas▸ what happens if you use 'drill' to look up the record? 'drill sr.ht' does it change if you switch up your dns server? 'drill sr.ht @8.8.8.8' ? 2021-03-22 16:58:02 programmerq, works just fine 2021-03-22 16:58:05 both 8.8.8.8 and 1.1.1.1 give me seemingly correct responses for sr.ht with drill. ping and wget are also able to grab it. 2021-03-22 16:58:32 try slipping those into your /etc/resolv.conf and see if ping/wget do any better? 2021-03-22 16:58:35 both with default nameserver and with 8.8.8.8 2021-03-22 16:59:02 what version of alpine are you using? 2021-03-22 16:59:23 3.13.2 2021-03-22 16:59:42 but changing the nameserver to 8.8.8.8 fixed it 2021-03-22 16:59:45 so dunno 2021-03-22 16:59:54 thanks! 2021-03-22 17:00:50 Hello all, quick question on the alpine-extended iso, is there any reason why the python3 package has been dropped from that list of included packages? 2021-03-22 19:22:17 is there some meta package for installing qemu/kvm/libvirt? 2021-03-22 19:23:02 donoban: not that I'm aware of 2021-03-22 19:23:41 I installed a lot and still missing things :) 2021-03-22 19:26:32 virt-manager complains about missing 'qemu-kvm', should it be a symbolic link to qemu-x86_64? 2021-03-22 19:43:28 I get this error: "qemu: unknown option 'S'" maybe is a deprecated option 2021-03-22 19:44:00 In a debian host I checked the manpage and see: -S Do not start CPU at startup (you must type 'c' in the monitor). 2021-03-22 20:28:08 uhM it works now, I installed qemu-x86_64 but not qemu-system-x86 2021-03-22 21:23:10 Hello, 2021-03-22 21:23:10 I would like to contribute to a How-to on Wiki. 2021-03-22 21:23:11 I found out that that I can register easily and change content of Wiki pages. 2021-03-22 21:23:11 Is there a way how I could ask for a review? 2021-03-22 21:25:57 Not really, but you could edit the page and then ask if someone can take a look at it here 2021-03-22 21:28:26 oh please 2021-03-22 21:28:36 don't edit or add there anything 2021-03-22 21:29:17 it can just engourage somebody to keep it alive 2021-03-22 21:34:36 Could anybody review the Tomcat How To, please? 2021-03-22 21:34:36 https://wiki.alpinelinux.org/wiki/Tomcat 2021-03-22 21:34:37 Are there any typographic conventions, for example, how to write file and directory names, file content, commands, etc? 2021-03-23 00:14:47 I've run into this bug with fcron: https://gitlab.alpinelinux.org/alpine/aports/-/issues/12376 2021-03-23 00:15:24 A guy in the comments suggests running `usermod -g fcron -G "" fcron`, but fcron isn't installed on Alpine for me 2021-03-23 00:21:02 NEVERMIND. I needed to install shadow 2021-03-23 00:55:39 I just noticed that Alpine has pacman, xbps, rpm, dpkg, and maybe one or two more I'm missing. Is it a good idea to actually use any of those instead of apk? (I'm guessing not) 2021-03-23 00:56:03 (I'm guessing these are just for things like debootstrap or pacstrap) 2021-03-23 00:57:45 they can come handy when you want to extract stuff from a foreign package 2021-03-23 00:58:08 not that most of those come with musl compiled binaries. but anyhow sometimes useful 2021-03-23 00:58:19 s/most/any 2021-03-23 00:59:56 Void is optionally musl-based, so maybe xbps will work. Maybe 2021-03-23 01:03:08 why would somebody use xbps on alpine :? 2021-03-23 01:04:31 chroot 2021-03-23 01:04:32 Proxmox offers LXC image templates for alpine linux but not void linux. 2021-03-23 02:02:40 chroot 2021-03-23 02:02:40 probably for the same reason I use apk on arch 2021-03-23 02:11:39 sounds good, use apk on arch and pacman on alpine :/ 2021-03-23 02:11:52 that's the one reason all software suck 2021-03-23 02:14:36 I assume all is virtualized even on something else :/ 2021-03-23 09:16:29 <[diablo]> Good morning #alpine-linux ... hope you're all well. Just curious, does anyone know if Podman https://podman.io is heading to Alpine? 2021-03-23 09:17:27 We have podman in edge 2021-03-23 09:45:58 <[diablo]> hi Cogitri ohh 2021-03-23 09:46:32 <[diablo]> Cogitri atm I use Docker from community 2021-03-23 09:47:10 Podman is in testing 2021-03-23 10:08:30 <[diablo]> good morning ikke 2021-03-23 10:08:32 <[diablo]> ah cool 2021-03-23 10:56:42 update on the Power9/ppc64le software raid failure, it seems to not have anything to do with the error from grub-install complaining about 'error: ofpathname' and related 'not supported on the PowerNV Host platform' situation, as I went back and installed to a single nvme drive without mdraid and the system had the same grub-install error but rebooting brought up a working system. when using 2021-03-23 10:56:43 mdraid and rebooting the error was originally: 2021-03-23 10:56:46 'mount: mounting UUID=c7349bf1-cc4e-4be9-b4bb-38412ff4d7ad on /sysroot failed: No such file or directory Mounting root failed.' 2021-03-23 10:57:53 so it seems there's a bootloader issue when using mdraid on alpine. for comparison, fedora 33 server will setup mdraid but IIRC it uses systemd-boot 2021-03-23 10:58:28 in any case, I've ordered a hardware raid card and will use that instead. 2021-03-23 10:59:35 so if anyone ends up searching freenode logs for ppc64le and alpine, that's the current status 2021-03-23 11:14:19 well what to expect from bootloader bigger than whole OS 2021-03-23 11:16:47 yep. systemd is a travesty. openrc is one of several preferred features of alpine imo. 2021-03-23 11:18:17 I still have to use systemd at work on x86_64 prod systems but it doesn't have a place on any of my laptops or home lab systems. 2021-03-23 11:19:23 I was talking about grub :/ 2021-03-23 11:19:41 it's his little brother 2021-03-23 11:20:09 systemd however crap it is, at least it works 2021-03-23 12:18:52 anyone has experience with auditd in alpine? I would like to monitor file access on some directories, maybe there are better alternatives 2021-03-23 14:00:04 donoban: inotifywatch from inotify-tools package? 2021-03-23 14:07:36 uhM could be an option, I'm testing auditd it seems pretty simple but I read some complains about performance 2021-03-23 14:09:35 donoban: have auditd running here but not done much with its config yet (got sidelined looking into apparmor) 2021-03-23 14:10:25 have you looked at auditbeat? (part of the Elastic Beats family). It hooks into the kernel audit stuff separately from auditd 2021-03-23 14:10:54 doh, forgot auditbeat is not currently packaged for Alpine 2021-03-23 14:14:02 need to look again at sorting the beats packaging - filebeat and metricbeat are being packaged separately when they (plus auditbeat) should be creating from a single APKBUILD 2021-03-23 14:32:43 donoban: I think any auditd performance issues relate to how complex your auditd config is 2021-03-23 15:47:42 minimal: yes and I read that the performance penalties is more related to syscall's rules than filesystem watchers 2021-03-23 15:49:14 donoban: the filesystem watcher stuff hooks into kernel inotify so if its just that you need then inotify-tools might be a simpler solution 2021-03-23 15:50:30 ok I will take a look, with audit is pretty simple, create audit.rules, add "-w /path/ -p rwx -k ruleKeyName", and reload auditd service 2021-03-23 15:51:22 this talk has reminded me to start work again on sorting out a beats package (and so packaging auditbeat) :-) 2021-03-23 15:59:49 hehe 2021-03-23 16:00:21 is there a std way for set some hooks on apk? I would like to create btrfs snapshots before upgrade/install 2021-03-23 16:05:35 audibeat looks great :D 2021-03-23 16:18:15 Hello! I have decided to use Alpine Linux for as an OS solution for my server, but am having difficulties installing it in a way desired by me. My goal is to have a xen dom0 that is set up with LVM on LUKS and has Xfce4 DE, but the documentation on my specific desired setup is not very certain, as I am unsure in which order the steps described in respective guides for xfce4 and xen dom0 are supposed to be performed in relation to LVM on LUKS 2021-03-23 16:18:15 guide. Could anyone please clarify that? Thanks in advance. 2021-03-23 16:21:56 well, first focus on doing a sys install on LUKS/LVM and then install xfce 2021-03-23 16:27:08 Well, let us suppose, I finished installing Alpine Linux following steps described in LVM on LUKS guide, what next? I need more specifics, if that's possible 2021-03-23 16:27:57 And, am I supposed to run all the install scripts (for both xorg and xen) while still chrooted into /mnt/ or not? 2021-03-23 16:28:47 uhM, once you have the install done and the bootloader installed, you should reboot and continue with your new install 2021-03-23 16:29:26 maybe you have problems booting it with xen but I suppose that the install scripts should be aware of it 2021-03-23 16:29:53 are you using syslinux or uefi? 2021-03-23 16:33:28 >are you using syslinux or uefi? UEFI 2021-03-23 16:34:41 But I have no problems rebooting to the new install if xen is not installed and it is just LVM on LUKS 2021-03-23 16:35:29 But during my last attempt at installing all of the aforementioned components (basically just running setup-xenp 2021-03-23 16:35:36 I'm not sure of how xen uefi/grub should look 2021-03-23 16:35:48 A sec 2021-03-23 16:36:00 but welll if you don't have problem try rebooting and check what happens 2021-03-23 16:36:37 or take a look in your grub.cfg 2021-03-23 16:36:38 But during my last attempt at installing all of the aforementioned components (basically just running setup-xen-dom0 at the very beginning of the install and setup-xorg-base after, outside of chroot), I fail to boot into it 2021-03-23 16:38:25 The question is, should I follow all the steps mentioned in xen dom0 guide outside of inside of chroot? 2021-03-23 16:40:46 sorry I'm not sure, I haven't run alpine/xen in years, I guess setup-xen-dom0 should be run outside the chroot (I suppose that there is nothing for chroot before) 2021-03-23 16:41:53 then it says about adding xen-hypervisor and bridge, this should have to be done inside the chroot but I'm not sure 2021-03-23 16:42:15 maybe I can help if you share some error log about your failed boot 2021-03-23 18:37:43 hi everbody, has somebody ported alpine linux to pinephone? 2021-03-23 18:38:11 postmarketos? 2021-03-23 18:39:19 postmarketos is Alpine based, alpine runs with a bit of effort on the pine boards though 2021-03-23 18:39:20 :), i didnt know that build on alpine linux 2021-03-23 18:40:06 i will try that out, i use at the moment archlinux but with gnome :( 2021-03-23 18:40:50 gnome is not my favorite, its to heavy, slow, bloated 2021-03-23 18:42:33 thanks 2021-03-23 18:44:32 do we know where all the netdata modules are that are normally shipped with stock netdata? latest apk doesn't have them 2021-03-23 18:49:09 https://git.alpinelinux.org/aports/tree/community/netdata/APKBUILD is missing the go stuff 2021-03-23 21:37:31 hey someone that confirms if from APKBUILD the _flavor=${FLAVOR:-lts} retrieve the kernel current build dir version? 2021-03-23 21:37:49 It's an environment variable 2021-03-23 21:37:58 if it's not set, it defaults to lts 2021-03-23 21:38:00 -lts 2021-03-23 21:38:19 so hoy can i retreive the kernel curren from CI bvuild? 2021-03-23 21:38:58 there is no 'current' kernel flavor 2021-03-23 21:39:15 holy sh*t .. puff arggaaaaa 2021-03-23 21:39:31 again a curse to find a way.. o9k thanks ikke 2021-03-23 21:44:45 PICCORO: there is 'linux-edge' package with slightly changed kernel config but in edge repo 2021-03-23 21:45:11 MY-R: PICCORO is trying to build virtualbox 2021-03-23 21:45:14 with kernel modules 2021-03-23 21:45:16 MY-R: i am isn edge with sodomon trying to build virtualbox but is a pain with that limited package system 2021-03-23 21:45:26 ah ok 2021-03-23 21:45:35 i alrteady páss <--disable-kmods \ but is stil building.. i cannot understand why 2021-03-23 21:45:48 i am tired of tha apkbuild limitations 2021-03-23 21:45:51 PICCORO: your limited understanding of the package system* :) 2021-03-23 21:45:59 ^ :P 2021-03-23 21:46:12 puff.. i package more complicated thing in alpine.. but the kernel touched is very limited 2021-03-23 21:46:37 how can you MY-R explain that i passed --disable-kmods and still build the modules! puff 2021-03-23 21:46:46 We manage to build a distro with it 2021-03-23 21:46:55 uh ikke ? i pass --disable-kmods and does still build that mods.. 2021-03-23 21:47:23 averuybody can build a distro.. but noty everybody can make a good api and CI env 2021-03-23 21:49:12 I can't build a distro 2021-03-23 21:55:14 PICCORO: I can give you only one hint: first get it to work locally before you try to getting it to build in the CI system 2021-03-23 21:55:36 locally works and compiles.. we pusblished the picture with virtualbox working .. 2021-03-23 21:56:03 ikke: locally load and works flameless of course.. cos is not a limited not configured CI environment 2021-03-23 21:56:34 https://t.me/s/alpine_linux/322 2021-03-23 21:57:27 [diablo]: I use Podman from testing on my Alpine host. If you have any questions about it let me know. 2021-03-23 23:30:33 are there good instructions on how one might run the Alpine image for Raspberry Pi on QEMU? I have searched a bit online but gotten mostly conflicting/vague answers 2021-03-24 00:21:49 So I just installed Alpine on a vintage-2014 Chromebook. And I'm loving it so far. My only concern is that I'm not sure how to get my Yubikey to work with Firefox. The Arch Wiki says to install libfido2. What is the equivalent Alpine package? 2021-03-24 01:06:46 I do not think there is one, due to the build deps. 2021-03-24 01:06:58 "On Linux, libudev (part of systemd) is also required." 2021-03-24 03:00:25 i assume it could work with eudev 2021-03-24 03:00:42 if we had it packaged and firefox was built against it 2021-03-24 03:24:44 is there any way to get the contents of a package (as in `apk info -L `) without installing it? 2021-03-24 05:31:22 outerpassage: not through apk 2021-03-24 05:31:53 You can with https://pkgs.alpinelinux.org/contents 2021-03-24 05:32:23 there is an unofficial cli tool called apk-file as well 2021-03-24 06:19:36 ikke: thanks! 2021-03-24 07:30:13 <[diablo]> xordspar0 morning, sorry I've just noticed your comment. I was really just curious about Podman, I see quite a lot of hype from RH.... 2021-03-24 08:18:56 Hi. Coming from other distros, i used to use firewalld (and firewall-cmd) to setup a simple firewall for a workstation or server. What do you guys use in alpine ? bare nftables ? i started with a simple nft file, and `rc-service nftables save`, works for me. But i was wondering, since it's a very common usecase if there was a prefered way. I've seen Alpine Wall in the wiki. Is that still maintained ? The page 2021-03-24 08:18:56 last update is from 2017. 2021-03-24 08:21:47 Yes, a wall is still maintained 2021-03-24 08:22:08 awall uses iptables? 2021-03-24 08:22:12 The page, I'm not sure 2021-03-24 08:22:14 Yes 2021-03-24 08:22:54 I use ferm because of familiar syntax (perl) to me 2021-03-24 08:31:13 is the mechanism by which abuild separates built-time dependencies in an APKBUILD `makedepends` exposed anywhere? would be cool to be able to enter a shell with some development packages installed, for example, without modifying /etc/apk/world 2021-03-24 08:31:41 build-time* 2021-03-24 11:16:51 I'm experimenting frequent segfaults with plasma/wayland altought I don't know how to reproduce them 2021-03-24 11:16:54 [ 2641.695792] kwin_wayland[3896]: segfault at 7f9964298020 ip 00007f99f525ea7e sp 00007ffcd0a80d30 error 4 in libKWaylandServer.so.5.21.3[7f99f520b000+88000] 2021-03-24 11:17:08 is tis very hard to debug? :S 2021-03-24 11:26:23 what can go wrong with 1GB suite 2021-03-24 11:28:07 hehe 2021-03-24 11:28:16 I'm just testing my new future desktop 2021-03-24 11:28:45 OK: 4447 MiB in 1059 packages :D 2021-03-24 11:29:40 the kernel segfault message is notoriously useless 2021-03-24 11:30:04 turn on core dumps and wait for it to happen again 2021-03-24 11:30:20 uhM ok 2021-03-24 11:35:39 Hello71: I never did this before, should this be enough? 2021-03-24 11:35:44 medion:/usr/libexec$ ulimit -c 2021-03-24 11:35:46 unlimited 2021-03-24 11:42:04 Hello71: segfault msg is not debugging 'tool', just to notice bug 2021-03-24 11:47:08 just finished setting up Mate desktop on a thinkpad x1 and installed Zoom via Flatpak, which loads and doesn't throw any errors but the login screen doesn't allow clicking "join meeting" or "login". has anyone run into this before and know how to fix it? 2021-03-24 11:59:56 donoban: should be 2021-03-24 12:00:02 depends how you start plasma 2021-03-24 12:00:07 you need to set before starting 2021-03-24 12:01:42 ah, so I will reboot 2021-03-24 12:53:12 ncopa, ikke how can i removed the hardened compilations!? it seems are by default and i so have TEXTREL in my shared libs.. we dont need hardened in non networking things! 2021-03-24 12:54:57 -fno-PIC(CORO), jk 2021-03-24 12:55:44 mps is that all i must do? .. virtualbox needs non hardening cos i have TEXTREL problems.. reloacions with absolute refs 2021-03-24 12:56:52 no-pie and/or no-pic maybe can help, not sure because I forgot. last time I fixed textrel was about year or two ago 2021-03-24 12:58:26 puff mps that it does not help so much arrgggg XD 2021-03-24 12:59:24 read this https://wiki.gentoo.org/wiki/Hardened/Textrels_Guide 2021-03-24 12:59:38 i already read XD 2021-03-24 13:00:43 it does not help neither mps cos due performance issues we nedd to diasble all hardened compilations in virtualbox.. virtualbox is not server production software 2021-03-24 13:00:50 hmm, better option is 'don't use virtualbox' 2021-03-24 13:01:09 yeah.. mps you have right.. but i in a compromise with sodiomon packager XD 2021-03-24 13:01:19 we have qemu that works flameless 2021-03-24 13:01:22 "use qemu, Luke" :) 2021-03-24 13:01:30 XD XD XD XD 2021-03-24 13:02:21 don't understand why people likes to waste time on bad things 2021-03-24 13:08:42 Hello... 2021-03-24 13:09:36 Is there any reasons ntop package has been dropped from Alpine in 3.12 ? It was available in 'main' in 3.11 ... Is there a similar alternative? 2021-03-24 13:09:52 whattt? .-. no ntop package? 2021-03-24 13:10:39 umm what about ntoppng danci1973 ? 2021-03-24 13:10:56 ntopng 2021-03-24 13:13:12 No, nothing that starts with 'ntop'. Currently looking in 3.13 main / community and edge main / communtiy repos. 2021-03-24 13:13:48 pufff.. nonsense.. i always said.. i'm tyre.. err wait.. maybe we need a packager.. let me take into that! 2021-03-24 13:14:04 mps what happened to notp package? 2021-03-24 13:14:14 Check git 2021-03-24 13:14:43 git log --grep ntop 2021-03-24 13:14:51 git is a sea orf directoris.. very hard to browse.. any usefull link ikke to property make a collaboration cos SEEMS IS NEED!? 2021-03-24 13:14:56 I didn't followed ntop, but if it is removed that is good move 2021-03-24 13:15:11 i not will checkout a thousand of gigas just to check a log ikke 2021-03-24 13:15:37 aports git log is about 350MB 2021-03-24 13:15:51 git repo* 2021-03-24 13:16:08 mps yeah i have limi9ted connection.. umm ntop removed as good move.. there's a more better soluition? please help right? 2021-03-24 13:17:30 unmaintained/ntop: move from main 2021-03-24 13:17:48 but is the package the problem or the lack of mantainer? 2021-03-24 13:18:10 testing/ntopng: move to unmaintained 2021-03-24 13:18:26 7c6239b112238719c308c1640f9cf50fcf44a148 2021-03-24 13:18:29 mps is better to use ntopng is that the reason? or lack of manpower/mantainer? 2021-03-24 13:18:37 checking link 2021-03-24 13:18:40 mps is better to use ntopng is that the reason? or lack of manpower/mantainer? 2021-03-24 13:19:06 as I told I don't know 2021-03-24 13:19:23 ok thanks.. i will follow and do my task.. 2021-03-24 13:19:24 but I stopped to use ntop long ago 2021-03-24 13:19:44 it was in 'not good' shape long ago 2021-03-24 13:19:46 umm just by curiosoty mps.. why? illuminate me! 2021-03-24 13:20:32 try to build from source and use it, you will know 2021-03-24 13:20:41 mps XD XD XD XD 2021-03-24 13:21:03 and I'm zen master, so my teachings are not conventional :P 2021-03-24 13:22:07 oh checked debian patches.. and libc is a pain.. now understand.. there are better solutions.. 2021-03-24 13:22:56 for teaching I like to use keisaku stick ;) 2021-03-24 13:23:26 mps XD XD XD CAUCUAUACUA 2021-03-24 13:23:55 not so many people know that is that wood rule! 2021-03-24 13:24:34 I don't know what is 'CAUCUAUACUA' 2021-03-24 13:37:43 yay! specing! 2021-03-24 14:09:00 Hey is anyone here using certbot for Let's Encrypt ? 2021-03-24 14:21:33 I'm still receiving emails after having renewed the certificates. Everything seems okay except for that... 2021-03-24 14:22:03 did you restart server 2021-03-24 14:22:59 danci1973: iirc ntop is unmaintained and ntopng is super fat and not great license 2021-03-24 14:24:18 someone need to create ntopngng, to infinitum 2021-03-24 14:24:30 Yeah, I've just been looking at ntop.org and came to the same conclusion... 2021-03-24 14:24:47 Hello71: Are you answering to me or not at all ? 2021-03-24 14:25:06 yes 2021-03-24 14:25:07 TBH, it's not an essential package... 2021-03-24 14:25:31 ntop seems not really in "alpine spirit" 2021-03-24 14:25:44 not sure what is better though 2021-03-24 14:32:52 danci1973: Depends on how detailed traffic breakdown you want. There is plethora of simple tools (eg. darkstat, munin, etc.) that just read interface counters and there is a tooling to do more detailed statistics eg. via connection tracking subsystem but nothing that works "out of the box" without extensive setup like ntop did AFAIK. 2021-03-24 14:33:19 But then, I haven't looked for an alternative for a while. 2021-03-24 14:34:57 CcxWrk: That's true, of course, but I meant 'essential' on my new firewall. I'm updating from a relatively old Alpine and I had 'ntop' on it, so I wanted to install it on the new one as well. But to be honest, I don't remember ever using it there. 2021-03-24 14:37:37 I think darkstat and smokeping about cover the amount of statistics I want exposed over web interface for basic "does it work" debugging, for the rest I'd drop down to terminal. 2021-03-24 14:47:51 Hey, I wrote a little guide on getting Alpine on my APU boards, https://blog.markpash.me/posts/alpine-apu-iso/ 2021-03-24 14:48:02 I'm wondering if there is a way to make it easier though 2021-03-24 14:49:25 Is there any way to customise the extlinux config that gets installed by the ISO? So I could master an ISO that enables serial console during setup-alpine 2021-03-24 14:52:29 markpash: yes, add console=ttyS0 console=tty0 to kernel command line 2021-03-24 14:52:56 that should be default on install media 2021-03-24 14:54:03 this reminds me there is bug/wish report assigned to me 2021-03-24 14:54:50 mps: Right now it's really manual. So I make the ISO with the serial console settings so that I get output for the install media, but after running setup-alpine I have to manually mount the rootfs and go and tweak files for it to continue after reboot. 