2025-12-01 01:19:01 hii 2025-12-01 01:20:43 anyone help me dor xdg_runtime_dir unset i cant start pipewire or others.. i enabled elogind, dbus and polkit alredy installed dbus-x11 polkit-elogind i using lightdm but xdg runtime is doesnt setting. i was look void handbook and artix forum but i cant 2025-12-01 01:20:58 anyone help me 2025-12-01 01:21:28 sorry for my english 🤭 2025-12-01 01:21:57 i was try a script with startx but doesnt work 2025-12-01 01:23:05 ohh and not wayland i using i3wm 2025-12-01 01:24:01 I've been trying stupid things for hours for two days now, and it still hasn't worked. It's quite frustrating. 2025-12-01 06:01:52 ssm: stateful firewalls are probably faster than /etc/hosts tbh, as they can look in a preexisting hashtable or similar, while /etc/hosts has to be re-parsed for every process, and probably even every time (I don't think musl caches it) 2025-12-01 06:02:34 ssm: so I think you're comparing a single hashtable lookup with a hundred or a thousand of sscanf()s and strcmp()s... 2025-12-01 06:04:03 ssm: (sorry, what I had in mind was things like ipsets or nft sets the actual table of states -- wasn't saying this about the case where there's just a bunch of individual rules) 2025-12-01 06:04:44 ssm: but that aside, the other issue is that 0.0.0.0 is not a special case as you probably expect -- it doesn't "reject" or "block", it gets returned as a regular IP address, and once the program attempts to connect it, 0.0.0.0 is treated the same way as 127.0.0.1 (localhost), as per BSD legacy 2025-12-01 06:05:23 though it doesn't matter until you run a dev server on localhost:80 and then all of the ad iframes are suddenly loading your dev website 2025-12-01 09:12:15 I think I will do it like before, nftables and some shitty script to adding IPs to some list or god knows, regular linux fuckery 2025-12-01 09:37:36 oh man, par is in testing repo 2025-12-01 09:37:38 dang 2025-12-01 12:04:20 well that was an easy compile 2025-12-01 15:49:18 pretty dead here 2025-12-01 15:49:25 maybe we could discuss trojan horses? 2025-12-01 15:49:37 https://i.ytimg.com/vi/ICH9KPVDeRA/maxresdefault.jpg 2025-12-01 15:49:44 hups wrong link ;/ 2025-12-01 15:52:40 Hm, is there any facility for LBU to generate hashes of new apkovls and for the initramfs to check that the hash matches before loading them? 2025-12-01 16:31:54 hey can someone help me configure cosmic-greetr 2025-12-02 02:10:48 Do cloud images (https://alpinelinux.org/cloud/) not use the serial console? 2025-12-02 02:14:25 which cloud are you running them in 2025-12-02 02:14:32 also which image are you using 2025-12-02 02:14:58 I'm using the nocloud images, running them on qemu-system-x86_64 with a seed generated with cloud-localds 2025-12-02 02:16:16 Basically running them via: https://paste.sr.ht/blob/8b90fe3c270157b4e8433a34b3a8f8ec363295f4 2025-12-02 02:18:27 I'm not sure if I'm using this correctly or not, but my main blocker right now is that I don't see any output, both with and without `-nographic` 2025-12-02 02:18:41 The only output is `Booting from Hard Disk...` 2025-12-02 02:21:45 from previous issues, it looks like the cloud images *only* use serial console (which can be a problem with clouds that don't have serial) 2025-12-02 02:22:42 My initial test was only using serial port, but the last output that I see is "Booting from Hard Disk...". CPU then remains at 100% 2025-12-02 02:22:54 (the above paste is a later test using a graphics console instead of serial) 2025-12-02 02:25:02 you are specifying -serial, right? (presumably -serial stdio ?) 