2021-03-24 14:55:32 Otherwise I lose serial console, because the default extlinux config that gets installed to the rootfs doesn't enable that by default. 2021-03-24 14:56:29 What I'd like to be possible is, making an install media that automatically puts a modified extlinux config on the rootfs, so I don't have to. 2021-03-24 15:02:13 Ideally you'd want full system installation and configuration script suite for that. I've made one for myself but it's not really Alpine centered. 2021-03-24 15:05:33 PICCORO: thin -fno-pie is the proper way 2021-03-24 15:06:19 oh well he lefrt 2021-03-24 15:09:06 markpash: I'm not 100% sure what would be right solution, but I think setup-alpine should detect tty on which it run and set that one in extlinux.conf or grub.cfg 2021-03-24 15:12:12 mps: yeah, I don't know either, I'm happy to help come up with a solution if it's desireable for others. 2021-03-24 15:13:34 maybe use /proc/consoles and set them all dynamically 2021-03-24 16:08:27 I think it should already copy args to /etc/update-extlinux.conf 2021-03-24 16:08:40 which is included by 10_linux 2021-03-24 16:16:57 Is there a package for posix manpages? 2021-03-24 16:26:37 CcxWrk: not aware of any 2021-03-24 16:27:26 :( 2021-03-24 16:42:56 Can I just say how awesome alpine-setup is? I don't know what kind of magic it uses, but the WiFi setup part of the script is the best command line implementation I've seen for any installer. 2021-03-24 16:49:01 CcxWrk: i think `man-pages` includes posix manpages 2021-03-24 16:49:06 not sure how to open them though 2021-03-24 16:49:41 they are under man0? 2021-03-24 16:52:10 Yeah I tried man 1p ls and it doesn't work, but it's there somewhere 2021-03-24 16:52:34 So I guess just the categories are wrong? 2021-03-24 16:52:54 should be 3p 2021-03-24 16:53:14 this seems to work: man /usr/share/man/man3/fopen.3p.gz 2021-03-24 16:53:31 i kinda expected fopen.3p to work, but it doesnt 2021-03-24 16:53:40 3p is libc functions, 1p standard commands 2021-03-24 16:55:52 man -l /usr/share/man/man3/posix_madvise.3p.gz, example 2021-03-24 16:55:53 `man ls` giges me the 1p 2021-03-24 16:56:05 s/giges/gives/ 2021-03-24 16:56:13 man /usr/share/man/man1/ls.1p.gz 2021-03-24 16:56:23 same for man read 2021-03-24 16:56:29 gives 3p 2021-03-24 16:57:26 for the syscall, use `man 2 read` 2021-03-24 16:58:27 it should be fixed, maybe put them in /usr/share/man/man3p/ dir 2021-03-24 16:59:54 and there are two of versions, /usr/share/man/man3/posix_madvise.3p.gz and man -l /usr/share/man/man3/posix_madvise.3.gz 2021-03-24 17:00:11 and there are two of versions, /usr/share/man/man3/posix_madvise.3p.gz and /usr/share/man/man3/posix_madvise.3.gz 2021-03-24 17:00:17 Well, read is a shell builtin as well as libc function as well as kernel syscall. Plus you can have POSIX specification and actual implementation documentation (with vendor extensions documented) installed. 2021-03-24 17:01:22 posix_madvise.3p.gz is posix man page but /usr/share/man/man3/posix_madvise.3.gz is a linux programmers manual man page 2021-03-24 17:05:09 xaltsc: Let's Encrypt emails you if a certificate that it knows about is about to expire. It doesn't check you server configuration. You should look at your server and check if it's working correctly. For example, navigate to it in a web browser and check the cert expiry date. If that's fine then it's safe to ignore the email. 2021-03-24 17:06:40 Sometimes you will get an email for an old cert that you don't care about. If you add or remove a domain, it will issue a new cert and let the old one expire. Anyway, the email is just a reminder. Check the actual cert that your clients get to know for sure. 2021-03-24 17:31:47 bl4ckb0ne: I upgraded linux-edge to 5.10.9 and enabled DWC MMC drivers in-kernel. Would be nice if you have time to test it and report, maybe we made it to work on your SBC 2021-03-24 17:32:02 to 5.11.9* 2021-03-24 17:33:12 is it in the repos? 2021-03-24 17:33:39 https://pkgs.alpinelinux.org/package/edge/testing/x86_64/linux-edge 2021-03-24 17:33:54 it is 2021-03-24 17:34:01 gonna give it a whirl when my little one is asleep 2021-03-24 17:35:15 heck the uboot config is still not upstream 2021-03-24 17:35:33 xordspar0: thanks, so should I unsubscribe to the list, or will it prevent me from getting new exipry mails for the new certificate ? 2021-03-24 17:36:24 If you replace the certificate quick enough, you should not receive those reminders 2021-03-24 17:37:32 xaltsc: It's normally useful as a backup for when the cert renewal fails for some reason and your own monitoring doesn't catch it. Normally you won't get those emails when things are working correctly. 2021-03-24 17:37:57 ikke: I hope so, I've set up a monthly cronjob now, but just in case something doesn't work, I want to receive emails 2021-03-24 17:38:45 xordspar0: my question was more something like: does the list I'm subscribed to corresponds to one acme account or not 2021-03-24 17:38:46 Yes, that should still happen 2021-03-24 17:38:48 bl4ckb0ne: yes, u-boot support is not yet merged 2021-03-24 17:39:02 xaltsc: these alerts are not related to a subscription to a list 2021-03-24 17:39:24 I had a hope that it will be in next u-boot stable, but now my hope vanishing 2021-03-24 17:39:27 they are sent to the e-mail address belonging to the account that registered the domain 2021-03-24 17:41:25 ikke: They are ? There are list headers in the email and they say I can unsubscribe. Or were you talking about non-expiry related emails ? 2021-03-24 17:42:56 xaltsc: ah, like that. 2021-03-24 17:43:12 xaltsc: it makes sense you can unsubscribe from them 2021-03-24 17:44:33 But it's not a mailing list in the traditional sense 2021-03-24 17:45:43 well, yes, I'm the only one being sent these mails ofc :D 2021-03-24 17:46:15 But if you unsubscribe, you would not get any mails in the future 2021-03-24 19:54:27 mps: fails to boot 2021-03-24 19:55:10 https://paste.sr.ht/~bl4ckb0ne/7b2ef2c9f713284ee5c75bb9c2da9d62351cd477 2021-03-24 19:57:26 bl4ckb0ne: why do you have this in kernel command line 'init=' 2021-03-24 19:57:40 leftover from last try 2021-03-24 19:57:44 do you want me to run it without? 2021-03-24 19:57:51 mmm... that's odd. 2021-03-24 19:58:05 "[ .6dwmmc_rockchip fe320000.mmc: DW MMC controller at irq 33,32 bit host data width,256 deep fifo" 2021-03-24 20:00:46 bl4ckb0ne: successful cmdline was 'root=/dev/mmcblk1p2 rootwait rootfstype=ext4 console=ttyS2,1500000 loglevel=7' 2021-03-24 20:01:13 RootWyrm: probably web formating 2021-03-24 20:01:26 gonna try without 2021-03-24 20:01:52 no, it looks like it's trying to init the mmc twice 2021-03-24 20:02:54 this one log line looks strange 'mount: mounting /dev/mmcblk1p2 on /sysroot' 2021-03-24 20:03:04 same error mps 2021-03-24 20:03:24 it shouldn't appear when not using initramfs 2021-03-24 20:03:27 nah, that's not particularly interesting... it's not initializing the mmc correctly though 2021-03-24 20:03:43 oh 2021-03-24 20:03:57 bl4ckb0ne: remove INITRD from u-boot config 2021-03-24 20:04:08 from edge? 2021-03-24 20:04:12 yes 2021-03-24 20:04:30 bl4ckb0ne: this is sercons or hdmi? 2021-03-24 20:04:52 RootWyrm: https://wiki.kobol.io/helios64/hardware/ that 2021-03-24 20:05:34 RootWyrm: serial console 2021-03-24 20:05:47 well i figured it was a rock... hmm... that's... oh, might be. 2021-03-24 20:05:59 helios64 SBC 2021-03-24 20:06:20 hangs on `[ 0.264857] Waiting for root device /dev/mmcblk1p2...` 2021-03-24 20:06:22 hardware init's going wrong 2021-03-24 20:06:41 bl4ckb0ne: could you post this new boot log 2021-03-24 20:06:47 sure 2021-03-24 20:06:49 without initramfs 2021-03-24 20:07:09 https://paste.sr.ht/~bl4ckb0ne/63e767ff0915458c756e3b71f657377637b65496 2021-03-24 20:08:22 thanks 2021-03-24 20:08:42 (holly $whatever what is missing?) 2021-03-24 20:08:45 won't make any difference... it's not initiailizing the mmc controller correctly 2021-03-24 20:08:49 it's a kernel issue 2021-03-24 20:09:19 ive had kernels starting with success on that thing 2021-03-24 20:09:25 RootWyrm: yes, we know that, but don't know which driver must be built in-kernel and not as module 2021-03-24 20:09:34 bl4ckb0ne: yeah, hence my 'wtf'. it's just an rk3399. 2021-03-24 20:10:52 is this trying to boot mmc or sd card? 2021-03-24 20:10:53 I cannot annoy bl4ckb0ne for every idea/module to test it 2021-03-24 20:11:06 sd card 2021-03-24 20:11:20 ah 2021-03-24 20:11:24 not mmc? 2021-03-24 20:11:29 isnt it the same? 2021-03-24 20:11:37 no, sd card is connected to the mmc controller 2021-03-24 20:12:01 is there an mmc attached? wonder if it's inconsistent on the device number. 2021-03-24 20:12:07 I can't remember for this SBC, maybe sd card is connected over other interface than mmc 2021-03-24 20:12:29 nah, it's connected via mmc, has to be on those for boot. 2021-03-24 20:12:44 fwiw the config we need is prolly inthere https://gitlab.com/postmarketOS/pmaports/-/blob/master/main/linux-postmarketos-rockchip/config-postmarketos-rockchip.aarch64 2021-03-24 20:13:01 rk3399 have three different mmc interfaces 2021-03-24 20:13:32 yes, but to be a boot candidate has to be attached to mmc (unless they pulled a realtrash.) 2021-03-24 20:13:45 at least outside of some ... fun stuff. 2021-03-24 20:13:50 right 2021-03-24 20:16:14 well i can't imagine any of the kernel config is wildly different.. i think it's something in first stage 2021-03-24 20:17:43 RootWyrm: aarc64_defconfig works fine 2021-03-24 20:18:16 that's why i'm thinking first stage. 2021-03-24 20:18:32 RootWyrm: here is the loog https://paste.sr.ht/~bl4ckb0ne/98ff29256eb50cfc0515ae3ff8eafd7b808723a2 2021-03-24 20:18:56 maybe you spot problematic point 2021-03-24 20:19:03 you can* 2021-03-24 20:19:09 oh wait... yeah... i see something. 2021-03-24 20:19:14 becaue I have to to go out 2021-03-24 20:19:31 dma-pl330 2021-03-24 20:20:59 could be, will check later on my rk3399 2021-03-24 20:22:29 missing dma-pl330 and i don't see the CQHCI either 2021-03-24 20:23:49 so yeah, that would keep it from properly initializing the mmc, which would result in the mount hang 2021-03-24 20:24:27 ah, yes, my rk3399 have dma-pl330 ff6d0000.dma-controller: Loaded driver for PL330 DMAC-241330 2021-03-24 20:24:34 https://git.alpinelinux.org/aports/tree/testing/linux-edge/config-edge.aarch64#n7249 could be worth trying 2021-03-24 20:25:04 could try with loading the module in the kernel cmdline 2021-03-24 20:25:19 bl4ckb0ne: I will build new one in about 2 hours 2021-03-24 20:25:28 ack 2021-03-24 20:26:35 yep. CONFIG_PL330_DMA=m, not y 2021-03-24 20:26:53 bl4ckb0ne: don't hesitate to ask me whatever you want with this, I'm really happy if I can help you 2021-03-24 20:27:07 :) 2021-03-24 20:27:31 ikke: bl4ckb0ne is very nice and kind 2021-03-24 20:27:42 im still jobless so I have all the time I want between my kid's naps 2021-03-24 20:28:02 bl4ckb0ne: so yeah, missing module, question is if it's *just* pl330_dma or if there's more 2021-03-24 20:28:19 what's the cmdline to load the module again? `module=pl330`? 2021-03-24 20:28:23 or is it modules 2021-03-24 20:28:29 bl4ckb0ne: oh, sorry for your job status 2021-03-24 20:28:45 dont worry, i got some good leads 2021-03-24 20:30:26 tbh, i can't recall myself. <.< 2021-03-24 20:31:57 trying modules=pl330 2021-03-24 20:32:27 I doubt that this module is in initramfs 2021-03-24 20:32:38 neh 2021-03-24 20:33:50 yeah its not in initramfs 2021-03-24 20:37:16 building new kernel 2021-03-24 20:41:14 my *hope* is that it's just pl330's missing. lacking a dma controller driver would definitely impede dma. 2021-03-24 20:42:13 we've done pretty much everything for the mmc controller config already, thanks for the new look on the issue RootWyrm 2021-03-24 20:42:55 sure thing... us driver people are a rare breed (for a reason) 2021-03-24 20:43:48 so how can a dma driver impact mmc stuff? 2021-03-24 20:45:10 mmc is all dma operation 2021-03-24 20:45:30 makes sense 2021-03-24 22:28:16 I had another crash but I'm not sure where I should have the core dump 2021-03-24 22:30:21 You could install corecollector, that manages the coredumps for you 2021-03-24 22:30:40 oh, let's try, thanks 2021-03-24 22:30:45 Otherwise it's in the CWD of the program 2021-03-24 22:31:14 bl4ckb0ne: https://dev.alpinelinux.org/~mps/linux-edge-5.11.9-r1.apk 2021-03-24 22:31:24 After the install of corecollector you'll have to either restart or reload your sysctl values btw 2021-03-24 22:31:52 bl4ckb0ne: when you find time 2021-03-24 22:33:56 https://tpaste.us/5Vgy 2021-03-24 22:34:32 is the kernel.core_pattern error expected? 2021-03-24 22:34:38 Ah yes, you'll probably need a reboot for that sysctl value to be set 2021-03-24 22:34:55 ah 2021-03-24 22:35:43 well, is probably that I've already had plasma running before setting ulimit 2021-03-24 22:36:09 It seems that it ignores /etc/security/limits.conf 2021-03-24 22:36:29 I reboot but 'ulimit -c' returns 0 2021-03-24 22:46:11 I will try to get it tomorrow 2021-03-24 22:46:34 u can always uninstall whole plasma fuckery 2021-03-24 22:46:46 haha yes 2021-03-24 22:47:10 but I'm happy with it 2021-03-24 22:48:17 what dekstop do you suggest helby? 2021-03-24 22:49:47 I don't any, I use window manager 2021-03-24 22:52:37 hehe, well at least I would like to find the crash cause 2021-03-24 23:05:20 it will not help you 2021-03-24 23:05:41 whatever u will remove it will remove 500 own dependencies and u will end just with calculator 2021-03-24 23:26:35 well I meant the code bug 2021-03-25 00:43:12 mps RootWyrm https://paste.sr.ht/~bl4ckb0ne/3c9fe0813a9177aa17fdf07a89e2cee04b7684ae 2021-03-25 00:43:40 > [ 2.404368] dw-apb-uart ff1a0000.serial: forbid DMA for kernel console 2021-03-25 00:45:10 bl4ckb0ne: does it have /dev/mmc* ? 2021-03-25 00:46:51 nope 2021-03-25 00:46:59 mmmmm.. missing qhci 2021-03-25 00:47:43 er cqhci 2021-03-25 00:53:32 welp, tomorrow then 2021-03-25 00:58:48 yeah, i would have thought pulling in pl330 would fix cqhci.. not sure where that is 2021-03-25 01:43:15 i installed transmission-daemon 2021-03-25 01:43:33 but i see nothing related to transmission in the output of rc-status -s 2021-03-25 01:43:48 does transmission-daemon not include a transmission service? 2021-03-25 08:09:42 I'm getting an error with my python3.7-alpine: error: [Errno 2] No such file or directory: 'cargo': 'cargo' https://pastebin.com/5zQ3iKiH anyone experienced this? 2021-03-25 08:11:04 RootWyrm: CONFIG_MMC_CQHCI=y is already 2021-03-25 08:12:17 something else is problem 2021-03-25 08:13:49 will try to 'merge' some other options from my working rk3399 device 2021-03-25 12:15:48 mps: means something the CQHCI is attached to isn't initting 2021-03-25 12:24:24 RootWyrm: I'm building new version of kernel with more options enabled 2021-03-25 12:24:48 problem is I don't have this SBC 2021-03-25 12:25:32 if I have it then it would be easier and faster to find what is missing 2021-03-25 12:26:01 and I don't like idea to enable too much drivers in alpine kernels 2021-03-25 12:49:36 'dwmmc_rockchip fe320000.mmc: Got CD GPIO' and 'mmc0: CQHCI version 5.10' but this boots from mmc1 2021-03-25 12:55:10 what's the issue with that? 2021-03-25 13:05:21 bleb: There should be a transmission-daemon service: https://pkgs.alpinelinux.org/contents?file=&path=&name=transmission-daemon-openrc&branch=v3.13&repo=community&arch=x86_64 2021-03-25 13:05:45 RootWyrm: it is activated for emmc, but not for mmc 2021-03-25 13:06:20 if emmc is hot, mmc is hot. 2021-03-25 13:06:57 no necessary 2021-03-25 13:08:01 on the board in question, yes. they're sharing a block. i don't think the mmc4.5 will fully detect without sd inserted though. (mind, i am working from memory on the rk3399.) 2021-03-25 13:10:30 block diagram of this SBC is not clear about this 2021-03-25 13:14:34 per datasheet, there's a single emmc phy which can supply >1 downstream with a common strobe, clock, and command line 2021-03-25 13:15:52 ah, no, i take that back.. it does have separate sdmmc pins. huh. 2021-03-25 13:18:40 RootWyrm: it's cumbersome to understand, for me at least 2021-03-25 13:19:03 mps: not just you. rockchips documentation sucks ass. 2021-03-25 13:37:45 afaict the sdmmc should be same driver as emmc though. whether or not that's accurate, 2021-03-25 13:41:52 I'm not sure, at least for this hellios64 board 2021-03-25 14:11:41 it's not a board-level choice, it's the rk3399 2021-03-25 14:14:37 at least 3 years passed from time when I lastly looked at rk3399 datasheet 2021-03-25 15:27:43 we could annoy MartijnBraam for rk3399 knowledge 2021-03-25 15:28:22 the mmc interfaces are not exactly the same afaik 2021-03-25 15:28:46 at least because the mmc bus is wider than the sd one 2021-03-25 15:32:29 yes, I copied same options from my rk3399 chromebook but it didn't booted on hellios64 2021-03-25 15:32:55 so I think e/mmc interface is not same 2021-03-25 15:35:20 MartijnBraam: yeah, but _should_ be same driver i would think. 2021-03-25 15:36:18 bl4ckb0ne: anyway, I will upload new kernel in about 10-15 minutes, when it finish building 2021-03-25 15:36:57 Hello, everyone. I am trying to install Alpine Linux on my homeserver as a Xen dom0 with LVM on LUKS, however am having difficulties doing that. My attempts were so numerous I have resorted to using scripts written in accordance with official guide. Current versions of the scripts are: https://textbin.net/raw/xl2xdsojit and https://textbin.net/raw/t0ypnn2gnl first is the one run initially and the second one is run after getting to chroot 2021-03-25 15:36:57 environment. The steps described in post-setup part of Xen guide are done manually for now. However, after I boot, I get "mounting root system failed" error. Can anyone possibly help? 2021-03-25 15:37:16 neat 2021-03-25 15:37:19 To be specific, "mounting /dev/mapper/vg0-root on /sysroot failed: no such file or directory" 2021-03-25 15:38:39 sleep 2s every line is for me to catch any errors that might appear during script execution 2021-03-25 15:48:46 bl4ckb0ne: https://dev.alpinelinux.org/~mps/linux-edge-5.11.9-r2.apk 2021-03-25 15:50:34 nice 2021-03-25 15:50:38 powering up the board 2021-03-25 15:51:45 lets hope this time, crossing fingers :) 2021-03-25 15:53:06 apk static is very nice 2021-03-25 15:55:11 no luck :( https://paste.sr.ht/~bl4ckb0ne/8b685a15db48216b260ad70c775bcc66e90187c5 2021-03-25 15:55:48 uh 2021-03-25 15:56:10 you don't need this 'modules=pl330' in kernel cmdline 2021-03-25 15:56:36 weird. [ 2.387713] Synopsys Designware Multimedia Card Interface Driver 2021-03-25 15:56:36 oh right, leftover sorry 2021-03-25 15:56:43 so it's finding it. 2021-03-25 15:57:25 RootWyrm: yes, weird 2021-03-25 15:58:18 last idea I have is to copy my config from linux-gru and add/enable serial/console driver for this board 2021-03-25 15:58:46 linux-gru is alpine kernel for rk3399 gru-kevin chromebook 2021-03-25 15:59:09 wait, i see what's up 2021-03-25 16:00:25 O_O 2021-03-25 16:00:51 rv808 2021-03-25 16:01:34 808? 2021-03-25 16:01:34 er, sorry, rk808. in the good boot there's an rk808-regulator that appears to init vcc. 2021-03-25 16:02:25 aha 2021-03-25 16:02:32 haha, yep, that's it. 2021-03-25 16:02:46 crap, yes 2021-03-25 16:02:46 also why it's not detecting the e1000's or the rest. 2021-03-25 16:02:55 it is enabled in linux-gru 2021-03-25 16:03:57 rest is not important for now, we want it to boot correctly first, rest will be added as modules (if it is not already there) 2021-03-25 16:04:39 e1000's y default, last i looked. question is why rk808 isn't initting. 2021-03-25 16:04:53 but builders are loaded too much right now, will add CONFIG_MFD_RK808=y later and build again 2021-03-25 16:05:29 oh, it's not actually in there? yeah, that'd do it. had me worried this thing was running customized f/w and registers shifted. 2021-03-25 16:05:37 and now going to eat something, lunch time is nearly passed 2021-03-25 16:07:39 GentooChad: did it ask for luks passphrase? if yes it's missing vgscan after open luks device 2021-03-25 16:09:38 Yes it did, but after I enter it, I get this message 2021-03-25 16:10:47 Well, any way to fix it? 2021-03-25 16:11:35 what modules do you have in '/etc/mkinitfs/mkinitfs.conf' 2021-03-25 16:13:47 modules/features 2021-03-25 16:46:58 ey guys, this EOL timeline is only valid for main repository right? 2021-03-25 16:54:54 yes, community is only supported for one release period 2021-03-25 17:00:22 would be a good idea to warn about it when doing apk update? E.g. You are using a not supported repository! 2021-03-25 17:00:43 Yeah, that'd be nice 2021-03-25 17:01:58 but maybe the repository should include some signed field about support expiration 2021-03-25 17:02:16 the apkindex is singed 2021-03-25 17:02:19 signed 2021-03-25 17:02:34 donoban: might be a feature for apk 3 2021-03-25 17:03:20 hehe nice, I have to check some VM's that probably are running packages from community 2021-03-25 17:41:56 Hi, anyone know if rshd/rexecd is available on alpine? Need it for some legacy software on the local network. 2021-03-25 18:04:50 Hi there. 2021-03-25 18:06:48 I have a general question regarding packaging. On 22th March crystal lang hited 1.0.0 how many days does it ususally take to update package in Alpine linux? 2021-03-25 18:07:32 Or is the package already in some sort of community testing phase? 2021-03-25 18:07:52 frojnd: it doesn't pass tests 2021-03-25 18:09:07 OK. It's officially supported which is good, but how many days does it usually take to update package? (any package) 2021-03-25 18:09:22 depends on the maintainer 2021-03-25 18:09:34 frojnd: https://gitlab.alpinelinux.org/alpine/aports/-/issues/12518 2021-03-25 18:10:00 on edge, it can range from hours to days to weeks, depending on the maintainer, package and other circumstances 2021-03-25 18:10:30 and this https://gitlab.alpinelinux.org/alpine/aports/-/merge_requests/17747 2021-03-25 18:10:34 https://repology.org/repository/alpine_edge 2021-03-25 18:11:05 mps: Oh I see it's still on 0.35 2021-03-25 18:11:17 yes 2021-03-25 18:11:31 ikke: yeah was thinking to switch to Alpine from Arch, love the edge / rolling philosophy 2021-03-25 18:11:58 compared to arch, edge is slightly less stable 2021-03-25 18:12:12 but, arch is rolling as well 2021-03-25 18:17:30 alpine is not rolling 2021-03-25 18:19:01 edge is 2021-03-25 18:20:02 that's why my firefox is one month old? 2021-03-25 18:20:19 that's not rolling 2021-03-25 18:21:04 rolling does not mean everything is instantly updated 2021-03-25 18:22:08 well if browser is not updated one month ... that's far from rolling 2021-03-25 18:22:30 maybe there are issues that prevent it from being upgraded 2021-03-25 18:22:35 browser is not some package who use 2 users during testing 2021-03-25 18:22:41 does not mean it's not rolling 2021-03-25 18:22:49 rolling release doesn't mean fast updates, it just means continuously updated 2021-03-25 18:23:00 exactly 2021-03-25 18:23:10 yes, my browser does not continuously updating :/ 2021-03-25 18:23:45 not sure what's more used than web browser in desktop ... 2021-03-25 18:23:56 terminal :P 2021-03-25 18:24:06 yes 2021-03-25 18:24:09 obviously 2021-03-25 18:24:24 it does get continuously updated, but that doesn't mean it happens fast 2021-03-25 18:24:36 but browser should get special attention anyway 2021-03-25 18:24:54 that has no bearing on whether or not alpine edge is a rolling release 2021-03-25 18:24:54 Browsers get lots of attention 2021-03-25 18:25:01 well so rolling + edge but old browser, that's edge :D 2021-03-25 18:25:38 well if grub or bash is updated fast that's another story :/ 2021-03-25 18:25:53 my browser is not rolling and not edge 2021-03-25 18:26:30 it is, you just refuse to accept what that means 2021-03-25 18:29:39 if somebody is on edge, it's one who follows wiki, bc. that's living on edge of centuries :/ 2021-03-25 18:30:16 qman__: no idea where to get info about firefox, mailing list is just full of spam, otherwise also info from past decade 2021-03-25 19:07:09 How does upgrading between releases work? I see that v3.13 is supported untill 2022-11-01. After that date you are advised to upgrade? But how is the update? Before arch cca 10 years ago I was using Ubuntu and upgrading from one release to another was painful and almost always broke something.. 2021-03-25 19:09:51 frojnd: generally? `sed -i -e 's/3\.12/3\.13/g' /etc/apk/repositories && apk update && apk upgrade` 2021-03-25 19:24:45 good news for today is, that I got rid of grub 2021-03-25 19:24:56 did you install lilo 2021-03-25 19:25:16 will be fun after kernel update when I will not boot :/ 2021-03-25 19:25:34 loosh: no, booting using efistub from uefi bios without fucking bootloader 2021-03-25 19:26:48 Never got, how that gurb can be so ubiquitous in linux, even tiny distros use that 2021-03-25 19:31:13 what permission do u have in /boot/* for these nonsenses? somehow during copying it all got 755 2021-03-25 19:31:30 644 should be okay for all? 2021-03-25 19:31:31 did you copy files from fatfs? 2021-03-25 19:31:59 copied them several times :/ there and back 2021-03-25 19:32:15 oh can create them them agan reinstalling linux probably 2021-03-25 19:33:05 fuck I have feeling sooner or later I will not boot 2021-03-25 19:35:14 well it created it with 755 2021-03-25 19:36:45 before it was not the case 2021-03-25 19:37:07 but yes, /boot is vfat 2021-03-25 19:39:42 but it's probably all right as by default boot crap is mounted with fmask dmask 022 2021-03-25 19:48:29 helby: and now you know why i made an imager :P 2021-03-25 19:49:43 /b 6 2021-03-25 20:01:23 RootWyrm: what imager? 2021-03-25 20:02:17 helby: https://github.com/rootwyrm/alpine-imager - can easily build x86/x86_64 too, just rpi is what i use it for. 2021-03-25 20:02:33 and it's on the interminably long backlog 2021-03-25 20:05:14 well install alpine is fast, but fuckering with config again and again so time consuming 2021-03-25 20:05:26 did it several times this week already :/ 2021-03-25 20:06:14 with my tiny system is faster to fresh install than dig into issues whole night :/ 2021-03-25 20:16:19 I have to check lbu backup, what it can do, never touched it 2021-03-25 20:22:25 a lot of fuckery about diskless mode, never got how it can be used 2021-03-25 20:23:14 RootWyrm: I written something similar myself for building x86_64, RPI, and general aarch64 images too. Like you using qemu-user inside a docker container, in my case I'm driving it all from an Ansible playbook 2021-03-25 20:26:31 I will copy your message probably for next generations :/ 2021-03-25 20:26:40 qemu user inside docker from ansible 2021-03-25 20:27:52 helby: I didn't say it was pretty ;-) 2021-03-25 20:28:11 if I include in lbu directory will it backup also anything inside of it? ... I doubt 2021-03-25 20:32:15 not bad that lbu 2021-03-25 20:35:16 pretty easy, just have to remove a lot of nonsense from /etc 2021-03-25 21:05:43 if manualy source .ashrc by '. .ashrc' then it somehow add weird file with name of some part of alias inside of .ashrc ... any idea? 2021-03-25 21:06:32 ... just saying, don't worry :/ 2021-03-25 21:12:26 Hi all :-) 2021-03-25 21:12:50 What are the criterias to include an app into community repo ? 2021-03-25 21:21:25 kmmndr_: be a good maintainer! 2021-03-25 21:24:32 artok: What are the critirias of a good maintainer :-D 2021-03-25 21:41:13 well it is good to follow issues and do updates for the package 2021-03-25 21:41:59 first it will live in testing repo, then it might take itself into community, at least if the license is ok 2021-03-25 21:42:19 what is the package? 2021-03-25 21:43:14 I was thinking to gcompris-qt 2021-03-25 23:27:03 frojnd: alpine mostly uses upstream packages 2021-03-25 23:27:23 ubuntu needs that crap mainly because they like to add broken patches to everything 2021-03-25 23:27:41 see: adding local privilege escalation to gnome 2021-03-25 23:27:45 ...and thereby giving "Linux" a reputation for being buggy 2021-03-25 23:27:50 however update-conf is an important step 2021-03-25 23:27:54 and reading release notes 2021-03-25 23:39:23 but linux is buggy anyway 2021-03-25 23:41:55 s/linux/Software/ 2021-03-25 23:42:08 Software is shit. Go back to logging. 2021-03-25 23:42:40 yes 2021-03-25 23:42:55 except logging is also buggy 2021-03-25 23:43:18 wood logging? 2021-03-25 23:43:31 ah, that one is ok :) 2021-03-25 23:43:37 :) 2021-03-25 23:43:41 I thought about log books 2021-03-25 23:45:07 when nature/god created this world first it created bugs, after then other animals ;) 2021-03-25 23:45:26 that* 2021-03-25 23:47:37 hmm, and at last red wine 2021-03-25 23:48:33 not a terrible idea... I have a 2009 Chianti waiting to be decorked one day 2021-03-25 23:49:54 heh, synchronicity at work, I have macedonian kratoshija also 2009 2021-03-25 23:50:11 Right, I need to get some quality alcohol to drink after working on software for the day 2021-03-25 23:50:16 Now I know why I feel so unbalanced 2021-03-25 23:51:36 spirit(us) is latin term for alcohol, I think with good reason 2021-03-25 23:51:58 machine spirit(us) 2021-03-25 23:52:01 There, it's in the name! 2021-03-25 23:53:15 breathe 2021-03-25 23:54:29 I don't know proper english word for it 2021-03-25 23:55:24 soul, mind? 2021-03-25 23:55:48 "spirits" can refer to a beverage in English as well 2021-03-25 23:57:13 s_piriti, exhale air in serbian, breathe out 2021-03-25 23:57:47 make file by exhaling air 2021-03-25 23:58:10 true meaning is 'internal fire' 2021-03-25 23:58:12 similar to "pneuma" 2021-03-25 23:58:29 a Greek concept popular with Stoic philosophers 2021-03-25 23:59:18 pneuma is greek translation, but bad one, as usual with greek translations 2021-03-26 00:00:42 Greeks borrowed a lot of stuff from further East 2021-03-26 00:01:13 no, they borrowed from north, hyperboreia 2021-03-26 00:02:07 you know their mythology, they come there in 3 waves, all from north 2021-03-26 00:03:06 Plato explained in Cratylus about their language 2021-03-26 00:03:22 yeah, it's hard to determine what is factual and what is shrouded in mythology 2021-03-26 00:04:12 it is not hard if you understand old languages, then it is simple 2021-03-26 00:04:24 lol... yes, I should probably put some more effort into that 2021-03-26 00:04:48 tehcloud: do you know etrurian name ethymology 2021-03-26 00:05:35 and meaning of words pediater, psyhiater, geriater etc ... 2021-03-26 00:06:14 I didn't wrote them properly in english but I hope you understand 2021-03-26 00:06:52 'iater' is key 2021-03-26 00:07:51 notice that it is 'similar' with (i)etr(urian) 2021-03-26 00:08:42 and you probably have seen 'liver' artefact from Etruria 2021-03-26 00:08:55 :) 2021-03-26 00:09:23 granted, all scholar to day don't understand this 2021-03-26 00:09:48 oh yes, the Liver is very interesting 2021-03-26 00:09:55 but is 'bloody' simple 2021-03-26 00:10:09 translate liver to serbian ;) 2021-03-26 00:11:36 ohhhh.... 2021-03-26 00:11:44 yeah, I see the link there 2021-03-26 00:11:56 hehe 2021-03-26 00:12:07 ietra 2021-03-26 00:12:22 yes, this is the second time you've made a very good point about languages in regards to studying ancient cultures 2021-03-26 00:12:35 well, the second time in my presence ;) 2021-03-26 00:13:52 etruscan are known by using liver for their spiritual life and for medicine, and prophecy 2021-03-26 00:14:48 in serbian 'ietrari' means those who use liver for all above activities 2021-03-26 00:15:11 remove 'i' and you got all 2021-03-26 00:15:56 I told you that serbian is foundation for all european languages 2021-03-26 00:17:03 and to add, also I couldn't accept this as true for long time 2021-03-26 00:18:05 you've made compelling points supporting it, so it at least has a very strong link to the foundational language 2021-03-26 00:19:06 but it is not nowadays serbian language, but one from 19th century 2021-03-26 00:20:09 I was grown up in part where old language is preserved so i was lucky to understand these 'fine nuances' 2021-03-26 00:21:12 I think it was in the 19th century that an understanding of the link between Serbian epic poetry and Homeric poetry was established in the minds of academics 2021-03-26 00:22:27 yes, Pisistrat ordered someone (forgot name) to translate Illiad from barbaric lang to greek 2021-03-26 00:23:08 barbarians for greek were at north (again north) 2021-03-26 00:25:02 and I hope you know about vincha script 2021-03-26 00:25:10 yes, because of you ;) 2021-03-26 00:25:27 ah, sorry for repeating 2021-03-26 00:25:38 lol, don't be sorry 2021-03-26 00:28:30 I think you know how brahmans in india 'keep' vedas in intact state for millennia 2021-03-26 00:30:32 you mean through preserving the language? 2021-03-26 00:30:56 reciting and mentally memorizing them 2021-03-26 00:31:15 and language also by this method 2021-03-26 00:32:40 I used this method to locate Troy (Trois) and I found it 2021-03-26 00:34:51 and not only I, few centuries ago poet Hector Hectorovich wrote poem about where Troy is 2021-03-26 00:36:07 yeah, not everyone agrees on it being at Hisarlik 2021-03-26 00:37:09 Petar Hecktorovich, sorry 2021-03-26 00:37:16 wine works :) 2021-03-26 00:37:24 but there was a lot of money and glamour around Schliemann 2021-03-26 00:37:36 yes 2021-03-26 00:37:37 so you aren't allowed to ask these questions ;) 2021-03-26 00:38:20 Schlieman was merchant, not archeologist 2021-03-26 00:39:00 I think his setup in Greece was turned into some sort of numismatics museum too 2021-03-26 00:39:44 someone must be idiot to believe in Schlieman 'discovery' 2021-03-26 00:40:33 Sirius can't be over sea in Hisarlik and never could be 2021-03-26 00:41:48 and a loot of eels in such small creek in Hisarlik 2021-03-26 00:43:09 and a lot of stupid facts 2021-03-26 00:44:12 Achiles followed goats and by foot come to Troy from Greece, around Black Sea :) 2021-03-26 00:44:53 or goats are swimed over Dardanelly :) 2021-03-26 00:45:30 lol... yeah, but there are other things to consider too 2021-03-26 00:45:53 geography does change over time, and if you look at Thermopylae as an example, there is a lot more land next to the mountain pass now than in the time of the 300 Spartans 2021-03-26 00:47:06 heh, in battle were engaged about 500-600 Spartans and about 400 Athens 2021-03-26 00:47:35 yeah, the entire army from a smaller city state was also there and got totally slaughtered :p 2021-03-26 00:47:52 right 2021-03-26 00:50:04 they say that battle is lost because of betrayal (usual motive used in mythology) but actually most of soldiers are died before that 2021-03-26 00:51:16 maybe they are betrayed at the end, but they were exhausted already, and not much of them were alive at final crush 2021-03-26 00:52:17 but yes, geography is changed for some degree from this time 2021-03-26 00:52:32 but Dardanelly not much 2021-03-26 00:54:08 and my monthly accounting work just finished, so I can go to bed 2021-03-26 02:47:00 Hi, is "Alpine WSL" on the windows store official/supported/etc ? 2021-03-26 02:55:27 anyone? 2021-03-26 04:32:36 wth is with freenode not liking my connect. wow 2021-03-26 04:32:44 Anyway, is "Alpine WSL" on the windows store official/supported/etc ? 2021-03-26 07:55:02 I'm trying to do alpine-setup from vbox, and it keeps failing with chronyd time out 2021-03-26 07:55:11 I checked and networks is working fine for eth0 2021-03-26 07:56:48 wget: can't connect to remote host: Operation timed out 2021-03-26 07:57:00 This is after starting chronyd ... 2021-03-26 07:58:28 I thought there was goona be more mirrors? Just one? https://imgur.com/a/ykFSVRr 2021-03-26 08:00:11 frojnd: why not qemu 2021-03-26 08:06:28 mps: to run in qemu? 2021-03-26 08:06:57 I guess because of the performance I Could but... why it won't run in vbox? 2021-03-26 08:08:29 https://imgur.com/a/IAKbNeP 2021-03-26 08:08:44 frojnd: well, not much of us have experience with vbox, we mostly use qemu 2021-03-26 08:09:14 so I'm not sure if we can help much with vbox 2021-03-26 08:11:03 Ok let me try in qemu 2021-03-26 08:12:36 and it is 'setup-alpine' not 'alpine-setup' iirc 2021-03-26 08:22:15 Yeah, setup-alpine 2021-03-26 08:37:16 Well fuck 2021-03-26 08:37:28 It's native performance :D 2021-03-26 08:37:32 us 2021-03-26 08:38:11 And it works.. 2021-03-26 08:40:35 :) 2021-03-26 10:05:35 I see open zfs is part of alpine 2021-03-26 10:05:43 Gotta find full encryption guide 2021-03-26 10:10:26 Hello, why duplicity, s6-overlay and redis package didn't updated in 3.13.3 ? 2021-03-26 10:35:27 Note: In order to setup GRUB with UEFI, you are required to use the edge branch with the main and community repository. The reason for this is that efibootmgr is not available in the stable branch. If you do not want to switch completely over to edge you can do something called repository pinning. You will need to do this after the setup-apkrepos step. 2021-03-26 10:35:43 https://pkgs.alpinelinux.org/packages?name=efibootmgr&branch=v3.13 2021-03-26 10:36:25 So no switching to edge and no pinning? 2021-03-26 10:37:47 grub should be not available in stable branch as it has to be stable 2021-03-26 10:38:15 https://pkgs.alpinelinux.org/packages?name=grub&branch=v3.13 2021-03-26 10:39:24 yes, so both efibootmgr and grub are in main stable 2021-03-26 10:39:34 what's your story about then? 2021-03-26 10:39:48 I should update wiki or someone? 2021-03-26 10:40:10 I'm installing alpine in qemu and would like to try full disk encryption with zfs: https://wiki.alpinelinux.org/wiki/LVM_on_LUKS#Setting_up_Alpine_Linux_Using_LVM_on_Top_of_a_LUKS_Partition 2021-03-26 10:40:37 update wiki? delete it 2021-03-26 10:41:16 Just that section right? :) 2021-03-26 10:41:27 Or u meant the whole wiki is obsolete? 2021-03-26 10:41:28 which one? I don't see it 2021-03-26 10:41:45 yes, it is obsolete whole 2021-03-26 10:42:10 don't relly on anyting from there 2021-03-26 10:42:15 docs? 2021-03-26 10:42:22 there is not much info 2021-03-26 10:42:34 there is just everything what u don't need to know 2021-03-26 10:42:49 In wiki or docs? or Both? 2021-03-26 10:42:52 how to dd iso to usb and how to plug in your machine 2021-03-26 10:43:00 docs are okay 2021-03-26 10:43:06 but nothing there 2021-03-26 10:43:10 wiki is full of ancient crap 2021-03-26 10:43:31 Docs don't have search option? Can't find luks 2021-03-26 10:44:08 there is search 2021-03-26 10:44:19 but in docs u find nothing, bc. nothing there :) 2021-03-26 10:44:39 if u need something u have to check arch wiki gento wiki 2021-03-26 10:44:55 Ah... ok then ;) used to arch's wiki 2021-03-26 10:45:10 But at least it gave me some info about setup-this setup-that :) 2021-03-26 10:45:22 i don't understand the package update strategy of Alpine. there is duplicity minor upgrade. why 3.13.3 didn't get it ? (0.8.13 to 0.8.18) 2021-03-26 10:45:24 then next year during election, vote for alpine wiki deletion ;) 2021-03-26 10:45:50 Hehe 2021-03-26 10:46:03 Just link it to Arch wiki hehe 2021-03-26 10:46:58 frojnd: yes, it gives some info, setup this and setup that, the issue is, sometimes u don't have to yet, it was actual 10 years ago ;) 2021-03-26 10:49:04 I see that it actually defaults to ext4 data type 2021-03-26 10:49:10 On that wiki... 2021-03-26 10:51:20 nobody know ? ^^ 2021-03-26 10:51:33 or i just want a link to the package update strategy of alpine 2021-03-26 10:52:38 GreyXor: I did not get what u asked 2021-03-26 10:53:14 helby i didn't understand why this package it isn't updated to alpine 3.13.3 https://pkgs.alpinelinux.org/packages?name=duplicity&branch=edge 2021-03-26 10:54:08 how it is not? 2021-03-26 10:54:34 oh, u mean, why it's in edge, not 3.13 2021-03-26 10:54:47 duplicity edge is on 0.8.18 and 3.13 still in 0.8.13 2021-03-26 10:55:01 oh 2021-03-26 10:55:13 release is not rolling what I know 2021-03-26 10:55:19 since months actually 2021-03-26 10:55:20 or whatever it is calling like stable 2021-03-26 10:56:08 bc. it's something like 'stable' so packages are not updated like in edge I assume 2021-03-26 10:56:17 ok 2021-03-26 10:56:39 so i understand that it will be available in alpine 3.14 2021-03-26 10:56:43 so probably it get just some kind security fixes during 6 months of release 2021-03-26 10:57:54 I need this version in alpine 3.13. Is it a good idea to force the update from the edge repository or it's bad idea? 2021-03-26 10:58:53 uhM, are you the package maitainer? 2021-03-26 10:58:57 GreyXor: stable packages are update in case there are bug or security problem 2021-03-26 10:59:07 why don't try to just build the package on 3.13 and install if all runs ok? 2021-03-26 10:59:56 https://git.alpinelinux.org/aports/commit/?id=11708fedb1d77b3d72b4541a7a955dfa003750a3 is this commit yours? 2021-03-26 11:00:10 why buuild if it's in edge 2021-03-26 11:00:12 just install it 2021-03-26 11:00:57 Yes it's mine 2021-03-26 11:02:08 helby: but can you install a package from edge without upgrading all on 'apk upgrade'? maybe manual downloading it? 2021-03-26 11:03:30 u can somehow use edge just for one package I think 2021-03-26 11:04:48 in wiki it's 'apk add stableapp newapp@edge bleedingapp@testing' 2021-03-26 11:05:50 oh nice, I didn't know 2021-03-26 11:06:29 check that package management in wiki, not sure about this command I typed, looks ancient 2021-03-26 11:13:05 --force-broken-world 2021-03-26 11:13:11 actually i want to understand why the package didn't in 3.13.3 ^ 2021-03-26 11:14:52 duplicity 18 is a minor update. so it should logically be in 3.13.3. whatever i will wait for 3.14 u_u 2021-03-26 11:19:24 maybe updated late for 3.13.3 ? 2021-03-26 11:22:17 GreyXor: if you want package to be backported you should open wish request on aports, or ask maintainer, or even ask on irc with reasoning why it should be backported 2021-03-26 11:22:19 GreyXor: regarding s6-overlay, I'm the maintainer for it. I am working on dealing with an issue with it in 3.13 and so it is likely to be updated in the near future. 2021-03-26 11:23:26 GreyXor: keep in mind that some packages are relatively easy to backport but some not, for example libs in case of ABI/API changes 2021-03-26 12:01:55 ey minimal, I just remember what you said about auditbeats and have another idea, what about an acf-audit? 2021-03-26 12:02:33 donoban: have never used any acf stuff 2021-03-26 12:04:42 hehe, I used it a lot for iptables, I think it's nice and simple 2021-03-26 12:05:42 I've packaged up auditbeat for Alpine before locally. I started working on a "beats" package as currently the filebeat and metricbeat packages are a mess - they're built as 2 distinct packages despite using the same source tarfile (and they're currently on differing versions). Just got a bit stuck trying to ensure that only the OSS versions of the beats were being built (so no Elastic licensed code enabled) 2021-03-26 12:07:13 donoban: what was your idea about acf-audit? for it to handle both auditd and auditbeat config? 2021-03-26 12:08:00 no no, only auditd 2021-03-26 12:08:18 configure audit.rules and see logs in a simple but nice format 2021-03-26 12:10:01 I think that auditbeat alone is already over acf 2021-03-26 12:10:14 have been meaning to talk to the audit Alpine maintainer - its still on 2.8.x whereas 3.0.1 is out which supports some syscalls added in recent kernels 2021-03-26 12:11:59 donoban: there is no auditbeat (yet) packaged for Alpine so I wouldn't expect acf to support it 2021-03-26 12:13:49 I don't know very much about auditbeat but I saw that it has his own web reports right? Is not very likely that some auditbeat user would need acf for it, but for auditd alone it could be useful 2021-03-26 12:14:33 I will try to do something when I finish my acf-docker attempt :D 2021-03-26 12:16:06 auditbeat (like all the Elastic beats) is designed to collect data and forward it to Redis/Kafka/Logstash/Elasticsearch for central processing and/or storage 2021-03-26 12:17:18 yeah, it seems a server oriented tool 2021-03-26 12:17:40 I'm trying to setup auditd for my desktop so I don't need too much features 2021-03-26 12:18:02 typically yes, as a way to feed data to a central monitoring/management/security system 2021-03-26 13:02:51 donoban: I have a local packaged version of audit 3.0.1, it builds but haven't done any testing of it so far. 2021-03-26 13:04:38 oh don't worry, I think that I'm fine with auditd and file access, I have another things to improve before switching my main desktop to alpine 2021-03-26 13:14:52 minimal: was referring to audit/auditd - current Edge package is 2.8.5, I've locally built 3.0.1 for newer kernels support 2021-03-26 13:15:40 donoban: was referring to audit/auditd - current Edge package is 2.8.5, I've locally built 3.0.1 for newer kernels support 2021-03-26 13:16:10 ahh, I understood that you refer to auditbeats and alpine 3.0.1 :D 2021-03-26 13:16:15 donoban: Re: desktop, are you using Wayland with Alpine or X? 2021-03-26 13:16:42 I'm testing KDE with wayland, but having frequent segfaults 2021-03-26 13:18:03 I configured Linux for get coredumps but I get some different process, I expect to get win_wayland and get... 2021-03-26 13:19:03 https://tpaste.us/pr5W 2021-03-26 13:19:44 I'm running edge, do you think that I would get too much unstability there? 2021-03-26 13:20:40 donoban: haven't really done anything with desktop in Alpine so far but want to get Wayland working, ideally with MATE 2021-03-26 13:21:27 uhM Mate, I could switch, some years using KDE, new OS new desktop :D 2021-03-26 14:03:33 wayland and edge works well for me on desktop 2021-03-26 14:06:52 don't know what can go wrong with 500 KDE packages :/ 2021-03-26 14:12:37 hehe, I installed whole KDE by accident, I will reinstall a minimal setup 2021-03-26 14:12:53 I'm testing MATE now but I feel that is not running plasma 2021-03-26 14:15:44 these monsters still install and use xorg no? 2021-03-26 14:16:19 https://mate-desktop.org/blog/2020-02-10-mate-1-24-released/ 2021-03-26 14:16:21 and every other package available in repository obviously :/ 2021-03-26 14:17:07 what monsters? 2021-03-26 14:19:05 these desktops packs 2021-03-26 14:19:06 helby: what window manager do you use? 2021-03-26 14:19:23 sway as there is not much choice in wayland 2021-03-26 14:19:53 tried build hikari but somehow failed so 2021-03-26 14:20:09 I'm gonna try it 2021-03-26 14:20:55 that works out of box 2021-03-26 14:23:27 uhm lol 2021-03-26 14:23:31 I can't open anything 2021-03-26 14:23:53 it's keyboard based like i3 2021-03-26 14:24:03 right click, alt+f2. ctrl/alt f1,esc... 2021-03-26 14:24:03 if u don't know key binding then ... :/ 2021-03-26 14:24:27 is there a key binding for run custom command? 2021-03-26 14:24:32 yes 2021-03-26 14:24:36 u can bind anything 2021-03-26 14:24:51 but default does it have something binded? 2021-03-26 14:25:03 yes ctrl alt enter - alacritty terminal 2021-03-26 14:25:23 of course u have ot install it or change keybiding for other terminal 2021-03-26 14:26:00 default config is in /etc/sway/config u have to copy it to ~/.config/sway/config then u can customize it 2021-03-26 14:26:08 ok 2021-03-26 14:26:11 ty 2021-03-26 14:30:01 puff I will try again with kde :S 2021-03-26 14:30:46 :/ 2021-03-26 14:32:04 dont like gnome? 2021-03-26 14:32:17 that monsterd at least usually works 2021-03-26 14:33:08 still if I would want to use these Desktop environments I would probably don't go with alpine 2021-03-26 14:33:09 not too much, I really like kde 2021-03-26 14:33:21 but something like arch 2021-03-26 14:34:54 well, I'm decided to try, I don't need very much because I will use virtual machines, flatpack, containers, if needed 2021-03-26 14:35:38 if KDE is too unstable I will try customizing sway 2021-03-26 14:36:04 well 'very much' ... that's what KDE is I think :/ 2021-03-26 14:37:23 well, it is already there :) 2021-03-26 14:38:17 Hey Cogitri I did several tests for vscode because the solution on flatpack somehow I don't like, I'm trying now with a version of Codeserver installed with yarn and so far it seems to be fine. I have to do more tests, I also tried to start it in a webview with pywebview, it works but it is not light as I hoped, too many dependencies. 2021-03-26 14:39:24 I also found a version of vscode to install with npm but from what I understand it is missing something 2021-03-26 14:45:41 I'm having this error with podman/docker: "Error: OCI runtime error: cgroups in hybrid mode not supported, drop all controllers from cgroupv2" 2021-03-26 14:52:21 any plans to drop openrc for say s6 2021-03-26 14:52:23 ? 2021-03-26 14:58:02 what's the status on the latest kernel config mps ? 2021-03-26 15:01:25 bl4ckb0ne: didn't had time yesterday 2021-03-26 15:01:43 I think I will make it this evening 2021-03-26 15:02:03 all good 2021-03-26 15:02:21 Naib: There are some s6 fans around here. There has been discussion in the mailing list about it a few times. 2021-03-26 15:02:27 For example: https://lists.alpinelinux.org/~alpine/devel/%3C3LLUI2KOULSYM.359WA6HATX45B%408pit.net%3E 2021-03-26 15:14:24 xordspar0: ok thanks. I am primarily a gentoo user (alpine in containers and I am going to change my headless to alpine). OpenRC ... appears to have some odd development making a few nervous. Supervisor-Daemon is quite nice though 2021-03-26 15:16:11 some of us don't like supervisors by default 2021-03-26 15:18:11 which is fair, make it optional 2021-03-26 15:18:36 I am surprised this isn't in the openrc init scripts controlled via a conf.d field 2021-03-26 15:19:33 That was discussed in the above thread. :) Search for "supervise-daemon does not offer such setting" 2021-03-26 15:20:20 :) 2021-03-26 15:20:36 comes back to william not implementing it - was rewuexstb 2021-03-26 17:13:47 bl4ckb0ne: https://dev.alpinelinux.org/~mps/linux-edge-5.11.10-r1.apk 2021-03-26 17:14:09 bl4ckb0ne: when you find time 2021-03-26 20:11:04 mps: on it 2021-03-26 20:18:30 nope :( 2021-03-26 20:18:35 https://paste.sr.ht/~bl4ckb0ne/28c198fb92b6f4595714ed8df1551d3484b69a0b RootWyrm mps 2021-03-26 20:19:49 and that has the rk808 driver? 2021-03-26 20:20:05 i suppose 2021-03-26 20:20:11 no rk808 when grepping the log tho 2021-03-26 20:20:51 i think its missing 2021-03-26 20:22:44 module is there though 2021-03-26 20:23:03 bl4ckb0ne: you have initramfs 'enabled' in u-boot, though removing it wouldn't help 2021-03-26 20:23:36 RK808 is built in-kernel 2021-03-26 20:24:11 but this method we used to find working config doesn't work 2021-03-26 20:25:24 I will try to use linux-gru config with serial driver for hellios64 as a last idea 2021-03-26 20:26:06 there's still the pmos linux rockchip config 2021-03-26 20:26:15 I have to finish with crystal lang first 2021-03-26 20:26:38 bl4ckb0ne: you have separate media for both kernels, yes? 2021-03-26 20:26:48 everything is on the sd card 2021-03-26 20:27:08 bl4ckb0ne: this pmOS config will not 'fit' for alpine 2021-03-26 20:28:32 how come? 2021-03-26 20:29:11 because they enable too much 'things' 2021-03-26 20:29:37 agreed 2021-03-26 20:29:48 do you have the config for 5.11.10 somehwere? 2021-03-26 20:29:53 i could try to diff it 2021-03-26 20:30:01 and not sure will it be good for alpine config for big arm64 servers 2021-03-26 20:30:41 bl4ckb0ne: if you installed it you have it in /boot dir 2021-03-26 20:30:51 ah yes 2021-03-26 20:31:13 gonna diff that tonight 2021-03-26 20:31:49 i wonder if I could make a three way diff with the armbian config 2021-03-26 20:31:58 and check what is enabled in pmos and armbian but not ours 2021-03-26 20:32:43 Hm, yo - I just noticed when I updated my router apk update says 1 error 2021-03-26 20:32:48 but there's no output 2021-03-26 20:32:51 or anything in the logs 2021-03-26 20:33:23 er, apk upgrade 2021-03-26 20:33:25 not update 2021-03-26 20:33:29 1 error; 1176 MiB in 297 packages 2021-03-26 20:33:33 apk fix 2021-03-26 20:34:04 ... i feel stupid 2021-03-26 20:34:08 thank you 2021-03-26 20:37:55 bl4ckb0ne: when I finish with current works I will try to build it using linux-gru config 2021-03-26 20:38:20 sounds good 2021-03-26 20:38:25 gonna try some stuff on my side too 2021-03-26 20:38:58 this one works fine on rk3399 but only have to enable serial console driver 2021-03-26 20:39:36 i do all by serial 2021-03-26 20:40:12 ofc, but I don't have serial console on chromebook 2021-03-26 20:40:37 because that I didn't enabled it 2021-03-26 20:43:26 bl4ckb0ne: yes but are they different sd cards? it's not finding the rk808. 2021-03-26 20:43:43 so i'm wondering if there's a first stage binary component missing 2021-03-26 20:44:02 its the same sd card 2021-03-26 20:44:41 ok, then there's got to be something rk808 depends on missing... ugh... 2021-03-26 20:51:11 yes, this is very strange. we enabled mmc driver in-kernel but it didn't set automatically needed options for it to work 2021-03-26 20:51:33 no, it did that fine, because they are two completely separate things. 2021-03-26 20:54:18 mmc can't and shouldn't assume rk808 because that's a voltage regulator component and not common to boards. what's got me scratching my head though is what rk808 could be depending on unless they did something funky with gpio. 2021-03-26 20:55:13 yes, gpio also can be issue 2021-03-26 20:56:39 i.e. pine64 uses an rk805 in the rock64, axp803 in the pine a64, nada in the quartz64. 2021-03-26 21:01:14 ok, started build new kernel, lets see 2021-03-26 21:14:09 good, builder is not busy now 2021-03-26 21:14:16 hi 2021-03-26 21:14:44 nice 2021-03-26 21:14:45 when I run startx the screen turns off (not even backlight is on) and switching tty does not help recovering 2021-03-26 21:16:28 Get Xorg logs 2021-03-26 21:17:14 if someone from kobol.io is here please talk 2021-03-26 21:17:28 or point me to the schematic 2021-03-26 21:18:19 maybe more luck in armbian 2021-03-26 21:18:29 since its the only supported distro 2021-03-26 21:18:48 i could mail them tbh 2021-03-26 21:19:46 RootWyrm: https://wiki.kobol.io/helios64/i2c/ 2021-03-26 21:20:30 bl4ckb0ne: i need the actual full electrical schematic 2021-03-26 21:20:38 but that is interesting. 2021-03-26 21:21:14 'syr837' 'syr838' yet neither appears on the block diagram. 2021-03-26 21:21:59 the helios4 has schematics but not the helios64 2021-03-26 21:22:01 so i don't know if those are attached to the hdd power, the usb, or somewhere else. 2021-03-26 21:22:57 https://gist.github.com/theqp/c03df5089ce94e525d050b9156d7eebf 2021-03-26 21:24:58 gonna mail them then 2021-03-26 21:28:01 Thermi: do you have an idea based on the log? 2021-03-26 21:28:16 Didn't even notice your providing it 2021-03-26 21:29:16 kupi: What's the right resolution for the screen? 2021-03-26 21:29:32 1200x1600 2021-03-26 21:29:46 kupi: Also, see your own log lines 403-405, 419-432 2021-03-26 21:30:29 nvm, it's 1600x1200, just messed up the order 2021-03-26 21:31:28 kupi: Also, now armed with the log, you can use Google to find the solution 2021-03-26 21:32:41 bl4ckb0ne: https://dev.alpinelinux.org/~mps/linux-edge-5.11.10-r2.apk 2021-03-26 21:32:42 i have tried that 2021-03-26 21:33:07 bl4ckb0ne: this doesn't support initramfs 2021-03-26 21:33:09 "(EE) open /dev/dri/card0: No such file or directory" 2021-03-26 21:33:30 is this supposed to happen? 2021-03-26 21:34:04 kupi: https://arvanta.net/alpine/install-x/ 2021-03-26 21:34:20 this is for alpine in qemu but maybe could help 2021-03-26 21:34:41 kupi: Depends on the context. Generally, if you don't have a card, you don't have a kernel driver for it 2021-03-26 21:34:47 bl4ckb0ne: crossing fingers 2021-03-26 21:36:27 Thermi: I have an ATI Rage Mobility 128 in this 2021-03-26 21:36:37 That doesn't help me at all 2021-03-26 21:38:32 I have fixed 403-405, 419-432 by removing the xf86-video-r128 2021-03-26 21:38:48 oh already 2021-03-26 21:38:58 gonna eat, ill test it later tonight when my wee one's asleep 2021-03-26 21:39:19 it is a lot smaller then previous ones 2021-03-26 21:39:49 half size 2021-03-26 21:40:12 bl4ckb0ne: ok, see you tomorrow then 2021-03-26 21:40:22 see you tomorrow 2021-03-26 21:40:26 thanks for the kernels! 2021-03-26 21:40:37 heh 2021-03-26 21:40:50 thanks for testing 2021-03-26 21:40:57 we'll get it! 2021-03-26 21:41:52 this latest config surely works for few years on my rk3399 2021-03-26 21:42:30 ok, good night 2021-03-26 21:42:43 nn 2021-03-26 21:42:56 I guess the root of the problem is that the /dev/dri does not exist 2021-03-26 21:44:10 but I see no dri related errors here https://gist.github.com/theqp/40ebad512a4e440d6c44ddbe88f46d8c 2021-03-26 21:48:10 turns out my wee one is still napping 2021-03-26 21:48:44 https://cateee.net/lkddb/web-lkddb/DRM_R128.html 2021-03-26 21:49:04 i need this module but it does not exist on my system, i even installed linux-firmware 2021-03-26 21:49:14 mps RootWyrm https://paste.sr.ht/~bl4ckb0ne/8abccf864f7f2bd713203391e3ca9ff8c27d8fc3 2021-03-26 21:49:21 SUCCESS \o\ 2021-03-26 21:49:41 \o/ 2021-03-26 21:50:08 i dont see my sata drives tho, but its booted 2021-03-26 21:50:31 > lsblk: failed to access sysfs directory: /sys/dev/block: No such file or directory 2021-03-26 21:52:07 sata is not enabled in this kernel 2021-03-26 21:58:07 bl4ckb0ne: we can continue tomorrow, to enable needed drivers for this kernel. I'm tired now and have to go to sleep 2021-03-26 22:05:27 bl4ckb0ne: that's to be expected... hmm.. i do see some other missing devices, but nothing major 2021-03-26 22:15:04 no problem, we'll tackle that tomorrow 2021-03-26 22:16:26 hey, what's up? 2021-03-26 22:16:40 the sky 2021-03-26 22:17:11 odd. I'm looking up and all I can see is some concrete! 2021-03-26 22:17:38 is there something wrong with my installation of gnu/wall? 2021-03-26 22:19:29 seriously though, I was interested in Alpine and I wanted to know what kind of issues I would get with running proprietary software on it 2021-03-26 22:20:28 not many issue 2021-03-26 22:20:37 except for dependencies on glibc 2021-03-26 22:20:54 yes, just one, it will not work, but otherwise everything will be okay 2021-03-26 22:21:31 is there a way to work around this? 2021-03-26 22:21:41 yes, using windows 2021-03-26 22:21:42 only need zoom and microsoft teams to work, really 2021-03-26 22:21:49 oh, lmfao 2021-03-26 22:21:53 really 2021-03-26 22:22:00 what's the point to run it on linux? 2021-03-26 22:22:45 using linux as a daily driver, unfortunately, I am required to use teams for meetings 2021-03-26 22:24:39 then probably choose some septic distro for running shit :/ 2021-03-26 22:36:07 You could use the flatpak of them 2021-03-26 22:39:14 Cogitri, I saw some people doing that 2021-03-26 22:39:28 could give it a go through a virtual machine, thanks! 2021-03-27 00:41:28 installed openbsd, xorg works 2021-03-27 00:56:46 kupi: why it should not? 2021-03-27 00:58:08 helby on alpine I get the error "(EE) open /dev/dri/card0: No such file or directory" 2021-03-27 00:58:52 kupi: in alpine u probably just did not install necessary driver or did not add user or root /who knows/ to video group or maybe did not installed font, maybe did not run setup-eudev ... 2021-03-27 00:59:13 i have installed linux-firmware 2021-03-27 00:59:31 well u still need driver for gpu 2021-03-27 00:59:35 also even "sudo ls /dev/dri" did not list dri 2021-03-27 01:01:55 for example if I install xorg by setup-xorg-base it installs mesa driver, but I still hvae to install xf86 intel crap 2021-03-27 01:01:59 what driver? a long time ago it worked out of the box with linux firmware, this is what I have http://www.thinkwiki.org/wiki/ATI_Rage_Mobility_M3 2021-03-27 01:02:05 of course at least one shitty font 2021-03-27 01:02:20 helby: on openrc it works with the vesa driver 2021-03-27 01:02:25 *openbsd 2021-03-27 01:05:11 there is xf86-video-r128 2021-03-27 01:05:18 not sure if u got that 2021-03-27 01:05:31 I doubt that setup-xorg installed it 2021-03-27 01:06:34 well openbsd works well on thinkpads, used that OS for years 2021-03-27 01:06:39 but fucking performance ... 2021-03-27 01:07:56 r128 is buggy 2021-03-27 01:08:15 https://usercontent.irccloud-cdn.com/file/k39lWLM8/IMG_20210327_000018.jpg 2021-03-27 01:08:23 this is how it looks with openbsd 2021-03-27 01:08:55 but as you can see here I have tried with r128 on alpine too https://gist.github.com/theqp/c03df5089ce94e525d050b9156d7eebf 2021-03-27 01:09:54 so even broken on openbsd? 2021-03-27 01:10:21 with r128 yes, but at least it starts 2021-03-27 01:10:46 on alpine the screen turns black and only reboot can make the system usable (not even swiching tty) 2021-03-27 01:11:04 with vesa on openbsd it works perfectly 2021-03-27 01:11:23 a few year ago it worked on alpine too 2021-03-27 01:12:09 u get the same result on alpine if u have that driver or uninstalled ? 2021-03-27 01:12:14 or did not try? 2021-03-27 01:12:42 i tried without r128 on alpine too 2021-03-27 01:15:17 try wayland :/ 2021-03-27 01:15:51 wayland does not work on ancient stuff like this :P 2021-03-27 01:17:13 lol this is from Oct 1999 2021-03-27 01:17:13 what year is that? 2021-03-27 01:17:18 oh nice shit 2021-03-27 01:17:29 21 years 2021-03-27 01:18:02 bought this laptop for 6.5 usd 2021-03-27 01:18:19 hear that netbsd works on ancient shit maybe ... for me it never even boot 2021-03-27 01:18:25 and i love it 1600x1200 2021-03-27 01:18:43 somehow good resolution for that year 2021-03-27 01:18:48 how that machine looks? 2021-03-27 01:20:05 good screen then, not so stupidly wide like most of laptops 2021-03-27 01:21:45 https://imgur.com/a/u1Ijirc 2021-03-27 01:24:20 what's the name of that? 2021-03-27 01:24:37 I used 3:2 ratio on pixelbook and it was amazing compared to these wide crap 2021-03-27 01:24:51 but love keyboard on my T420 2021-03-27 01:26:19 a21p 2021-03-27 01:26:58 it's biggest weakness is low ram (362M only) 2021-03-27 01:27:01 it's really old, never heard of it 2021-03-27 01:27:05 it has maybe even floppy? 2021-03-27 01:27:08 yes 2021-03-27 01:27:10 haha 2021-03-27 01:27:23 I think openbsd has even floppy installer 2021-03-27 01:27:55 helby: QNX ran off a floppy 2021-03-27 01:28:20 Complete with GUI and browser 2021-03-27 01:28:28 well probaly not enough for browser, but with tiny wm enough ram 2021-03-27 01:28:38 i searched for floppy for sale and it showed me floppy disk shaped powerbanks lol 2021-03-27 01:28:41 All written in ASM 2021-03-27 01:29:13 browsing is possible with links 2021-03-27 01:29:18 or lynx 2021-03-27 01:29:25 true 2021-03-27 01:30:37 with wm and some terminal windows I use just around 100mb ram, so 2021-03-27 01:34:46 wyoung: what it can be used for? 2021-03-27 01:35:52 kupi: laso curious why would u want to run xorg with that machine 2021-03-27 01:36:22 1. i am experimenting with it 2021-03-27 01:36:39 2. i can't find how to run console mode with 1600x1200 resolution 2021-03-27 01:36:43 on openbsd 2021-03-27 01:37:44 just rewrite the kernel, no big deal 2021-03-27 01:42:24 kupi: not sure, but it's probably just matter of chaning size of font spleen? 2021-03-27 01:43:10 no 2021-03-27 01:43:17 if the resolution is wrong 2021-03-27 01:43:25 then the text is blurry 2021-03-27 01:43:41 wikipedia is usable with dillo 2021-03-27 01:43:51 only 140mb ram is used 2021-03-27 01:50:13 what resolution u get there if not your default? 2021-03-27 01:51:42 yes these browsers are nice, like lynx, if only web is not so much fucked up 2021-03-27 01:58:38 I would like to get machine like that for 6,5 EUR :/ at least for backup my configs and fuckering around 2021-03-27 02:54:29 helby: idk it's somewhere between 640x480 and 1024x768 2021-03-27 03:10:21 kupi: maybe u can try live usb something like manjaro, as it's bloated with all shit they found around ... to figure out if u missing some driver or your destiny is console :/ 2021-03-27 03:10:57 i want a non-systemd system on this laptop 2021-03-27 03:11:03 for diversity 2021-03-27 03:11:57 then do it 2021-03-27 03:12:04 :p 2021-03-27 03:13:09 try openwrt lol 2021-03-27 03:13:13 I meant it just to test if u missing some driver 2021-03-27 03:13:34 does it boot from usb tho? 2021-03-27 03:14:01 bc. if it will not work on some bloat, it will hardly work on tiny where u miss 90% of fuckery :/ 2021-03-27 03:14:25 openwrt is only used for routers isn't it? 2021-03-27 03:15:38 exactly right solution for your machine :D 2021-03-27 08:43:51 Hi there! I was under the impression that "apk upgrade && lbu commit -d" would save those version upgrades to disk. However after a reboot they are gone. Did I misunderstand something or is my system no configured correctly? 2021-03-27 08:56:25 bl4ckb0ne: do we know which sata drivers/options we need to enable for helios64 2021-03-27 08:56:39 and ethernet driver 2021-03-27 08:57:02 well, all drivers needed for this SBC 2021-03-27 09:11:33 troethe: lbu by default back up all configs in /etc/* nothing more 2021-03-27 09:12:50 troethe: 'lbu ls' will show you all of files 2021-03-27 09:13:12 troethe: so if u want someting more or less u have to add/remove it 2021-03-27 09:38:21 helby: yea I read about that, however I still haven't really figured out where packages are saved on the system. Because for some reason, packages I add *will* persist a reboot in /var/cache/apk. Are they re-downloaded every time on boot, because they are listed in /etc/apk/world? 2021-03-27 09:46:03 troethe: yes, /etc/apk/world decides what packages are installed 2021-03-27 09:46:09 but they are not pinned there 2021-03-27 09:46:20 so a package upgrade does not result in a change of the system that lbu persists 2021-03-27 09:46:58 If you enabled caching of packages (/etc/apk/cache), then the upgraded packages should be store there 2021-03-27 09:53:43 ikke: Yes I do have the symlink /etc/apk/cache -> /var/cache/apk. However updates do not persist. 2021-03-27 09:55:17 ikke: Doesn't lbu follow symlinks? 2021-03-27 09:57:10 lbu is not meant to persist the actual packages 2021-03-27 09:57:17 they should be stored on persisted media 2021-03-27 09:57:41 so the default setup is that /etc/apk/cache points to a directory on a usb drive, sdcard or disk 2021-03-27 10:11:42 ikke: Oh I see, thank you. However there also seem to be some essential packages loaded from the SD card that I boot from. Like linux or busybox. Is there some canonical way to upgrade those directly? 2021-03-27 10:12:40 busybox is just a package, nothing special 2021-03-27 10:13:02 the kernel is present on the bootmedia, and requires to write a new / updated image to the boot media 2021-03-27 10:13:24 (eg. with setup-bootable 2021-03-27 10:18:01 yea nevermind the bit about the kernel, I mixed up something there. I guess my question was, how I could upgrade the packages in my boot image. But I guess I just can't/should't and I will do that via the package cache. Thanks again! 2021-03-27 10:20:16 apk would install the latest versions of packages on boot anyway, I believe 2021-03-27 10:20:23 the cache is just that, a cache 2021-03-27 10:29:07 idk it doesn't seem to update them for me. After every reboot, I have to manually upgrade the packages again. This is especially annoying, since I then also have to manually restart the upgraded services, like sshd. 2021-03-27 10:39:15 troethe: do you have also cache enabled? 2021-03-27 10:41:18 ah lol, just read it now 2021-03-27 10:42:37 I was trying yesterday to setup a tiniproxy for URL filtering and apk seems ignoring environment vars, is it supposed to work with an http proxy? 2021-03-27 10:50:28 donoban: it should 2021-03-27 10:51:53 ok 2021-03-27 10:52:07 probably I missconfigured something 2021-03-27 11:00:32 first problem was running apk with sudo, it doesn't load proxy.sh profile 2021-03-27 11:01:52 but now it gets stuck: "fetch http://alpine.42.fr/edge/main/x86_64/APKINDEX.tar.gz" while I can download it perfectly with firefox 2021-03-27 11:02:53 do you get an error or is it just hanging? 2021-03-27 11:04:52 stills hanging 2021-03-27 11:05:31 strace / tcpdump? 2021-03-27 11:05:50 let me check 2021-03-27 11:07:45 uhM 2021-03-27 11:08:57 there is some traffic 2021-03-27 11:10:23 but it starts using IPv6 2021-03-27 11:10:37 uHm 2021-03-27 11:10:41 I think it is stucked here 2021-03-27 11:10:45 let me tpaste 2021-03-27 11:12:04 https://tpaste.us/raQx (I'm inspecting only lo sice I'm logged with ssh and there is a lot of noise in wlan0) 2021-03-27 11:18:10 ouch, I filtered all ssh trafic and there is nothing when run apk update 2021-03-27 11:18:47 uhM, I tried to ron apk upgrade instead and it generated some DNS request 2021-03-27 11:19:11 maybe apk is trying to resolv names without the proxy? 2021-03-27 11:19:38 well is not DNS traffic really 2021-03-27 11:23:49 uhM, firefox worked but curl doesn't 2021-03-27 11:36:11 ey ikke I've fixed, it seems problem with tiny proxy 2021-03-27 11:36:21 aha, ok 2021-03-27 11:37:15 puf, I'm not sure now I've done what I chanved and stills working 2021-03-27 11:37:24 crazy computers 2021-03-27 11:38:31 uhM, if I don't specify a Listen interface 2021-03-27 11:39:05 I get: Unauthorized connection from "localhost" [::1] when trying to do apk update (but firefox works fine) 2021-03-27 11:39:59 but if I add "Listen 127.0.0.1" (which I wanted to do anyway), it works fine 2021-03-27 11:43:15 another problem is this warning message: "WARNING: logging deactivated (can't log to stdout when daemonized)" 2021-03-27 11:43:29 and the fact that the service appears as 'crashed' 2021-03-27 11:46:53 I fixed both problems editing 'tinyproxy.conf', should alpine package provide a configuration file that doesn't produce this errors? 2021-03-27 11:50:05 Would make sense 2021-03-27 11:52:02 what should be best, mail the package maintainer or opening an issue on gitlab? 2021-03-27 11:52:26 issue on gitlab 2021-03-27 11:52:33 ok, thanks ikke 2021-03-27 13:17:18 Hi all 2021-03-27 13:18:07 I have a question about qemu: it seems Alpine pkg is missing zstd compression support for qcow2 images, is it normal ? 2021-03-27 13:18:47 they should have been supported since qemu ver 5.1 2021-03-27 13:19:49 as reported here: https://www.qemu.org/2020/08/11/qemu-5-1-0/ 2021-03-27 13:23:40 I guess it's because we don't have zstd-dev as a makedepends 2021-03-27 13:23:47 (--enable-zstd) 2021-03-27 13:24:34 tnx, so a simple rebuild will do ? 2021-03-27 13:24:58 I would expect so 2021-03-27 13:28:50 trying now... 2021-03-27 13:47:41 qemu is huge beast to compile 2021-03-27 13:48:29 there is a way to build just the qemu-system-x86_64 binary and the qemu_img binary with aports ? 2021-03-27 13:48:48 the apkbuild lists what to build 2021-03-27 14:09:26 mps: ethernet is good iirc, sata is the JMB585 2021-03-27 14:09:54 eth is RTL8211F and RTL8156 2021-03-27 14:10:06 and usb hub is VL815 2021-03-27 14:12:03 yup it works by adding --enable-zstd to the APKBUILD file 2021-03-27 14:12:37 I think it should be added 2021-03-27 14:14:25 bl4ckb0ne: ok, will do later. now I'm working on some big payment system changes and installation for my $day_job 2021-03-27 14:15:25 sounds good 2021-03-27 14:20:01 huh, when running ping: sendto: invalid argument 2021-03-27 14:20:06 anyone a clue what that means? 2021-03-27 14:20:17 interface is up, but I cannot ping anything except 127.0.0.1 2021-03-27 14:23:12 hmmm maybe the jmb585 has something to do with uboot, since my sd card uses the generic rk3399 from pmos 2021-03-27 14:37:23 ikke: I think I got that when had blocked all in firewall 2021-03-27 14:37:54 no firewall 2021-03-27 14:39:06 Might have to do with the subnet. It's a /31 2021-03-27 14:39:18 But a similar setup works on a different machine 2021-03-27 14:44:47 arping works 2021-03-27 15:25:59 bl4ckb0ne: I think there is no jmb585 driver in kernel 2021-03-27 15:26:26 it works if I just use 1 interface instead of bonding :( 2021-03-27 15:28:40 also 'find drivers/net -name Kconfig | xargs grep -i RTL82' shows nothing 2021-03-27 15:37:52 bl4ckb0ne: aha, drivers/net/phy/realtek.ko:RTL8211F Gigabit Ethernet 2021-03-27 17:22:14 yeah idk where the JMB585 driver is, cant find it either 2021-03-27 17:30:21 bl4ckb0ne: so we don't know which sata driver to enable 2021-03-27 17:35:32 jmicron 2021-03-27 17:43:43 bl4ckb0ne: https://dev.alpinelinux.org/~mps/linux-edge-5.11.10-r3.apk 2021-03-27 17:44:00 blindly enabled some sata drivers and some ethernet 2021-03-27 17:58:21 is there a way how to open firefox tab from let's say terminal instead of getting error about already running firefox? without using dbus or session greeter and other nonsense? 2021-03-27 18:01:23 hi 2021-03-27 18:01:46 leave while u can 2021-03-27 18:01:58 i'm trying to compile gcc and getting the following error: https://pastebin.com/QCfafBEH 2021-03-27 18:02:09 linux/if_ppp.h exists, but not net/if_ppp.h 2021-03-27 18:02:41 /usr/bin/firefox --new-tab %s 2021-03-27 18:06:38 mps: it does not work, it always just shows that firefox error pop 2021-03-27 18:06:49 probably not possible with some kind of session 2021-03-27 18:07:15 ... in wayland/sway 2021-03-27 18:07:45 this is in my urlview for mutt, X and awesome wm 2021-03-27 18:11:24 yes, even running just 'firefox' will work normally if u are in user session 2021-03-27 18:12:04 if run sway with dbus-session-run it works, but why I would run dbus :/ 2021-03-27 18:12:49 I don't use any login/display manager or that kind of 'addons' ;) 2021-03-27 18:13:42 I recall it worked without dbus when I had patience to rebuild firefox without dbus and other useless/bad features 2021-03-27 18:14:04 that was probably cool week :/ 2021-03-27 18:14:40 months :) 2021-03-27 18:15:01 btw. sndio is disabled in firefox package? 2021-03-27 18:15:16 but I'm loosing battles and war against featurists 2021-03-27 18:15:23 yes 2021-03-27 18:15:52 if webkit works for me I would probably use that shit instead fire, but that shit perfom like ... 2021-03-27 18:15:58 even videos are freezing 2021-03-27 18:16:18 "мало руках малена и снага, сува сламка међу вихорове" :) 2021-03-27 18:20:32 I agree 2021-03-27 18:20:52 you translated it? 2021-03-27 18:21:17 nope, but it looks cool 2021-03-27 18:21:27 it is true I bet 2021-03-27 18:29:53 i dont see the eth interfaces 2021-03-27 18:30:28 which kernel 2021-03-27 18:30:46 5.11.10 r3 2021-03-27 18:31:02 its listed as `5.11.10-2-edge` tho 2021-03-27 18:31:10 does lsmod shows any realtek module loaded 2021-03-27 18:31:33 nothing 2021-03-27 18:32:04 no rtl ones or no modules 2021-03-27 18:32:14 no modules 2021-03-27 18:32:19 gonna insmod it 2021-03-27 18:33:07 add 'rc-update add hwdrivers sysinit' 2021-03-27 18:35:26 still no rtl modules but there's some 2021-03-27 18:36:55 helby: "little hands and little power, dry straw among the Whirlwinds" 2021-03-27 18:37:54 bad google translate, but better than nothing 2021-03-27 18:38:25 bl4ckb0ne: do you disks 2021-03-27 18:38:29 interfaces 2021-03-27 18:38:47 the sata disks? 2021-03-27 18:38:55 yes 2021-03-27 18:38:59 nothing 2021-03-27 18:39:06 i wonder if the kernel is instaleld correctly 2021-03-27 18:39:22 do you have /sys fs 2021-03-27 18:39:51 yep 2021-03-27 18:40:07 what shows 'rc-update' 2021-03-27 18:40:24 https://paste.sr.ht/~bl4ckb0ne/df20b9b498ed6a3f9f00d6741c7b8631ebdee620 2021-03-27 18:42:42 hmm, looks enough 2021-03-27 18:43:19 except I have modules twice 'modules | boot sysinit' 2021-03-27 18:43:45 btw, you don't have udev running 2021-03-27 18:44:43 strictly speaking udev is not needed, but could help in debugging problem 2021-03-27 18:45:44 i also dont have a fstab, lots of warning during init process 2021-03-27 18:46:34 yes, I saw 2021-03-27 19:06:34 'rolling edge' and still 2 major versions behind with firefox 2021-03-27 19:06:41 I think I will ask for money back 2021-03-27 19:09:19 helby: why didn't you supply a merge request? :) 2021-03-27 19:10:27 ikke: if I start to supply merge reguests, alpine would die sooner Biden 2021-03-27 19:10:34 than 2021-03-27 19:11:40 ... probably nobody using firefox, all of users are on windows and mac using virtual docker kuber and other nonsense 2021-03-27 19:12:11 hehe, https://flathub.org/apps/details/org.mozilla.firefox 2021-03-27 19:12:23 and -esr is also behind? 2021-03-27 19:12:41 why would anyone use esr? 2021-03-27 19:12:42 I installed it for netflix, but I also use the musl version 2021-03-27 19:12:58 well flat... I would rather use win really 2021-03-27 19:13:04 i mean who the hell wants all the crap mozilla is adding to the lastest firefox? they only add cruft and destroy good stuff. 2021-03-27 19:13:21 good? what's good there? 2021-03-27 19:13:31 mozilla is successfully waging war against its users 2021-03-27 19:13:33 I don't knwo what differences currently are between esr and last 2021-03-27 19:13:40 not much good stuff left, 2021-03-27 19:13:46 true 2021-03-27 19:13:46 is the sabotage palemoon static binary? 2021-03-27 19:13:55 esr was not a agreement with debian for avoid all the iceweasel...? 2021-03-27 19:14:01 but still firefox is usable without too much google crap 2021-03-27 19:14:11 webkit shit does not work me on this machine well 2021-03-27 19:14:47 midor? 2021-03-27 19:14:57 all webkit based somehow perform so badly, 2021-03-27 19:15:21 utube video even freeze and site loads so long 2021-03-27 19:15:39 helby: what do you see so bad with flatpak? 2021-03-27 19:16:04 donoban: I don't use linux/bsd that I install all kind of bullshit inside :/ 2021-03-27 19:16:09 why you watch youtube in a brwoser, mpv/yt-dl is meant for watching videos, not browsers... 2021-03-27 19:16:30 g8U4Z1xGzjlm: then why I would need a browser at all 2021-03-27 19:16:57 but at least is an isolated shit 2021-03-27 19:17:06 text web works well on lynx too, so if this monster can't handle video, thant it's ... 2021-03-27 19:17:24 donoban: but why to isolate it? 2021-03-27 19:17:38 browser never took my file somewhere 2021-03-27 19:18:24 hehe because it's bullshit no? 2021-03-27 19:18:42 flatpacks? obviously 2021-03-27 19:18:53 I see a nice alternative for using alpine as a desktop 2021-03-27 19:19:29 gtk-straw-viewer 2021-03-27 19:19:39 or straw-viewer 2021-03-27 19:19:57 apk add straw-viewer 2021-03-27 19:20:32 yes, it's only 50MB 2021-03-27 19:20:35 ytfzf 2021-03-27 19:20:37 perling 2021-03-27 19:20:46 looks nice 2021-03-27 19:21:17 Deprecated in favor of: pipe-viewer. (info) 2021-03-27 19:21:50 donoban: I know 2021-03-27 19:22:05 ah ok 2021-03-27 19:22:14 just seeing their github repository 2021-03-27 19:22:35 plan is to change it before next alpine release 2021-03-27 19:22:52 if I don't forget :) 2021-03-27 19:23:08 hehe, well, if it stills working ^^ 2021-03-27 19:23:34 straw works fine 2021-03-27 19:28:01 mps: could you put CONFIG_REALTEK_PHY=y in the next kernel build pls? 2021-03-27 19:28:24 its for the RTL8211F 2021-03-27 19:29:13 hi, use somebody alpine-linux as desktop? if yes how big is the dektop env? 2021-03-27 19:30:48 i think RTL8156 is not yet mainline, there's a patch for it in armbian 2021-03-27 19:31:11 bubblegum: as big as the packages you install, there is no desktop env, just a bunch of packages you have to pick yourself 2021-03-27 19:31:12 bl4ckb0ne: sure 2021-03-27 19:31:15 thanks 2021-03-27 19:31:21 gonna search more for the jmb858 2021-03-27 19:32:57 ok 2021-03-27 19:33:42 maybe it is related to some usb-c options 2021-03-27 19:34:09 from the diagram it seems to be pcie 2021-03-27 19:34:09 or PCIe 2021-03-27 19:34:16 heh 2021-03-27 19:34:26 but yeah the rtl8156 is somehow linked to the usb hub 2021-03-27 19:34:29 VL815 2021-03-27 19:35:33 https://cateee.net/lkddb/web-lkddb/BLK_DEV_JMICRON.html found this tho 2021-03-27 19:35:38 PCIE_ROCKCHIP_HOST=y 2021-03-27 19:35:43 it is enabled 2021-03-27 19:36:06 heck dont have lspci 2021-03-27 19:36:15 :) 2021-03-27 19:36:46 found this fwiw https://www.jmicron.com/file/download/996/JMB585.pdf 2021-03-27 19:37:36 is ahci enabled? 2021-03-27 19:37:42 yes 2021-03-27 19:41:41 huh, JMicron JMB36x support is in drivers/ide and not drivers/ata 2021-03-27 19:41:47 strange 2021-03-27 19:41:54 found some doc https://wiki.kobol.io/helios64/sata/ 2021-03-27 19:42:26 yes, I have this in FF tab 2021-03-27 19:42:40 kobol doc is piss poor 2021-03-27 19:43:06 agree 2021-03-27 19:44:49 can lspci give the driver in use? 2021-03-27 19:44:54 armbian recognizes it 2021-03-27 19:45:35 -k 2021-03-27 19:45:57 https://paste.sr.ht/~bl4ckb0ne/06decdcf1cb2313a747d429de6d32b40152196c2 2021-03-27 19:45:58 ahci 2021-03-27 19:46:48 heh, it is there 2021-03-27 19:47:11 whack 2021-03-27 19:47:22 lsblk ? 2021-03-27 19:47:39 found the other one for the rtl 2021-03-27 19:47:42 > /lib/modules/5.10.16-rockchip64/kernel/drivers/net/usb/r8152.ko 2021-03-27 19:47:57 yup they are showing on armbian 2021-03-27 19:48:33 heh, I think I have this eth on usb dongle 2021-03-27 19:49:43 yes, /lib/modules/5.11.8-0-gru/kernel/drivers/net/usb/r8152.ko 2021-03-27 19:49:54 rockchip-pcie ? 2021-03-27 19:50:12 no, it is in usb dongle 2021-03-27 19:50:31 https://paste.sr.ht/~bl4ckb0ne/82ba262c66f71a6669774b0afaea98bf51552894 2021-03-27 19:50:38 (for the jmb585) 2021-03-27 19:50:49 but look above, according you armbian have it at same place 2021-03-27 19:50:55 yeah 2021-03-27 19:52:00 so, you have all needed drivers, just have to find why they don't work 2021-03-27 19:52:36 and I know that r8152.ko is somewhat stubborn to work out of the box 2021-03-27 19:53:10 is CONFIG_ROCKCHIP_PHY enabled? 2021-03-27 19:53:45 CONFIG_PHY_ROCKCHIP_PCIE also 2021-03-27 19:53:52 yes 2021-03-27 19:53:57 and CONFIG_PHY_ROCKCHIP_USB 2021-03-27 19:54:21 CONFIG_PCIE_ROCKCHIP too, there's one without _HOST 2021-03-27 19:55:23 grep $what_I_want_to_see /boot/config-edge 2021-03-27 19:56:01 i was on armbian, gonna reboot and grep 2021-03-27 19:56:08 ah, ok 2021-03-27 19:56:30 but, fyi CONFIG_ROCKCHIP_PHY=y 2021-03-27 19:57:09 rockchip usb is m 2021-03-27 19:57:41 yes, but this is not important 2021-03-27 19:58:18 if you boot it will be autoloaded by hwdrivers if it is needed 2021-03-27 19:58:24 everything is y 2021-03-27 19:58:27 whack 2021-03-27 19:59:07 can you look /sys/block dirs when you boot linux-edge 2021-03-27 19:59:42 > mmcblk1 mmcblk2 mmcblk2boot0 mmcblk2boot1 2021-03-27 19:59:49 hmm 2021-03-27 19:59:57 something is missing 2021-03-27 20:01:35 did you got answer to mail from kobol 2021-03-27 20:01:56 i forgot to mail them after the success last night 2021-03-27 20:02:05 aha 2021-03-27 20:02:17 so we need schematics and kernel config for sata? 2021-03-27 20:02:28 no 2021-03-27 20:02:45 we need minimaly needed kernel options 2021-03-27 20:03:20 schematic? if you want it sure. I wouldn't waste time on it 2021-03-27 20:03:30 i think RootWyrm wanted to take a look at it 2021-03-27 20:03:52 I'm not against :) 2021-03-27 20:06:00 https://github.com/armbian/build/blob/master/config/boards/helios64.wip found this fwiw 2021-03-27 20:06:21 what are u two building? alpine wasching machine? 2021-03-27 20:06:32 https://github.com/armbian/build/blob/eefad69215557708b151a5d9244617a4ffd1281c/patch/u-boot/u-boot-rockchip64-mainline/add-board-helios64.patch#L2026 2021-03-27 20:06:38 alpine on the helios64 2021-03-27 20:08:25 i dont see CONFIG_AHCI_PCI in the edge conf 2021-03-27 20:08:40 oooh that uboot config 2021-03-27 20:08:49 i still havent put the proper uboot config on the sd card 2021-03-27 20:08:53 gonna try to do that tonight 2021-03-27 20:09:45 ok 2021-03-27 20:10:31 with u-boot we can't do anything, except to build it locally with patches 2021-03-27 20:10:52 sounds like ipxe 2021-03-27 20:10:53 lol 2021-03-27 20:11:10 but I will not add these patches to alpine u-boot without really strong reasons 2021-03-27 20:11:24 99% of the time to fix ipxe... you recompile it. 2021-03-27 20:11:28 D: 2021-03-27 20:11:29 i agree with you 2021-03-27 20:12:35 good, going to drink wine, cheers to you and care for your little :) 2021-03-27 20:13:00 boy or girl? 2021-03-27 20:13:03 girl 2021-03-27 20:13:06 18 months 2021-03-27 20:13:11 nice :) 2021-03-27 20:13:50 I like people who cares for children 2021-03-27 20:14:11 which cares? 2021-03-27 20:14:24 huh, english /o\ 2021-03-27 20:15:07 zfs on armbian is so borked 2021-03-27 20:21:18 > and have 5748 and 1 different commits each, respectively. 2021-03-27 20:21:27 kobol-io/uboot is embarassingly late 2021-03-27 20:24:26 I wouldn't use zfs for anything serious 2021-03-27 20:24:48 what do you use instead? 2021-03-27 20:24:52 fat32? :P 2021-03-27 20:25:09 kobol.io should send one board to me 2021-03-27 20:26:25 heck they dont sell the board alone 2021-03-27 20:26:38 maybe I have one external usb sata case/adapter with jmicron interface 2021-03-27 20:27:02 I have to look in garage 2021-03-27 20:27:38 will try to connect it to my rk3399 to see will it work 2021-03-27 21:19:41 why is alpine linux not so popular for desktop use? 2021-03-27 21:20:41 1) that was not the main thing it was developed for. 2) desktop environments are complex and cost a lot of time and energy to setup 2021-03-27 21:20:44 i typically have only used it as a base image for docker .. wanted to practice some networking configs and had to spin up a few VMs so naturally i was looking for somehting lightweight .. decided to go with alpine.. as I use it i am falling in love with apk and cannot understand why this is not a popular option for desktop 2021-03-27 21:20:54 ah gotcha 2021-03-27 21:20:57 as a result, people need to setup a lot of things manually 2021-03-27 21:20:59 so lack of DE support for alpine? 2021-03-27 21:21:17 We do have the popular DEs 2021-03-27 21:21:25 my initial install is running at 50MB ram lol ! 2021-03-27 21:21:28 KDE, Gnome, and others 2021-03-27 21:21:37 something like 350MB on the drive 2021-03-27 21:21:41 but people have to manually set things up 2021-03-27 21:21:47 i see 2021-03-27 21:21:49 so it's not popular 2021-03-27 21:21:59 what would be an example of manual setup? 2021-03-27 21:22:10 zfs on linus surely works with alpine right? I don't see why not 2021-03-27 21:22:10 installing the correct packages 2021-03-27 21:22:13 linux* 2021-03-27 21:22:34 gotcha 2021-03-27 21:22:54 configuration, making sure the right services are started 2021-03-27 21:23:33 im going to make this the default distro for my rpi stuff as I don't care about DE there 2021-03-27 21:23:51 what about general server use? 2021-03-27 21:24:10 should be fine 2021-03-27 21:24:20 i see that things like docker are not in the package repo ... but nothing is stoppping you from git clone and building 2021-03-27 21:24:29 it's in community 2021-03-27 21:24:38 https://pkgs.alpinelinux.org/packages?name=docker&branch=edge 2021-03-27 21:25:03 most popular software is packaged 2021-03-27 21:26:00 this is really making me rethink some of my previous decisions 2021-03-27 21:26:21 i run some ubuntu servers but they are merely hosts and everything is dockerized .. the rest is just firewall rules etc 2021-03-27 21:27:19 We run services in docker on alpine ourselves 2021-03-27 21:27:42 how do you go about adding the community repo? 2021-03-27 21:27:54 add it to /etc/apk/repositores 2021-03-27 21:27:57 add it to /etc/apk/repositories 2021-03-27 21:28:03 ty 2021-03-27 21:28:23 just fyi, community only has about half a year of support as opposed to 2 years for main 2021-03-27 21:29:00 does that mean i should update more often? 2021-03-27 21:29:12 or the chance of something going out of support sooner is bigger 2021-03-27 21:29:28 hoestly i fon't mind compiling things like docker from scratch 2021-03-27 21:29:44 it basically means that we do not automically backport security / backport fixes 2021-03-27 21:30:10 bugfixes* 2021-03-27 21:30:15 ah so in order to get the latest patches you want to be a version within the 6 month window of latest 2021-03-27 21:30:34 actually that confuses me too 2021-03-27 21:30:45 i don't think i fully grasp the support concept on a rolling distro 2021-03-27 21:31:17 alpine is not a rolling distro 2021-03-27 21:31:25 only the edge repository is roll;ing 2021-03-27 21:31:35 we have releases every 6 months 2021-03-27 21:31:50 https://alpinelinux.org/releases/ 2021-03-27 21:31:56 ah gotcha that makes sense 2021-03-27 21:32:13 so 3.13 community is currently supported 2021-03-27 21:32:20 when 3.14 comes out, it's officially EOL 2021-03-27 21:33:15 apk.static upgrade is how to get on 3.14 when it comes out? 2021-03-27 21:33:34 would have been cool to have released 3.14 on 3/14 :P 2021-03-27 21:34:44 heh 2021-03-27 21:35:14 Generally for these releases, you do not need to use apk.static 2021-03-27 21:35:29 setup-apkrepos is enough? 2021-03-27 21:35:51 this is real nice though "apk.static upgrade --no-self-upgrade --available --simulate" 2021-03-27 21:35:55 You just change 3.13 to 3.14 in /etc/apk/repositories, and run apk ugprade -Ua 2021-03-27 21:36:21 i love the simplicity 2021-03-27 21:36:34 i don't know what i have been doing all this time 2021-03-27 21:38:11 i do see some benchmarking comparisons where it seems to not perform better than ubuntu but that could be due to configuration i suppose 2021-03-27 21:38:50 and if I am using docker in general i wonder if those things even matter 2021-03-27 21:39:00 Yeah, one point is that we typically compile with in -Os, to reduce the size 2021-03-27 21:39:24 where -O2 has better performance in certain cases 2021-03-27 21:39:33 since docker will have it's own base image .. but in the end of the day it is running linux tricks on the host machine so it should amtter somewhat 2021-03-27 21:39:51 No, it should not matter 2021-03-27 21:40:03 i didn't set up manuall i ran setup-alpine i do remember there was an option when selecting sys/lvm for something that was typically used for databases 2021-03-27 21:40:10 it just sounded like a bare disk but i cant recall 2021-03-27 21:40:30 i mean shoot 2021-03-27 21:40:48 is there any downfall to swithching to alpine from ubuntu albeit a biased place to ask that question? lol 2021-03-27 21:40:59 i foresee a project coming in my horizon 2021-03-27 21:42:26 I would say: start with something small first and get a bit more familiar with things 2021-03-27 21:42:45 i guess i don't really know what a distro brings to the table as packages are packages .. any security patches etc are up to the 'thing' such as docker let's say to be updated and you just update your version on your system .. same with the kernel .. it is not like alpine is changing the kernel in any way is it? unless it's about kernel options 2021-03-27 21:43:35 We used to run a hardened kernel, but since grsec closed, we no longer have that 2021-03-27 21:43:47 We do have compiler hardening options 2021-03-27 21:44:11 running musl is also beneficial generally (apart from compatibility issues from time to time) 2021-03-27 21:44:25 the musl malloc implementation for example recently got hardened 2021-03-27 21:44:54 yeah im going to start small and change up my personal server its just a vpn, ssh with docker for the things i like to host the rest is firewall configs 2021-03-27 21:44:58 i think gentoo uses musl right? 2021-03-27 21:45:13 ther eis mention of emphasis on security with alpine what do they really mean? 2021-03-27 21:45:26 mostly what I mentioned 2021-03-27 21:45:35 reducing attack surface 2021-03-27 21:45:40 gotcha 2021-03-27 21:46:11 hardened kernels really are about reducing attack surface NOT on the networking layer right? 2021-03-27 21:46:41 or could some decisions have implications on how low level networking works 2021-03-27 21:46:56 I would not expect so 2021-03-27 21:46:58 but Not sure 2021-03-27 21:47:22 hey thanks a lot for the insight im going to give it some thought 2021-03-27 21:47:24 i like lean and mean 2021-03-27 22:06:42 i must say though apk is faaaaaast 2021-03-27 22:07:12 im pretty convince i don't it doesn't just "feel that way" 2021-03-27 22:07:33 it could be that my default mirror is closer than what I have for other arch/ubuntu systems but man it feels really fast 2021-03-27 22:09:36 apk streams everything, which contributes to it's speed 2021-03-27 22:09:43 no separate download phase 2021-03-27 22:09:59 like fall in love, but after some time 'thing changes' :) 2021-03-27 22:10:00 oh wow 2021-03-27 22:12:33 do folks typically go Xorg if they try to use alpine for desktop use? 2021-03-27 22:15:13 some use xorg, others use wayland 2021-03-27 22:24:16 i'm trying to compile gcc and getting the following error: https://pastebin.com/QCfafBEH 2021-03-27 22:24:26 linux/if_ppp.h exists, but not net/if_ppp.h 2021-03-27 22:27:20 gcc 10.2 fails with a different error, btw: https://pastebin.com/vfk4yRBf 2021-03-28 13:05:55 i'am dump or does apk & https urls in an alpine:edge docker container didn't work? 2021-03-28 13:06:09 weiti: it does work 2021-03-28 13:06:30 weiti: you might be running into seccomp issues 2021-03-28 13:06:41 ikke: hmm ... then it must be in combination with drone ci ... 2021-03-28 13:06:59 what arch? 2021-03-28 13:07:26 x86_64 2021-03-28 13:07:42 what is the error you are getting? 2021-03-28 13:08:27 140693287865160:error:1416F086:SSL routines:tls_process_server_certificate:certificate verify failed:ssl/statem/statem_clnt.c:1913: 2021-03-28 13:08:36 and ERROR: https://dl-cdn.alpinelinux.org/alpine/edge/main: Permission denied 2021-03-28 13:09:06 if i change the repo urls to http it works in the same setup (sed -i 's/https/http/' /etc/apk/repositories) 2021-03-28 13:25:53 Anyone an idea why I get sendto: invalid argument for ping when using an LACP bond? THe same ip/subnet/gateway combination works when just configuring a single interface. I can ping 127.0.0.1, but nothing else. 2021-03-28 13:26:14 A similar setup works on another host 2021-03-28 13:27:53 I have no clue how to troubleshoot this. 2021-03-28 13:28:34 (strange thing is that the same configuration did work in a rescue os version of alpine 2021-03-28 13:29:27 does it also not work with only one interface in the bond? 2021-03-28 13:30:18 nope 2021-03-28 13:33:28 arping does work 2021-03-28 13:33:35 (to the default gateway) 2021-03-28 13:40:55 example of a host where it works: 2021-03-28 13:40:58 bond0: mtu 1500 qdisc noqueue state UP qlen 1000 2021-03-28 13:41:00 doesn 2021-03-28 13:41:03 doesn't work: 2021-03-28 13:41:14 bond0: mtu 1500 qdisc noqueue state UP mode DEFAULT group default qlen 1000 2021-03-28 13:42:58 and how does /proc/net/bonding/bond0 looks like? Different on the hosts? 2021-03-28 13:45:41 Working: Ethernet Channel Bonding Driver: v3.7.1 (April 27, 2011) 2021-03-28 13:45:57 Not working: Ethernet Channel Bonding Driver: v5.10.26-0-lts 2021-03-28 13:46:04 different bonding driver it seems 2021-03-28 13:50:21 Trying an older kernel, see if it makes a difference 2021-03-28 13:53:27 yup, old kernel works 2021-03-28 13:53:34 oh, no, wait :_) 2021-03-28 13:53:59 first need to switch back to bond 2021-03-28 14:02:50 older kernel works, but I had to manualy setup the bonding 2021-03-28 14:03:53 strange 2021-03-28 14:07:20 had only some 4.19 kernels on bonding machines ... 2021-03-28 14:07:49 5.4 is working, 5.10 not 2021-03-28 15:22:44 intent alpine to force https on the repository in the near future and disable http? :) 2021-03-28 15:28:43 no 2021-03-28 16:37:31 Hello all, just a quick question: Are the "Power Management" options in KDEs settings in 3.13.3 supported? I'm getting the error "Power Management configuration module could not be loaded. The Power Management Service appears not to be running." Or is the plasma-metapackage just missing a dependency? 2021-03-28 16:38:21 PureTryOut[m]2: ^ 2021-03-28 16:39:03 is this https://wiki.alpinelinux.org/wiki/Alpine_Linux_in_a_chroot the best way for build alpinelinux and break things? 2021-03-28 16:39:46 pingberlin: I think that it it's working fine on my edge install, let me check it 2021-03-28 16:40:57 uhM it works fine but I installed whole kde 2021-03-28 16:44:02 pingberlin: check if you have installed powerdevil pacakge 2021-03-28 16:45:09 uh 2021-03-28 16:45:17 it should be installed by both plasma/plasma-desktop 2021-03-28 17:03:46 pingberlin: ah there is a bug which breaks powerdevil sadly. I made a workaround on edge but seems I forgot to do the same on stable. Will push a fix in a bit 2021-03-28 17:06:25 ey PureTryOut[m]2 , I have frequent segfaults with kwin_wayland, I installed corectl for try to get a coredump, but I only get dumps from other process 2021-03-28 17:06:34 do you know what could I do for debug it? 2021-03-28 17:07:57 Attach to it with gdb, make sure you have debug symbols installed, and get a backtrace 2021-03-28 17:08:42 ok, I will try, it seems easy to crash it 2021-03-28 17:09:44 only need kwin-dbg-5.21.3-r2 ? 2021-03-28 17:20:50 For debug symbols yes 2021-03-28 17:21:18 And yeah the Wayland session is a bit too buggy for me to use atm. I can't even log out as kwin_wayland just crashes and then relaunches and throws you back in the session 2021-03-28 17:21:22 I try to run: medion:~/alpine$ sudo -E apk add -p /home/donoban/alpine/edge/ --initdb $(cat /etc/apk/world) and only get a lot of (no such package) errors 2021-03-28 17:34:44 ladies I have bad news :/ 2021-03-28 17:34:57 maintainer of firefox probably died 2021-03-28 17:35:08 :\ 2021-03-28 17:35:10 ... or maybe in a jail :/ 2021-03-28 17:35:29 Rasmus Thomsen? 2021-03-28 17:35:52 (that was sad joke) 2021-03-28 17:36:04 what? 2021-03-28 17:37:55 ? 2021-03-28 17:41:08 don't worry, nobody died ... it's just what my firefox is whispering me 2021-03-28 17:43:32 uhM, I get always this 2021-03-28 17:43:47 #0 __cp_end () at src/thread/x86_64/syscall_cp.s:29 2021-03-28 17:43:49 29 in src/thread/x86_64/syscall_cp.s 2021-03-28 17:43:58 is n default breakpoint? 2021-03-28 17:44:28 well, it crashed but probably I caused it 2021-03-28 17:50:14 Thread 1 "kwin_wayland" received signal SIGSEGV, Segmentation fault. 2021-03-28 17:50:15 0x00007f024bf21a7e in ?? () from /usr/lib/libKWaylandServer.so.5 2021-03-28 17:50:17 (gdb) 2021-03-28 17:52:32 PureTryOut[m]2: could I get something more useful? https://tpaste.us/R4y1 2021-03-28 18:02:06 this has more symbols, https://tpaste.us/9jv1 2021-03-28 18:21:05 Thanks PureTryOut[m]2 and donoban for checking! :) 2021-03-28 19:07:02 weiti: huh, I now suddenly get that same error outside of docker 2021-03-28 19:07:17 oh, wrong datetime 2021-03-28 19:09:46 hi 2021-03-28 19:10:07 i have no mouse or keyboard when i startx, but when i sudo startx i do 2021-03-28 19:10:17 i'm in the wheel group 2021-03-28 19:10:38 you need to be in the input group 2021-03-28 19:10:43 that's it? 2021-03-28 19:10:45 and probably video as well 2021-03-28 19:11:07 usually that's the issue 2021-03-28 19:11:13 especially because it works when you use sudo 2021-03-28 19:11:17 you're awesome 2021-03-28 19:11:22 it worked 2021-03-28 19:11:46 i'll be back in a minute 2021-03-28 19:11:50 thanks 2021-03-28 19:14:46 somehow can't get to work seatd with sway 2021-03-28 19:15:20 i thought arch was light... this is better 2021-03-28 19:15:26 not sure what I have to set up as I assume it should work if seatd is running 2021-03-28 19:19:31 as without it, that shit is run by root 2021-03-28 19:21:34 ... just saying obviously 2021-03-28 20:01:13 Hi, Alpine Linux is in serious need of documentation. I would like to help with the user and developer handbooks https://docs.alpinelinux.org/user-handbook/0.1a/index.html , can someone point me to how to contribute? 2021-03-28 20:01:43 https://gitlab.alpinelinux.org/alpine/docs 2021-03-28 20:09:56 oooo a call to make a new init systm 2021-03-28 20:10:38 thamk ikke 2021-03-28 20:12:52 L1Cafe: there is also #alpine-docs 2021-03-28 20:13:05 roger roger 2021-03-28 20:13:14 I'll be drafting stuff and sending it your way 2021-03-28 20:13:23 Who can review drafts? 2021-03-28 20:15:34 I can, some in #alpine-docs 2021-03-28 20:17:59 Would docs for writing OpenRC units belong in developer or user handbook? 2021-03-28 20:26:27 does somebody know if the faccessat2 issue within docker also affects current versions of docker for mac? 2021-03-28 20:27:05 if i use alpine:edge i'm always getting Operation not permitted for various commands :( 2021-03-28 20:28:37 espeacially if i want to execute make tasks 2021-03-28 20:30:15 right, that's indeed the faccessat2 issue 2021-03-28 20:30:35 afaik, docker on mac is using a vm, right? 2021-03-28 20:31:40 yes, docker for mac is using a vm 2021-03-28 20:34:19 that's correct ;) 2021-03-28 20:35:09 L1Cafe: probably developers handbook, but, we should not repeat basic openrc documentation 2021-03-28 20:36:13 ikke: What do you mean by "repeat basic OpenRC documentation". Do you mean stuff like what's already on the user handbook for starting and stopping? 2021-03-28 20:36:56 L1Cafe: things like documented in https://github.com/OpenRC/openrc/blob/master/service-script-guide.md 2021-03-28 20:37:02 https://github.com/OpenRC/openrc/blob/master/supervise-daemon-guide.md 2021-03-28 20:38:01 ikke: I see. What's the rationale behind this? Projects like Red Hat Linux include documentation for all system tools, even when it overlaps with upstream docs 2021-03-28 20:38:18 L1Cafe: duplication of efforts, maintenance 2021-03-28 20:38:21 I find it a bit weird that, when Alpine is a single-init distro, you have to go somewhere else to get the docs 2021-03-28 20:38:28 But that's fair 2021-03-28 20:38:51 https://cloud.drone.io/promhippie/github_exporter/107/1/2 looks like the hosted drone ci instance also doesn't provide a proper seccomp profile :( 2021-03-28 20:39:13 L1Cafe: chance is that we will switch to a new init system in the futue 2021-03-28 20:39:19 future 2021-03-28 20:39:26 #alpine-devel 2021-03-28 20:39:27 What is it going to be? 2021-03-28 20:39:46 L1Cafe: something that is still being developped, but probably based on s6 2021-03-28 20:40:17 ikke: An Alpine fork, or a vanilla upstream project? and why is this change being done? 2021-03-28 20:40:39 s6 is great :) 2021-03-28 20:40:42 https://ariadne.space/2021/03/25/lets-build-a-new-service-manager-for-alpine/ 2021-03-28 20:40:55 still using it within some docker images, but getting rid of it within docker 2021-03-28 20:42:06 wait, wait 2021-03-28 20:42:10 s6 is fantastic, I was really happy to hear Alpine seems to be moving in that direction 2021-03-28 20:42:12 kaniini is ariadne? 2021-03-28 20:42:15 yes 2021-03-28 20:42:18 Ha 2021-03-28 20:42:21 I know Kaniini 2021-03-28 20:42:31 oh no 2021-03-28 20:42:32 From something unrelated, Pleroma 2021-03-28 20:42:39 i quit that 2021-03-28 20:42:42 I am aware. 2021-03-28 20:44:50 I have a problem with accents, keyboard seems fine but it writes the accents before the letter 2021-03-28 20:44:57 e.g. ´a 2021-03-28 20:45:09 does somebody know if github actions provide a proper seccomp profile for alpine:edge? 2021-03-28 20:45:35 tboerger: probably not :( 2021-03-28 20:45:47 ikke, Ariadne : Is there a todo list of stuff Alpine needs help with? 2021-03-28 20:45:54 ok... time to setup my own ci system again -.- 2021-03-28 20:46:03 tboerger: though in my experience, alpine does work fine in github actions 2021-03-28 20:46:11 tboerger: not aware 2021-03-28 20:46:37 Ariadne: have you also tried edge? :) 2021-03-28 20:46:53 tboerger: no, only latest-stable 2021-03-28 20:47:44 if 3.14 works github actions should be fine 2021-03-28 20:47:49 looks like i got to try it 2021-03-28 20:49:23 don't wanna run drone on my own anymore and i don't think that cloud.drone.io will get any update related to that. 2021-03-28 20:49:44 L1Cafe: what is "single-init" 2021-03-28 20:50:11 L1Cafe: aports issues 2021-03-28 20:50:26 for longer term projects i don't think we have a specific place 2021-03-28 20:50:55 Hello71: I meant like other projects (I think Void) allow you to choose your init 2021-03-28 20:51:08 ping: sendto: Network unreachable 2021-03-28 20:51:09 Alpine is OpenRC first 2021-03-28 20:56:28 ikke: wrong channel? 2021-03-28 20:56:40 yes 2021-03-28 20:57:15 probably wrong term entirely :p 2021-03-28 20:57:22 i hate doing that 2021-03-28 20:57:29 especially when its irc 2021-03-28 20:57:30 haha 2021-03-28 20:58:07 no, it was meant for #alpine-devel 2021-03-28 20:58:16 ah 2021-03-28 20:58:21 man i gotta figure this out, dhcp works fine on my network... except for windows clients. 2021-03-28 20:59:01 https://ariadne.space/2021/03/23/why-rms-should-not-be-leading-the-free-software-movement/ nice post 2021-03-28 20:59:12 that is just my opinion 2021-03-28 20:59:17 not the opinion of alpine 2021-03-28 20:59:23 ah well 2021-03-28 20:59:36 if that letter has anything to do with ANYTHING other than his views on software 2021-03-28 20:59:43 i might sperg 2021-03-28 21:00:18 oh cool more CoC nonsense 2021-03-28 21:00:24 ACTION rolls eyes and continues moving on 2021-03-28 21:00:47 RMS is incompetent and I wish it took more than a few crocodile tears to get that opinion 2021-03-28 21:00:59 fys: for what it's worth, i would have personally composed a different letter that covered *all* the bases 2021-03-28 21:01:09 i just want competent leadership at FSF 2021-03-28 21:01:15 I'm sure you do. 2021-03-28 21:01:28 But as someone who has hated the man since the late 90s. 2021-03-28 21:01:30 I'm switching my desktop distribution (and my few time and energy for colabotion) because I a pretty lack of empathy 2021-03-28 21:01:40 And is sick of CoC bullshit at the same time. 2021-03-28 21:01:49 because I felt* 2021-03-28 21:01:54 I find this recent RMS outcry annoying on SOOOOOOO many levels. 2021-03-28 21:02:16 --> #alpine-offtopic --> 2021-03-28 21:02:21 Hello71: Was about to ask that 2021-03-28 21:02:23 :) 2021-03-28 21:02:24 yes, please take this to -offtopic :) 2021-03-28 21:02:35 I wasn't the one who posted it. 2021-03-28 21:02:38 hehe 2021-03-28 21:02:58 i know, i am just saying this is not really an alpine user support discussion, so #alpine-offtopic is likely a better place for it 2021-03-28 21:03:01 sorry, just read the s6 post and continue reading the blog 2021-03-28 21:03:04 I don't enjoy the tone of force feeding an opinion to the channel and not allowing an opposing opinion. 2021-03-28 21:03:19 For the record. 2021-03-28 21:03:39 for the record, we should have asked donoban to discuss his opinion on #alpine-offtopic too :) 2021-03-28 21:03:39 I've said what I have to say. 2021-03-28 21:03:40 so except of being a male, what is the horrible thing RMS did except for promoting open source (and making Microsoft and Oracle want him killed)? 2021-03-28 21:03:41 Moving on. 2021-03-28 21:03:47 lets go ~offtopic channel :) 2021-03-28 21:03:56 Yes if we want to continue, let's go there./ 2021-03-28 21:04:04 i don't really want to continue 2021-03-28 21:04:18 i'm quite tired of this topic, really 2021-03-28 21:04:18 well that boat have sailed 2021-03-28 21:04:28 and it ended up in the suez canal ;) 2021-03-28 21:05:01 LOL 2021-03-28 21:05:11 sorry I really didn't refer to RMS itself, just I feel that alpine is a nice community, I hope don't break it with this topic :| 2021-03-28 21:05:18 speaking of which... would be fun if the bridge com is ever released from that fuckery 2021-03-28 21:05:40 "so ehm... captain?... yeah... so... there was this wind and ehm... we are now stuck here in the middle of the desert" 2021-03-28 21:06:17 docker have a new logotype ;) 2021-03-28 21:12:49 ... like fsf matters :/ 2021-03-28 21:13:13 It matters a LOT more than you think. 2021-03-28 21:13:41 Just because GNU doesn't prance around the stock exchange like RedHat doesn't mean they don't have an iron grip. 2021-03-28 21:13:43 FLM! 2021-03-28 21:14:36 imagine a world without alpine and other *nix dialects 2021-03-28 21:14:46 a world where backdoored windows is EVERYWHERE 2021-03-28 21:15:02 or only GNU is used .. 2021-03-28 21:17:21 still did not get that shit around fsf ... if that organization disappear tomorrow, nobody will notice 2021-03-28 21:31:16 sure you will 2021-03-28 21:35:40 for have an alpine-sdk chroot do I need to bind mount something? I created a chroot with chroot-install-script and now have a pretty crazy mount scheme :\ 2021-03-28 21:36:41 medion:~$ mount | wc 2021-03-28 21:36:42 65 390 6755 2021-03-28 21:37:05 is this not excesive? -_- 2021-03-28 21:37:56 ubuntu 20.10: 44 264 3700 2021-03-28 21:38:18 44 mounts 2021-03-28 21:38:32 do you have some chrooted system? 2021-03-28 21:42:46 and I have some strange things, like: 2021-03-28 21:42:54 /dev/mapper/root on /home/donoban/alpine/edge/home/donoban/alpine type btrfs (rw,relatime,ssd,space_cache,subvolid=262,subvol=/@home) 2021-03-29 05:31:02 ikke: so it was a time issue ? may i check the dates on the system or is it only a datetime issue outside of docker? 2021-03-29 05:32:26 weiti: yes, for me it was a time issue, but that resulted in that exact error 2021-03-29 05:33:53 281412551511112:error:1416F086:SSL routines:tls_process_server_certificate:certificate verify failed:ssl/statem/statem_clnt.c:1913: 2021-03-29 05:33:55 this one? 2021-03-29 05:39:21 yes, the same error 2021-03-29 06:30:14 hmmm ... but the docker host has correct time settings on my side ... strange. 2021-03-29 08:21:42 does `apk` allow breaking system updates? I assume not right since Alpine is a stable release? That is Arch Linux's pacman can do a partial system update and break, whereas Void Linux's XBPS will not allow a partial update which breaks the system. Alpine Linux's APK will not allow a partial update too? 2021-03-29 08:45:59 tsujp: that question is very strangely worded 2021-03-29 08:46:11 what are you wanting to actually know? 2021-03-29 08:47:29 I see an initial question followed by comments about other operating systems. 2021-03-29 08:47:39 So I guess I'll answer the relevant question I see. 2021-03-29 08:47:50 > does `apk` allow breaking system updates? 2021-03-29 08:48:03 Yes? No? Does your text editor let your write bad code? 2021-03-29 08:48:33 Where you define a package manager "allowing" a system breaking update? 2021-03-29 08:48:51 Are you referring to edge or a mainline version? 2021-03-29 08:49:28 Are you suggesting that a package manager somehow have temporal abilities to know if your update will fail? 2021-03-29 08:50:10 tsujp: apk works in 'all or nothing' mode in every one particular invocation 2021-03-29 08:51:00 mostly, tbh 2021-03-29 08:53:26 fys i think the question is pretty clear - will apk roll back changes if an update fails mid-way through? i.e. is it transactional 2021-03-29 08:53:49 Okay. 2021-03-29 08:53:53 I didn't get that. 2021-03-29 08:54:17 It was a very poorly worded question with lots of open-endedness. 2021-03-29 08:54:58 And I suspect intentionally so, tends to be the case when someone enters a distro channel and their first question involves another distro. 2021-03-29 09:06:15 I was thinking to setup a chroot installation for use abuild and try to fix some things but now I'm considering using docker images, I think that it would be cleaner and more flexible 2021-03-29 09:06:50 Probably. 2021-03-29 09:08:14 There is docker-abuild, or alpinelinux/build-base 2021-03-29 09:08:35 uhM, nice 2021-03-29 09:09:24 the pi4 alpine build is nice 2021-03-29 09:09:51 it "just works" lol 2021-03-29 09:11:40 but I don't visualize exactly the 'workflow', I should attach some folder in my home for have persistence... and then just try to build packages and have the results in this "shared" dir? 2021-03-29 09:12:32 solar: no, apk is not transactional at the moment 2021-03-29 09:13:36 it depends of view what is transaction 2021-03-29 09:14:36 There is no roll-back 2021-03-29 09:14:39 uhM, buildozer:x:1000:1000:Linux User,,,:/home/buildozer:/bin/ash 2021-03-29 09:15:24 ikke: apk use renameat(3) 2021-03-29 09:16:25 That's atomic updating of files 2021-03-29 09:16:28 I didn't analyzed its usage in deep 2021-03-29 09:16:40 yeah, I remember that user name from somewhere :D https://tpaste.us/9jv1 2021-03-29 09:16:48 pingberlin: I just pushed a fix to 3.13 for the Powerdevil issue. Should be in the repos in a bit 2021-03-29 09:17:45 donoban: that's the build user of the official builders as well 2021-03-29 09:17:55 I see 2021-03-29 09:20:56 hi PureTryOut[m]2, is the kwin_wayland backtrace useful? I could try to take a look but I suppose that will be pretty difficult code 2021-03-29 09:21:33 I don't know, ask the Plasma devs 2021-03-29 09:22:05 ah ok 2021-03-29 13:05:00 fys mps it's possible for pacman to partially update the system leaving it in some half-state resulting in an unbootable os, I assume `apk` is not capable of that (which would be a good thing) 2021-03-29 13:05:15 I'm not intimately familiar with the internals of package manages beyond a naive guesstimate 2021-03-29 13:05:41 I ask because without such a formal understanding I have no way to ask the question, I need an analogy to what I already know 2021-03-29 13:05:48 I ask in that manner* 2021-03-29 13:05:58 I'm a fan 2021-03-29 13:12:05 brb gym hopefully no suspend hoho 2021-03-29 13:12:35 tsujp: ime, there is no safe package manager, all they can make system unbootable 2021-03-29 13:13:54 Package manager could integrate an A/B booting scheme. But it can be implemented separately fairly easily. 2021-03-29 13:33:12 So I have a CA certificate bundle that I need to use, but `update-ca-certificates` just throws a warning that it doesn't contain exactly one certificate (duh) and skips it 2021-03-29 13:33:33 Is that really intentional? 2021-03-29 14:31:02 I just tried to boot alpine 3.13.3 on a Lenovo X1 Nano, however it seems to not find the usb stick - is it possible that the usb connector is too new? 2021-03-29 14:31:29 So I can boot, grub is displayed, and it stops with "Mounting boot medial failed" Followed by a fallback to initramfs 2021-03-29 14:47:12 Oh, no, it actually has /dev/sda1 2021-03-29 14:57:35 Ah, I forgot to set the label... try2 2021-03-29 15:26:48 I'm currently testing remmina and it seems to fail because shared libraries cannot be loaded 2021-03-29 15:26:57 on edge? 2021-03-29 15:26:57 [17:26:20:576] [7767:8331] [ERROR][com.winpr.library] - LoadLibraryA failed with Error loading shared library /usr/lib/freerdp2/librdpsnd-client-fake.so: No such file or directory 2021-03-29 15:30:52 uhM I saw it failed too but I thought it was due firejail problem 2021-03-29 15:32:29 telmich: does it work if you install freerdp-plugins? 2021-03-29 15:33:01 ikke: it does, indeed! 2021-03-29 15:33:05 Wow, you are amazing! 2021-03-29 15:33:11 https://pkgs.alpinelinux.org/contents?file=librdpsnd-client-fake.so&path=&name=&branch=edge 2021-03-29 15:33:47 So a dependency is probably missing 2021-03-29 15:33:51 yesw 2021-03-29 15:48:06 I checked and works fine without firejail but not with it 2021-03-29 16:24:59 Just installing texlive-full - it really resembles waiting for compiling quite well 2021-03-29 16:31:25 It seems like the wifi is a bit unstable on that machine, wondering if Intel Corporation Wi-Fi 6 AX201 are supposedly working fine already 2021-03-29 17:16:54 Hello ladies 2021-03-29 17:17:15 If somebody know how to get working 'seatd' I m listening :/ 2021-03-29 17:17:51 That shit does only that it runs, nothing more ;) 2021-03-29 17:18:40 Tried it in voidlinux, while running it automatically take care of running sway 2021-03-29 17:18:53 without any fuckery 2021-03-29 17:22:23 ... oh as always, thanks for listening :/ 2021-03-29 17:41:01 upgrade tow host, the bonding is broken 2021-03-29 17:41:30 ping: sendto: Invalid argument 2021-03-29 17:41:45 wener[m]: 3.13? 2021-03-29 17:41:52 en 2021-03-29 17:42:07 ? 2021-03-29 17:42:12 yes 2021-03-29 17:42:24 ncopa just pushed a new kernel version that should fix it 2021-03-29 17:42:33 5.10.26-r1 2021-03-29 17:43:12 emmmmm.... ok, I'll upgrade when the mirror synced 2021-03-29 18:06:40 ikke: so the bonding issue has been found? 2021-03-29 18:07:48 yes 2021-03-29 18:07:52 kernel bug 2021-03-29 18:08:04 after upgrade, https://gitlab.alpinelinux.org/alpine/aports/-/issues/12562 2021-03-29 18:34:29 upgrade, bonding works now 2021-03-29 18:46:04 cool 2021-03-29 18:46:17 I ran into this this weekend 2021-03-29 20:47:38 btw. got why seatd does not work with sway 2021-03-29 20:48:02 wlroots it's not build with libseat option 2021-03-29 20:48:15 as always ... thank you for your support :/ 2021-03-29 20:51:00 helby: thank you for your patience 2021-03-29 20:53:47 ikke: you welcome ;) 2021-03-29 20:55:06 still it answers just one question from many :/ 2021-03-29 20:56:11 helby: please keep in mind that this is all supported by volunteers 2021-03-29 20:56:41 we do our best to make sure things are working, but it's impossible to verify all sorts of setups 2021-03-29 20:56:46 ikke: please keep in mind, I know about it, I m just joking 2021-03-29 20:56:56 helby: it does not come over like that 2021-03-29 20:57:23 anyway any news about maintainer of firefox? already out of jail? :/ 2021-03-29 20:59:01 ikke: don't worry, even if it can sound I demand or expect something, I don't, just normally asking about something what I can't get answer about in alpine, I know that this is 'slow' channel 2021-03-29 20:59:48 still it looks like nobody run it daily on desktop, just me :/ 2021-03-29 21:01:18 given the regular reports we get, you are not the only one, believe me :) 2021-03-29 21:01:39 really? is there another one? ;) 2021-03-29 21:01:43 how do you all recommend making sure that nothing from my personal /etc/apk/world is leaking into the `makedepends` of an APKBUILD i'm writing? 2021-03-29 21:02:01 Containers 2021-03-29 21:02:09 https://gitlab.alpinelinux.org/alpine/docker-abuild 2021-03-29 21:22:31 kk thanks 2021-03-29 21:42:34 Is alpine linux bullet-proof against attacks over network where the attacker already got my password? 2021-03-29 21:46:15 lmfao no 2021-03-29 21:46:31 nothing is 2021-03-29 21:51:03 Are you sure? 2021-03-29 21:59:23 Back. I disconnected 2021-03-29 21:59:29 i can set it up so no one can access it 2021-03-29 21:59:31 even yourself 2021-03-29 22:04:46 anysystem can be setup in a way, where security of the system does not depend on a password stored for reference on that system. 2021-03-29 22:04:46 but pakes are actually a problem if the attacker already knows your password. 2021-03-29 22:04:46 disconnect it from the network 2021-03-29 22:04:46 pubkey systems work without passwords stored on the host. also pakes are a thing... 2021-03-29 22:12:26 Disconnected again 2021-03-30 00:28:55 Does anyone have any experience running SPEC CPU2006 on Alpine? 2021-03-30 02:34:07 Hi! 2021-03-30 02:34:39 Is there a package to generate a default self signed certificate? Similar to how debian does it (with snakeoil certs). 2021-03-30 03:00:00 is it possible to have a verbose init process? 2021-03-30 03:00:09 im stuck on ` * Caching service dependencies ... [ ok ]` 2021-03-30 03:16:53 bl4ckb0ne: /etc/rc.conf the rc_logger and rc_verbose directives might help you... Also consider rc_interactive. Good luck. 2021-03-30 03:17:01 thanks 2021-03-30 03:28:31 got it, I was missing https://wiki.alpinelinux.org/wiki/Alpine_Linux_in_a_chroot#Preparing_init_services 2021-03-30 03:34:19 hangs on ` * Starting busybox syslog ... [ ok ]` now 2021-03-30 03:56:13 Oh, is Alpine WSL "official"? 2021-03-30 04:30:53 do patches sent to the aports mailing list get automatically tested in CI, or does a maintainer have to trigger that themselves? 2021-03-30 04:31:25 (just submitted my first patch) 2021-03-30 05:46:47 piousminion: I don't think it is. It's just some dude who made it 2021-03-30 05:47:00 (I've been using it for work for about a year now) 2021-03-30 05:47:52 Any ideas? Or do I need to script this myself? 2021-03-30 05:50:59 wyoung: not aware of anything packaged 2021-03-30 05:52:59 ikke: I mean Debian has this type of thing in a package. I know Alpine isn't Debian but I assumed it would have something similar. If not that is fine, I will just add it to my docker image RUN clause. 2021-03-30 05:59:00 do you all do something to set a preferred language for man pages? i'm noticing that mandoc seems to just pick the first file it sees, which sometimes isn't in english if a package ships documentation in multiple languages 2021-03-30 05:59:47 and nothing in the mandoc documentation says anything about language/locale 2021-03-30 07:17:37 Never ran into non-english man pages 2021-03-30 07:32:52 @Fishie03 2021-03-30 07:53:02 outerpassage, what's your locale? 2021-03-30 07:53:15 honstely translated manpages are usually never up to date 2021-03-30 07:53:47 ikke: for example, weechat and alsaconf have non-english man pages 2021-03-30 07:53:57 markand: en_US 2021-03-30 07:54:09 I remember when I was younger and still using my fedora box in french, I read the rpmbuild manpage in french and it described an option that I tried to apply in command line and oops : unrecognized option :} 2021-03-30 07:54:16 since then, always in english everywhere 2021-03-30 07:54:35 oh i don't care about having non-english man pages, i just don't want to have to rebuild these *-doc packages without the other languages if possible 2021-03-30 07:54:43 outerpassage, and now if you type man weechat it appears in which language ? 2021-03-30 07:55:24 czech, I think 2021-03-30 07:55:35 let me have a look on mine 2021-03-30 07:55:46 because the first line of `man -w weechat` is /usr/share/man/cs/man1/weechat.1.gz 2021-03-30 07:56:41 yeah confirmed with translate it's czech 2021-03-30 07:58:58 $ man -w weechat 2021-03-30 07:58:59 /usr/share/man/man1/weechat.1.gz 2021-03-30 07:59:30 can you paste somewhere your env? 2021-03-30 07:59:38 ACTION looks surprised 2021-03-30 07:59:40 that's weird 2021-03-30 07:59:45 definitely 2021-03-30 08:02:00 https://paste.ee/p/gVAan 2021-03-30 08:02:05 env as in environment variables? 2021-03-30 08:02:12 yes 2021-03-30 08:02:15 ^ 2021-03-30 08:02:31 man isn't aliased either? 2021-03-30 08:02:48 nope 2021-03-30 08:04:05 mandoc version is mandoc-1.14.5-r5 2021-03-30 08:04:32 mine too 2021-03-30 08:06:18 so weechat-doc does include the other languages for you, it's just that man -w weechat lists the english one first? 2021-03-30 08:06:27 yes 2021-03-30 08:06:37 hmmmm i am stumped, that's strange 2021-03-30 12:42:35 Hello... 2021-03-30 12:43:50 Can LInux use different bond types at the same time? As in - bond0 should be 802.3ad (LACP), while bond1 should be 'normal' active-backup ? 2021-03-30 13:17:58 yes 2021-03-30 13:36:22 outerpassage: set LANG, not LC_CTYPE 2021-03-30 13:36:59 (don't set LC_ALL either) 2021-03-30 14:38:13 is there a script that setups openrc after a chroot install? 2021-03-30 14:39:52 dunno if there's a dedicated script 2021-03-30 14:39:58 but you can do something like this: https://gitlab.com/Durrendal/Droid4-Alpine-Bootstrap/-/blob/master/scripts/chroot-build-base.sh#L100-126 2021-03-30 14:41:03 what's the diff between syslog and syslog-ng? 2021-03-30 14:43:06 bl4ckb0ne: https://arvanta.net/alpine/install-alpine-on-acer-r13-chromebook/ 2021-03-30 14:43:11 syslog is busybox's implementation, syslog-ng is it's own thing 2021-03-30 14:43:26 look at part where are 'ln' 2021-03-30 14:43:53 oh cool you can just ln them into the runlevels 2021-03-30 14:44:16 i started from your script mps 2021-03-30 14:44:17 yes 2021-03-30 14:44:24 btw alpine-mirrors doesnt exist anymore 2021-03-30 14:45:05 bl4ckb0ne: this script is bare minimum to get bootable system 2021-03-30 14:45:06 are those links equivalent to rc-update add ? 2021-03-30 14:45:16 it boots but doesnt init 2021-03-30 14:45:21 got stuck at syslogd 2021-03-30 14:45:33 yes, rc-update just make (and remove) this links 2021-03-30 14:46:23 it sounds suspiciously like your problem is related to getty 2021-03-30 14:47:25 i have modified the inittab to get serial login 2021-03-30 14:47:43 but did you do it properly 2021-03-30 14:48:02 ttyS0::respawn:/sbin/getty -L ttyS0 115200 vt100 2021-03-30 14:48:04 ttyS2::respawn:/sbin/getty -L ttyS2 115200 vt100 2021-03-30 14:48:21 hm wait baud rate is wrong 2021-03-30 14:48:41 should be 1500000 2021-03-30 14:49:20 uh 2021-03-30 14:50:04 i don't think you can get that rate on a real serial line, or a real 16550 2021-03-30 14:50:19 got `console=ttyS2,1500000` in my cmdline already 2021-03-30 14:50:40 hah, now I know who is durrendal :) 2021-03-30 14:51:17 I'm super bad at consistently choosing a username :) 2021-03-30 14:52:56 I see you use printf, nice idea, thanks for hint 2021-03-30 14:53:22 i think it was urandom missing + the false baud rate in inittab 2021-03-30 14:56:45 on a real tty the getty baud rate will override the console parameter 2021-03-30 14:56:53 but if you're virtualizing then it shouldn't do anything 2021-03-30 14:57:03 its real hw 2021-03-30 14:57:38 mps: it's a little easier to get the formatting nice with printf I've found. That entire repo is a hot mess though, but if you have an old motorola droid4 it technically works 2021-03-30 14:57:51 wut root login incorrect 2021-03-30 14:58:29 ttyS2 in /etc/securetty? 2021-03-30 14:58:35 seems like it 2021-03-30 14:59:27 S0 and S1 2021-03-30 14:59:55 wsinatra: well, all script/notes I made are not so good, but they are just as starting point for someone who want to install/setup devices 2021-03-30 15:01:01 might be missing something else too 2021-03-30 15:01:05 my user gets logged out 2021-03-30 15:01:20 fair take, I'd say mine fall into the same bin. 2021-03-30 15:01:22 i have 2 lines of the motd 2021-03-30 15:01:50 bl4ckb0ne: this is on your helios64 2021-03-30 15:01:52 ? 2021-03-30 15:01:53 1yes 2021-03-30 15:02:03 i finally took some time to redo the sd card with a proper uboot 2021-03-30 15:02:12 comment out ttyS0 in inittab 2021-03-30 15:02:24 you don't need it, I think 2021-03-30 15:04:37 > snorlax auth.err getty[1721]: tcgetattr: I/O error^M 2021-03-30 15:04:42 at least there's an error 2021-03-30 15:06:07 could I be missing some dumb stuff like polkit 2021-03-30 15:06:35 if you don't use polkit you are smart and not dumb 2021-03-30 15:06:57 i recently got locked out of some archlinux because of dumb polkit stuff 2021-03-30 15:07:23 still cant login 2021-03-30 15:07:29 no polkit (or any *kit) on my boxes 2021-03-30 15:07:58 good 2021-03-30 15:08:04 i dont have it aswell 2021-03-30 15:08:45 could you look how armbian set getty for ttyS2 2021-03-30 15:08:55 i.e. parameters 2021-03-30 15:09:09 got it right last time with the pmos sdcard 2021-03-30 15:09:16 meh, they probably use agetty 2021-03-30 15:09:38 armbian ^ 2021-03-30 15:09:49 https://paste.sr.ht/~bl4ckb0ne/77dbb37e5f53a2fb77522f16cb87b10601b8cca3 2021-03-30 15:09:50 look at pmos then 2021-03-30 15:09:51 its really weird 2021-03-30 15:11:33 yes 2021-03-30 15:16:55 seems like getty is exiting 2021-03-30 15:17:10 > process '/sbin/getty -L ttyS2 1500000 vt100' (pid 1702) exited. Scheduling for restart. 2021-03-30 15:17:26 on agetty you can omit those extra arguments 2021-03-30 15:17:30 not sure if busybox getty allows 2021-03-30 15:17:51 you can try 0 instead 2021-03-30 15:18:13 ttyS0? 2021-03-30 15:19:22 i mean /sbin/getty -L ttyS2 0 vt100 2021-03-30 15:21:08 getty -h, 'BAUD_RATE of 0 leaves it unchanged' 2021-03-30 15:21:43 nup same issue 2021-03-30 15:22:39 are you sure you don't have two (or more) getty's for ttyS2 2021-03-30 15:23:43 i dont think so 2021-03-30 15:25:06 do you have getty enabled on any other tty in /etc/inittab 2021-03-30 15:25:44 the basic config yes 2021-03-30 15:26:56 not sure, but to me looks like two getty's runs on same tty 2021-03-30 15:27:36 have to go now, cul 2021-03-30 15:29:57 managed to log as root o/ 2021-03-30 15:30:04 added ttyS2 in securetty 2021-03-30 15:30:43 well back to square 1, no ethernet device and no sata drives 2021-03-30 15:33:18 mps: draft note for alpine on the helios64 https://paste.sr.ht/~bl4ckb0ne/82e17743b6498047e60ab20bad05ea74937ccfcf 2021-03-30 17:26:07 Is it necessary to SIGHUP busybox syslog after rotating the logs? logrotate is configured to rotate /var/log/messages by default, but it doesn't signal syslogd by default. 2021-03-30 17:28:59 xordspar0: I think so 2021-03-30 17:29:33 xordspar0: another method is to use something like cronolog to pipe the output through which then will automatically create file with proper date without rotating anything 2021-03-30 17:29:51 it just directs the output to next file according to your stringsyntax 2021-03-30 17:30:36 example: /usr/local/bin/cronolog /log/apache/error%y%m.log 2021-03-30 17:32:41 That's neat 2021-03-30 17:36:32 however I think cronolog uses slightly more cpu since it will check the time every now and then 2021-03-30 17:36:44 dunno if it does that line by line or have some kind of cache 2021-03-30 17:37:00 I assume line by line 2021-03-30 17:37:19 and probably an open call for each line too 2021-03-30 17:38:08 bl4ckb0ne: good for start, hope you will refine it a little 2021-03-30 17:38:47 xordspar0: look at metalog pkg, maybe it will be good for you 2021-03-30 17:43:55 Cool, thanks mps 2021-03-30 17:49:13 Oh, busybox syslogd actually ignores SIGHUP. https://git.busybox.net/busybox/tree/sysklogd/syslogd.c?id=3c13da3dab539eac948de48640d8862857d0c8d0#n1023 2021-03-30 17:50:04 Instead, it attempts to reopen the log file periodically. Interesting. https://git.busybox.net/busybox/tree/sysklogd/syslogd.c?id=3c13da3dab539eac948de48640d8862857d0c8d0#n701 2021-03-30 18:02:23 bl4ckb0ne: maybe this is better https://github.com/kobol-io/u-boot/commit/29d63b29550818992e3bcdb1ceb2a0db49d395cc 2021-03-30 18:02:59 it applies to current u-boot 2021.01 version 2021-03-30 18:03:25 though I didn't tried to build because don't have board to test 2021-03-30 19:28:24 mps: yeah this is just the rough notes, gonna make something nicer 2021-03-30 19:28:36 iirc i tried the one from the kobol repo and it failed on 2021.1 2021-03-30 19:28:48 forgot to mention the uboot mess in my mail tho 2021-03-30 19:34:35 bl4ckb0ne: aha, ok 2021-03-30 19:34:42 it was a bloody mess 2021-03-30 19:35:04 i also asked for extra boards but from their shop it seems the helios64 is out of order 2021-03-30 19:35:31 dead end? 2021-03-30 19:35:43 idk, ive sent the mail this morning 2021-03-30 19:37:07 everything about uboot is a bloody mess. :P 2021-03-30 19:37:21 ACTION drowning 2021-03-30 19:37:29 i cant find any track of kobol pushing their patch upstream 2021-03-30 19:38:25 that is how most SBC makers do 2021-03-30 19:38:28 their otp is full of 'oh dear sweet child no.' 2021-03-30 19:38:47 they are waiting for someone else to do their works 2021-03-30 19:38:55 classic 2021-03-30 19:39:08 well at least it works and I have a pmos-less sd card 2021-03-30 19:39:29 though samsung is somewhat ok 2021-03-30 19:39:45 there's just so much... wtf 2021-03-30 19:40:09 I had good communications with samsung people few years ago 2021-03-30 19:40:47 apparently rockchip is also trying to be the new realtrash. the gpio stuff is dumb but necessary because hardware kludge. that's fine. but WTF software controlled vdd *ARE YOU HIGH?* 2021-03-30 19:41:43 to fry some parts of SBC, as I did with audio on my one 2021-03-30 19:41:48 gee i can't imagine anything bad EVER happening with a software controlled, os accessible vdd in an operating system that has never ever ever ever had a single vulnerability ever, nope nope nope 2021-03-30 19:43:04 speaking of realtrash, gonna try to find the right config on my side now that I can build aports on the helios 2021-03-30 19:44:35 i'm not seeing a lot of truly fubar instruction stuff aside from the ethernet and a couple 'yeah mark it always okay.' most of it's gpio and attached hardware kludge crap. 2021-03-30 19:46:25 i don't think it will actually need anything special for aports, just standard aarch64 nightmares. 2021-03-30 20:05:00 bl4ckb0ne: you could install https://dev.alpinelinux.org/~mps/linux-5.11.6.tar.gz besided current one, and look if it can find sata and ethernet 2021-03-30 20:05:33 maybe this could give us hint what to enable 2021-03-30 20:06:25 sure, i'll send you the results later 2021-03-30 20:07:19 good, but no hurries 2021-03-30 20:08:48 and my love to your ... you know who :) 2021-03-30 20:09:05 hehe 2021-03-30 20:10:01 sorry if didn't wrote it properly, I'm self taught in english and really bad at fields which are not technical 2021-03-30 20:10:36 yeah dont worry im also self taught in english 2021-03-30 20:10:52 ah, you are not in USA? 2021-03-30 20:11:14 I thought you are american 2021-03-30 20:12:27 i live in canada but im french 2021-03-30 20:12:51 ah, ok :) 2021-03-30 20:13:31 your nick and time zone 'fooled' me 2021-03-30 20:13:56 hehehe :D 2021-03-30 20:26:17 https://www.arm.com/blogs/blueprint/armv9 Armv9 on horizon 2021-03-30 20:26:43 weeeeee 2021-03-30 20:27:32 I would like to read this but that it themed: Risc-V 2021-03-30 21:26:03 100mb was not enough for 3 kernels 2021-03-30 21:27:55 mps: 5.11.6 has the eth driver 2021-03-30 21:28:05 device is showing 2021-03-30 21:28:22 nothing for sata drives 2021-03-30 21:30:41 bl4ckb0ne: I usualy make 512MB boot partitions 2021-03-30 21:30:55 [ 5.000730] RTL8211F Gigabit Ethernet stmmac-0:00: attached PHY driver (mii_bus:phy_addr=stmmac-0:00, irq=POLL) 2021-03-30 21:30:58 [ 5.000740] RTL8211F Gigabit Ethernet stmmac-0:01: attached PHY driver (mii_bus:phy_addr=stmmac-0:01, irq=POLL) 2021-03-30 21:31:09 aha, good 2021-03-30 21:31:18 have to find in config 2021-03-30 21:31:36 something with realtek and phy 2021-03-30 21:31:42 yes 2021-03-30 21:32:14 for sata, I'm not sure that the driver is in kernel 2021-03-30 21:32:51 will try too look on pmos patches 2021-03-30 21:33:06 ill notify you as soon as kobol answers 2021-03-30 21:33:24 ok, and good 2021-03-30 21:34:43 lspci -k 2021-03-30 21:34:43 > [ 305.011190] rk_gmac-dwmac fe300000.ethernet eth0: PHY [stmmac-0:00] driver [RTL8211F Gigabit Ethernet] (irq=POLL) 2021-03-30 21:34:43 if that helps 2021-03-30 21:34:46 nothing shows 2021-03-30 21:34:55 might be missing a few packages tho 2021-03-30 21:35:06 apk add pciutils 2021-03-30 21:35:16 still configuring the eth0 interface 2021-03-30 21:35:26 aha 2021-03-30 21:35:39 i have it in /etc/network/interfaces but i dont have an ip 2021-03-30 21:36:04 wait i have an ipv6 2021-03-30 21:36:51 ip a a dev eth0 $ipaddr/mask brd + 2021-03-30 21:37:07 ip l s dev eth0 up 2021-03-30 21:37:18 ip r a default $router 2021-03-30 21:37:59 echo 'resolver $resolver_address' > /etc/resolv.conf 2021-03-30 21:38:11 echo 'nameserver $resolver_address' > /etc/resolv.conf 2021-03-30 21:38:24 ty 2021-03-30 21:39:25 > ip: ioctl 0x8914 failed: Operation timed out 2021-03-30 21:39:27 doesnt look so good 2021-03-30 21:39:51 uh 2021-03-30 21:40:59 seems to be a dma issue 2021-03-30 21:41:16 https://paste.sr.ht/~bl4ckb0ne/659718be2545b15b41ceb36cb68016407ba9ab48 2021-03-30 21:41:50 yes 2021-03-30 21:43:30 really have to look at pmos or armbian patches for this board 2021-03-30 21:43:37 will do too 2021-03-30 21:44:06 iirc they got 5.11.2 to run with no patch 2021-03-30 21:46:35 who? 2021-03-30 21:47:13 pmos 2021-03-30 21:47:25 on rockchip boards* 2021-03-30 21:47:52 their repo is on gitlab.com somewhere? 2021-03-30 21:48:21 hi guys, I'm trying to learn how to use dabuild. According to https://gitlab.alpinelinux.org/alpine/docker-abuild "Per normal usage, if you use the -K switch, then the build, source, etc directories will be left alone on completion" 2021-03-30 21:48:37 but I feel it doesn't work 2021-03-30 21:48:41 https://gitlab.com/postmarketOS 2021-03-30 21:49:04 where should I see all the source and build objects? 2021-03-30 21:55:54 bl4ckb0ne: hm, https://gitlab.com/postmarketOS/pmaports/-/tree/master/main/linux-postmarketos-rockchip says it is for pkgname=linux-postmarketos-rockchip 2021-03-30 21:56:03 this one works ? 2021-03-30 21:58:22 huh, to late is here, good night 2021-03-30 21:59:17 good nigth mps 2021-03-30 22:00:29 donoban: :) 2021-03-30 22:43:30 what the fuck did kim jong johnnie walker do this time 2021-03-30 22:43:45 .... 2021-03-30 22:43:50 obviously the wrong channel 2021-03-30 22:43:52 >_> 2021-03-30 22:44:11 i gotta work on my tact.. 2021-03-30 23:14:56 fys: :) 2021-03-31 00:58:33 nmeum: hi! i'm the author of this patch https://gitlab.alpinelinux.org/alpine/aports/-/merge_requests/19900, which still needs fixing. is the best way to submit a revised patch via the mailing list, or is there some way of adding to the merge request via gitlab? 2021-03-31 02:30:17 Hi! I was wondering if there's a way to bypass dbus? 2021-03-31 02:30:51 I'm using Alpine as a daily driver and I've noticed that some stuff, like notification daemons (on wayland) sort of need dbus 2021-03-31 02:31:15 And other things seems to work, but sometimes they don't 2021-03-31 02:31:25 And keep complaining about not finding dbus 2021-03-31 02:35:32 I believe the xfce install guide has you install it so, my best guess is that its assumed 2021-03-31 02:44:13 Ok, thanks! 2021-03-31 02:44:19 Ended up installing anyway 2021-03-31 03:10:08 get rid of notification daemon too 2021-03-31 03:11:36 helby: what do you mean? 2021-03-31 03:12:14 porcellis: get rid of anything what requires dbus 2021-03-31 05:52:31 outerpassage: yes, MR is a option which will give faster response than sending patch to ML 2021-03-31 05:52:57 and MR will trigger automatic build on CI 2021-03-31 05:54:38 thanks! i'll do that for future patches 2021-03-31 06:17:42 hello, I am trying to run nwipe in lxc debian, since it has not yet found it's way to apk. i get this error, either priviliged or unprivilged /sys/firmware/dmi/tables/smbios_entry_point: Permission denied (newfile) /dev/mem: No such file or directory 2021-03-31 06:20:14 bbl 2021-03-31 09:37:19 is there a release in the pipeline for fixing CVE-2021-28831, which seems to be fixed by https://github.com/alpinelinux/aports/commit/45b44a439b317a93ead58b7da727295a2f9407f5? ... 2021-03-31 10:10:37 Hi, I would like to learn about pubkey systems, is it a system where password login is disabled? 2021-03-31 10:11:45 davidreuss: wrong commit? 2021-03-31 10:12:08 Jackura: this is general knowledge, not specific to alpine linux 2021-03-31 10:12:20 Jackura: do you mean SSH keys? 2021-03-31 10:12:30 Jackura: search on duckduckgo about it 2021-03-31 10:12:54 there are a lot of good articles and free books on that subject 2021-03-31 10:13:28 nope, I heard about pubkey system, I thought it had nothing to do with ssh, tho i know of ssh keys. I am thinking about a soulution for desktop. 2021-03-31 10:14:10 I also know of keyfile for luks, but i was thinking about DE login 2021-03-31 10:15:27 Jackura: you mean token devices? 2021-03-31 10:15:41 token device? 2021-03-31 10:16:17 I wonder, how do one protect the keyfile when using desktop? 2021-03-31 10:16:30 Jackura: public-key cryptography is very generic, and can be used in all sorts of applications 2021-03-31 10:16:34 devices which contain secret/pubkey pair and are used as login tokens 2021-03-31 10:17:50 Does it verify the system before giving keyfile? 2021-03-31 10:18:25 Jackura: what system are you referring to? 2021-03-31 10:18:45 Desktop System, not remote.. 2021-03-31 10:18:56 alpine linux on a desktop 2021-03-31 10:18:57 Still not clear 2021-03-31 10:21:18 about what 2021-03-31 10:30:59 I would like a system where the device holding the keys, verify the system intergrity before handing over keys. and also prompts for password. 2021-03-31 10:40:38 Jackura: sounds like a TMP 2021-03-31 10:40:41 TPM 2021-03-31 11:23:46 ikke: does TMP USB/Storage devices exist? 2021-03-31 11:24:14 TPM is a chip that is part of the main board 2021-03-31 11:24:36 As an external decice it does not make a lot of sense 2021-03-31 11:24:55 What I mean is, how does the device giving the keyfile know who it's giving it to? 2021-03-31 11:25:34 say if the system has been comprimised 2021-03-31 11:26:24 signature checking of the boot chain 2021-03-31 11:26:32 and verifying the hardware 2021-03-31 11:26:40 that's what secure boot is about 2021-03-31 11:26:57 with signed bootloaders 2021-03-31 11:38:25 ikke: some boards have headers for tpm 2021-03-31 11:38:42 not sure if this decreases security but I assume so 2021-03-31 11:55:54 thanks 2021-03-31 13:01:40 Hey guys... 2021-03-31 13:04:11 I'm just finishing updating my router to Alpine Linux 3.13.3 and I noticed that on boot there is an error that goes something like 'ERROR: alpine-baselayout-3.2.0-r8: failed to rename var/.apk.xxxxxxx to var/run'. It boots OK, but then 'apk' complains about an error until I run 'apk fix', which reinstalls alpine-baselayout. 2021-03-31 13:07:40 Is this something I need to worry about? 2021-03-31 13:19:22 hi 2021-03-31 13:19:27 alpine linux rocks :D 2021-03-31 13:19:31 super fast install 2021-03-31 13:19:39 super small / root filesystem :) 2021-03-31 14:46:56 mps: im gonna try it 2021-03-31 14:48:20 bl4ckb0ne: I looked on pmos kernel config, can't find anything related to sata which we don't have already 2021-03-31 14:49:16 yeah, same with realtek stuff 2021-03-31 14:49:45 for realtek I think I should look once more 2021-03-31 14:51:59 gah no space left 2021-03-31 14:57:11 make new layout, and use 512MB for boot at least 2021-03-31 14:57:32 you will probably need it to test all this mess 2021-03-31 15:07:02 yeah 2021-03-31 15:07:06 i guess ill nuke my card then 2021-03-31 15:42:29 I'm seeing this error when running make https://bpa.st/UAEQ . What steps can I take to get past it? 2021-03-31 15:58:34 add a main function? 2021-03-31 16:19:13 Hello71: This is Wine. It's large and not a program I've created. Will doing something like "-nostartfiles -e _main" proposed here https://forums.freebsd.org/threads/problem-linking-with-libc.52488/ work or break everything? Can that be put into an APKBUILD? 2021-03-31 16:40:12 bovis: do you mean are you trying to build wine? there is already a package 2021-03-31 16:44:46 donoban: Yes. Building the latest version. Repos is still at v4. 2021-03-31 16:47:23 ah ok 2021-03-31 16:50:30 donoban: Do you know if a -nostartfiles flag could be used in this situation? The BSD blog I linked above talks about adding it to ld as a standalone command. I don't know how applying it to an APKBUILD would work or what it would do. 2021-03-31 16:52:14 I have no idea sorry 2021-03-31 16:52:35 got to go now, maybe I could take a look later 2021-03-31 16:52:42 see you 2021-03-31 16:53:38 try it and see what happens 2021-03-31 17:41:56 donoban c705 Unsure if this is exactly correct, but adding CXXFLAGS="$CXXFLAGS -nostartfiles" to the APKBUILD didn't help with the build. 2021-03-31 18:02:15 is maintainer of firefox still in jail? :/ 2021-03-31 18:03:38 There's a MR for FF pending 2021-03-31 18:04:12 And actually helping with the bump tends to be more productive than asking such questions 2021-03-31 18:28:04 not sure what 'helping with the bump' means, however good news that something is happening around firefox this year :/ 2021-03-31 18:42:34 ikke: do you know of any progress with that lvm scanning issue? it seems the issue report sees no more updates. 2021-03-31 19:01:12 Hi, the just pushed docker images for 3.13.4 seems to miss the amd64 arch for the tag 3.13.4 2021-03-31 19:13:08 anydot▸ there's #docker-library for docker official images. it's a pretty slow channel, but the folks who run the build/push for official images are there. 2021-03-31 19:16:05 thanks, I'll move it there 2021-03-31 19:20:10 hi 2021-03-31 19:20:31 does alpine linux have X ? can i install some lightweight DE, lxde lxqt ? :) 2021-03-31 19:22:00 spacedust, it does have X, and lxqt is packaged. I don't think lxde is 2021-03-31 19:22:24 XFCE, i3, and a few others are packaged. And package contributions are very much welcomed :) 2021-03-31 19:23:02 bovis: are you starting from the current APKBUILD version right? 2021-03-31 19:24:56 lxde is well what i used before lxqt :D now i use lxqt and i like it 2021-03-31 19:25:25 then it sounds like you're in luck! 2021-03-31 19:25:27 well i dont know much about alpine, but i tried it in the past and now again and it installs so fast, i mean what is not to like ? :) 2021-03-31 19:25:47 packages well i first have to get to know alpine linux better before i would venture there :) 2021-03-31 19:26:29 i like little space it takes up, how fast it gets you up and running, and that you can also run it as it without even installing it 2021-03-31 19:26:37 but as i said i have to play around with it a bit more :) 2021-03-31 19:26:42 is it used in production ? 2021-03-31 19:26:44 in many places ? 2021-03-31 19:27:27 i have an older machine and i was thinking i would make it into a vm host with alpine, can i install libvirtd ? 2021-03-31 19:30:19 donoban: https://bpa.st/S7GQ Here's my current APKBUILD. I'm using two updated patches from the 4.x one in our repos. And this build is for x86 (see the -no-pie flag). 2021-03-31 19:30:27 bl4ckb0ne: https://github.com/armbian/build/tree/master/patch/kernel/archive/rockchip-5.11 I don't see anything related to sata or ethernet 2021-03-31 19:34:18 mps: https://github.com/armbian/build/blob/master/patch/kernel/archive/rockchip64-5.11/board-helios64-remove-pcie-ep-pios.patch 2021-03-31 19:34:22 helios is rockchip64 2021-03-31 19:34:28 and still getting the missing main() error? 2021-03-31 19:36:04 donoban: Yes. -no-pie fixed a previous build error https://bpa.st/CTTA 2021-03-31 19:36:22 oh boy GNOME 4.0 is so buggy 2021-03-31 19:36:47 trying to click on a dash icon pops up the calendar view 2021-03-31 19:36:49 mps: still no answer from kobol 2021-03-31 19:36:55 i tried to ping the armbian peeps but no answer 2021-03-31 19:39:41 hello everyone, i was wondering why there is always a delay in releasing amd64 version of latest alpine release? 2021-03-31 19:40:14 for example right now from what i can see in https://hub.docker.com/_/alpine?tab=tags&page=1&ordering=last_updated 2021-03-31 19:40:40 latest and 3.13.4 both has been pushed just 24 minutes ago, but 3.13.4 is missing amd64 2021-03-31 19:40:53 i've faced this issue for both 3.13.3 and 3.13.2 iirc 2021-03-31 19:42:39 https://github.com/armbian/build/blob/master/config/kernel/linux-rockchip64-edge.config markand 2021-03-31 19:43:21 mps* 2021-03-31 19:44:30 how should I run 'dabuild' script for avoid it cleaning up? 2021-03-31 19:46:02 bovis: why you have a checksum for wine-6.5.tar.xz? 2021-03-31 19:46:19 is it not wine-6.0? 2021-03-31 19:46:58 mps: postmarketos-rockchip has ethernet and no sata drives 2021-03-31 19:47:44 spacedust: libvirtd runs on alpine excellently, I use it for my hypervisors. I'm also using it in production (mix of workstations, servers, embedded systems, and docker containers) 2021-03-31 19:48:09 that's been for a few years now, I've had less issues than when everything used to run on top of Debian 2021-03-31 19:48:20 Throw it on a machine and tinker, it's well worth it! 2021-03-31 19:50:33 bl4ckb0ne: does ethernet works with pmos kernel 2021-03-31 19:51:32 no error but no ip 2021-03-31 19:51:34 working on it 2021-03-31 19:52:07 device is up, i have a ipv6 but no ipv4 2021-03-31 19:52:19 donoban: I was switching them up to see if 6.5 performed any differently. 2021-03-31 19:52:42 ah I found it 2021-03-31 19:52:47 hmm, do you dhcp or static ip 2021-03-31 19:52:48 The APKBUILD would need an abuild checksum 2021-03-31 19:52:53 dhcp 2021-03-31 19:53:13 bl4ckb0ne: maybe dhcp server doesn't work 2021-03-31 19:53:20 yeah thats my guess 2021-03-31 19:53:25 maybe i am missing the package too 2021-03-31 19:53:53 no, by default udhcpc is installed 2021-03-31 19:54:09 welp it was not started by default 2021-03-31 19:54:17 pinging 8.8.8.8 2021-03-31 19:54:24 udhcpc -h 2021-03-31 19:54:45 ye 2021-03-31 19:54:52 let me add it to openrc 2021-03-31 19:55:07 udhcpc -i eth0 -f 2021-03-31 19:55:11 to debug 2021-03-31 19:55:25 yeah its working 2021-03-31 19:57:04 bl4ckb0ne: with pmos kernel? 2021-03-31 19:57:08 yes 2021-03-31 19:57:13 good 2021-03-31 19:57:27 you got network set? 2021-03-31 19:58:21 bovis: it fails patching, I suppose that you have different patch files 2021-03-31 19:58:35 yes 2021-03-31 19:58:52 but honestly I don't hope I could help you on this, at least more than you are already doing 2021-03-31 20:00:03 but I wanted to learn to use dabuild :) 2021-03-31 20:00:23 bl4ckb0ne: and doesn't work with latest linux-edge I posted to dev.a.o? 2021-03-31 20:00:29 it would be nice to have directly a dabuild package which installs docker and needed dependencies 2021-03-31 20:00:44 r3? 2021-03-31 20:00:53 donoban: Yes. If you'd like them, no-pie patch: https://bpa.st/MJ7Q winhlp32 patch: https://bpa.st/PIKQ 2021-03-31 20:00:53 which is latest 2021-03-31 20:01:27 bl4ckb0ne: yes, r3 2021-03-31 20:01:58 i have linux-edge-5.11.10-r3 yes 2021-03-31 20:02:00 let me reboot 2021-03-31 20:02:15 bl4ckb0ne: waiting ... 2021-03-31 20:02:36 donoban: Yes, if you're building it too, I expect you'd receive the same error. I'm willing to try other things, it's just over my head 2021-03-31 20:03:03 no eth0 on linux-edge 5.11.10-2 r3 2021-03-31 20:03:12 udhcpd failed 2021-03-31 20:03:40 > udhcpc: ioctl 0x8933 failed: No such device 2021-03-31 20:03:54 nice bug report 2021-03-31 20:03:59 ye 2021-03-31 20:04:13 phy probably, but which one 2021-03-31 20:04:19 is there some way for dabuild doesn't remove the container and you can try to rebuild without prepare again? 2021-03-31 20:05:01 CONFIG_STMMAC_ETH is missing 2021-03-31 20:05:07 it's compiling 2021-03-31 20:05:14 armbian has it with =m 2021-03-31 20:05:27 also pmos 2021-03-31 20:05:37 it is ok to be module 2021-03-31 20:05:44 yeah 2021-03-31 20:06:02 good, we probably found 'suspect' 2021-03-31 20:06:13 sounds sus to me m8 2021-03-31 20:06:14 armbian has almost all of the CONFIG_PHY_ROCKCHIP_* to =y 2021-03-31 20:06:59 I also set most in kernel I uploaded to dev.a.o 2021-03-31 20:07:01 i still need to put alpine on my espressobin.. i finally got around to doing it on my pi 2021-03-31 20:07:04 lol 2021-03-31 20:07:08 bovis: I hope it doesn't remove all when fails, with 2cores+ht should it delay too long? 2021-03-31 20:08:00 bl4ckb0ne: ok, I will try to add stmmac and build new kernel 2021-03-31 20:08:07 ty 2021-03-31 20:08:33 but I'm not sure will I finish it before bed time 2021-03-31 20:08:45 donoban: I'm not sure about that. I'm not using dabuild, only abuild. A full build of wine takes me a while. It's probably 10 minutes to get to the error that I've showed you. 2021-03-31 20:08:45 np 2021-03-31 20:08:45 hacker tip: don't sleep 2021-03-31 20:09:05 Geronimo :) 2021-03-31 20:09:29 ok so see you later I'm gonna have dinner 2021-03-31 20:09:36 I don't sleep, I go to bed for other reasons 2021-03-31 20:09:55 >grin< 2021-03-31 20:09:56 :P 2021-03-31 20:09:56 https://github.com/armbian/build/commit/edb45f9acf322fc4b27bf7efba3f14a3a432b617 interesstingstuff 2021-03-31 20:10:14 are you going to read some paper? 2021-03-31 20:10:50 bl4ckb0ne: aha, good 2021-03-31 20:10:55 patch is gone on master 2021-03-31 20:11:17 which master? kernel mainline? 2021-03-31 20:11:32 armbian buildmaster 2021-03-31 20:11:36 ah 2021-03-31 20:11:51 https://github.com/armbian/build/blob/master/patch/kernel/archive/rockchip64-5.11/board-helios64-remove-pcie-ep-gpios.patch there's still this one FWIW 2021-03-31 20:11:58 donoban: no, I close eyes and meditate all night :) 2021-03-31 20:12:19 hehe 2021-03-31 20:12:48 https://gitlab.alpinelinux.org/alpine/docker-abuild/-/issues/61 2021-03-31 20:12:58 bl4ckb0ne: these patches are too big for alpine 2021-03-31 20:13:46 I could try to apply them but only for kernels which I'm trying to build for your board 2021-03-31 20:14:15 nah dont, it's for an older kernel 2021-03-31 20:14:31 we should stay with upstream to avoid trouble 2021-03-31 20:14:41 bl4ckb0ne: I mean, patches which will work 2021-03-31 20:15:26 I don't see activity from kobol people on #linux-rockchip 2021-03-31 20:15:36 i could try to poke the kobol bloke directly if kobol dont answer 2021-03-31 20:15:42 same in armbian 2021-03-31 20:16:43 if I have board I would write to kobol without any waiting ;) 2021-03-31 20:17:06 i sent the mail yesterday morning 2021-03-31 20:17:24 aha, good then 2021-03-31 20:17:57 some patience is usually needed with companies 2021-03-31 20:18:00 yep 2021-03-31 20:18:11 gonna mail the bloke next week if i have no answer 2021-03-31 20:18:13 they are fast like slow loris 2021-03-31 20:20:16 there seems to be some issue with the 2.5gbps usb eth port 2021-03-31 20:20:29 there's a few issues on the armbian build project 2021-03-31 20:20:37 but it's the 1gbps that doesnt work on my side 2021-03-31 20:21:14 lspci -k would help 2021-03-31 20:21:32 rebooting on the pmos kernel 2021-03-31 20:23:03 no output 2021-03-31 20:23:32 the RTL8211F 1gbps is connected with RGMI 2021-03-31 20:24:28 huh, who designed this board 2021-03-31 20:25:42 ACTION shrugs 2021-03-31 20:25:55 https://elixir.bootlin.com/linux/v5.11.11/source/drivers/net/phy/realtek.c#L289 2021-03-31 20:25:59 something around there 2021-03-31 20:29:04 https://github.com/rockchip-linux/kernel/issues/116 found this also regarding rockchip pci 2021-03-31 20:29:51 https://bugzilla.redhat.com/show_bug.cgi?id=1704945 2021-03-31 20:30:09 CONFIG_PHY_ROCKCHIP_PCIE=m 2021-03-31 20:30:34 what is this mess, there's CONFIG_PCIE_ROCKCHIP and CONFIG_PHY_ROCKCHIP_PCIE 2021-03-31 20:33:13 wtf both are y in the config-edge 2021-03-31 20:35:32 I don't have any issue on my board with this 2021-03-31 20:36:48 bl4ckb0ne: ok, started new kernel build, hope it will be ready tomorrow morning 2021-03-31 20:36:55 awesome 2021-03-31 20:37:02 5.11.11? 2021-03-31 20:37:06 yes 2021-03-31 20:37:08 nice 2021-03-31 20:37:39 I will upload it tomorrow and inform you here 2021-03-31 20:37:49 till then, good night 2021-03-31 20:37:53 good night! 2021-03-31 20:40:47 Marquel: It's not easy to figure out what's going on, because you're in such an early boot state 2021-03-31 20:41:17 Marquel: personally I don't really know how to proceed, so I would need help from others 2021-03-31 20:43:16 i wonder if anyone uses dracut on alpine 2021-03-31 20:43:38 re: nlplug-findfs lvm 2021-03-31 20:46:59 bovis: uhM, it seems that I built it 2021-03-31 20:47:21 APKINDEX.tar.gz wine-6.5-r0.apk wine-dbg-6.5-r0.apk wine-dev-6.5-r0.apk wine-doc-6.5-r0.apk wine-libs-6.5-r0.apk 2021-03-31 20:47:52 donoban: Great! So that leaves us with the question of why it works for you but not for me. 2021-03-31 20:48:03 are you building it from edge? 2021-03-31 20:48:38 donoban: My build takes place in a 32-bit chroot inside my 64-bit system. Both are Alpine. 2021-03-31 20:48:48 I'm not using docker. 2021-03-31 20:48:54 leibniz: opencc fails on mips64 because libmarisa-dev is not available there 2021-03-31 20:49:04 maxice8: * 2021-03-31 20:50:25 could you try upgrading to edge? 2021-03-31 20:50:33 donoban: Previous builds of 5.0 have been successful in this configuration. 2021-03-31 20:50:47 donoban: I'm unsure how to do that. Wiki? 2021-03-31 20:51:02 edit your /etc/apk/repositories 2021-03-31 20:51:17 comment your current repo and uncomment the edge ones 2021-03-31 20:51:26 and then run apk update && apk uprgade 2021-03-31 20:51:32 do it in the chroot install only 2021-03-31 20:52:16 I'm gonna check if I can use dabuild script for 32bits 2021-03-31 20:52:45 * `DABUILD_ARCH=x86|x86_64|aarch64|armhf|armv7`. Specifies architecture for build container. :D 2021-03-31 20:54:42 donoban: Was your successful build in 64bit? 2021-03-31 20:55:01 I tried to set up a chroot install before discovering docker-abuild and it's really simple 2021-03-31 20:55:13 yes, now I'm trying x86 2021-03-31 20:55:27 it's pulling docker x86 images 2021-03-31 20:56:02 I've updated the repos, and I'm trying my build again. For comparison, if it would be helpful, I can try to build it on my 64-bit as well. 2021-03-31 20:56:58 ok 2021-03-31 20:57:33 This will take a while. I'll get back to you when it's done. 2021-03-31 20:57:58 ok 2021-03-31 21:13:54 bl4ckb0ne: https://dev.alpinelinux.org/~mps/linux-edge-5.11.11-r1.apk 2021-03-31 21:23:18 I have a strange problem with keyboard layout 2021-03-31 21:23:41 all local chars work fine ç,ñ... but I can't put an accent on a vocal 2021-03-31 21:23:44 ´a :S 2021-03-31 21:24:51 the accent appears when I type it 2021-03-31 21:32:57 bovis: /home/buildozer/aports/main/musl/src/v1.2.2/crt/crt1.c:18: undefined reference to `main' 2021-03-31 21:57:38 donoban: With a complete update and using edge, 32-bit still didn't build. On my 64-bit, the build completed successfully. 2021-03-31 21:57:57 donoban: Yes that's the error I get. 2021-03-31 21:58:40 hehe, so we are at the starting point 2021-03-31 21:58:50 Haha unfortunately 2021-03-31 22:01:25 I would like to look at the source code 2021-03-31 22:01:37 but it seems that all build data was cleaned up 2021-03-31 22:01:59 oh its there nice 2021-03-31 22:02:36 I have all my build output in a file. All 45k lines 2021-03-31 22:04:25 uhM 2021-03-31 22:04:52 like this? 40185 158242 1285723 config.log 2021-03-31 22:07:50 donoban: Well yes I have the config log. I captured the output of abuild too so I could reference the error. 2021-03-31 22:12:45 the configure sript has near 23k lines :\ 2021-03-31 22:14:57 oich 2021-03-31 22:15:27 tried to restart abuild and it cleared all the src dir 2021-03-31 22:15:45 maybe I should have done 'abuild build' 2021-03-31 22:23:04 What might cause a 32bit program not to build where a 64bit program does? Is there anything more generic I could try that's not necessarily specific to my error or this build? 2021-03-31 22:27:22 uhM 2021-03-31 22:28:04 if try to run dabuild build to avoid to clean up it seems that it doesn't install depednencies 2021-03-31 22:29:24 maybe there is some problem with arch dependad code and altough we are using chroot or containers /bin/arch returns x86_64 2021-03-31 22:29:48 maybe you could have look trying to build it on a virtual marchine 2021-03-31 22:30:10 or just fake /bin/arch return value :| 2021-03-31 22:31:19 I don't know :\ and it's too late for me 2021-03-31 22:32:20 see you bovis 2021-03-31 22:35:58 for a second i thought you said "see you bois" and thought i was on discord heh 2021-03-31 22:36:26 donoban: Thanks for the help. Any further ideas tomorrow or whenever, I'll be around. 2021-03-31 22:48:45 hehe lol 2021-03-31 22:49:12 yes tomorrow continue but maybe you have better help on wine support 2021-03-31 22:49:29 good night guys