2025-12-02 02:26:12 I've never used `-serial stdio`; the serial port is tied to stdio by default when using -nographic. 2025-12-02 02:26:32 I tried but "qemu-system-x86_64: -serial stdio: cannot use stdio by multiple character devices" 2025-12-02 02:26:55 Yeah, on all my preivous (successful) VM usages of all sorts, serial port to stdio is the default for qemu. 2025-12-02 02:27:04 ah 2025-12-02 02:27:18 (I don't use qemu directly) 2025-12-02 02:28:03 What do you use? 2025-12-02 02:28:29 proxmox or libvirt 2025-12-02 02:28:47 but also I don't run alpine, so I've never tested these images 2025-12-02 02:30:21 I'm mostly trying to use tiny-cloud+nocloud to script creation of VMs (for development/testing), but I'm not sure if I'm really taking the simplest approach here. 2025-12-02 02:50:12 Ah, I need to specify a firmware for the VM via "-drive if=pflash,format=raw,readonly=on,file=/usr/share/OVMF/OVMF_CODE.fd" 2025-12-02 05:37:13 Advert Advert 🚨🚨🚨... (full message at ) 2025-12-02 09:40:28 ye, kick them all, systemd users! 2025-12-02 13:06:34 lol what 2025-12-02 13:08:45 hope, they aren't all bad 2025-12-02 13:09:47 no, they are mostly innocent, like sheep 2025-12-02 13:11:35 ignorant? 2025-12-02 13:13:14 sheep are not ignorant, they have own destiny, they do what they were born for 2025-12-02 14:20:22 achill: could you look at aports!89115 once more? 2025-12-02 15:14:26 hi! does there happen to be an alpine 88x31 somewhere? 2025-12-02 15:46:05 lillianb_: what's that 2025-12-02 15:51:42 a logo/icon? 2025-12-02 15:54:11 weird size ;/ 2025-12-02 15:54:25 I assume that 1 at the end is important ;/ 2025-12-02 15:56:07 I believe those were those geocities style banners/buttons 2025-12-02 15:57:53 I found this: https://88x31.nl/history.html 2025-12-02 16:00:24 current value as emulating teletypwriter 2025-12-02 16:01:59 and ye, alpine should have own merch, selling 88x31 banners for 2,99 a piece. Maybe even 88x30 if somebody want safe some pixel 2025-12-02 16:02:11 Powerd by AlpineLinux 2025-12-02 16:02:41 or simply 80x24 with VGA colors :) 2025-12-02 16:03:11 obviously in sixel format 2025-12-02 16:06:15 lillianb_, https://88x31.nl/gifs/altavista.gif just edit the name on it 2025-12-02 16:07:45 calm down! 2025-12-02 16:07:57 this way nobody will make money! 2025-12-02 16:08:17 oops 2025-12-02 16:08:34 Well, I can change the name for 2.89 a piece 2025-12-02 16:08:56 he put on his website, somebody will click by mistake, server will crash 2025-12-02 16:10:28 ye, you can expand service, 2,89 handmade ready to publish logo or 1,49 chanding the name on premade one 2025-12-02 16:11:51 ok, you want to join services, could work 2025-12-02 16:13:43 Powered by OpenRC 2025-12-02 16:14:28 lillianb_: so what payment method do you prefer dear sir? 2025-12-02 16:14:48 https://neonaut.neocities.org/images/buttons/neocitizens/maxheadroom.gif that one is a bit more pricy though, because it's cooler 2025-12-02 16:15:56 premium 2025-12-02 16:39:24 it looks like wayland wm are dying? is there anything out of sway which does not require xwayland nonsense? 2025-12-02 16:39:30 lillianb_: will this work for you? https://0x0.st/KyHQ.png 2025-12-02 16:40:59 it's unofficial, based on website logo, you're welcome to have it as long as logo author is also okay with derivative use 2025-12-02 16:41:19 ye, it's unofficial! 2025-12-02 16:41:22 fake one 2025-12-02 16:41:38 if you want official one, authentic, handmade, pay 2025-12-02 16:44:13 Finally. I did build doublecmd. lazarus and fpc-src as well. U can test it. feel free to use the APKBUILD and public it to the repo. https://mega.nz/folder/CdZFzSyB#DCLTJo0pnTtnCymeqkaLHw 2025-12-02 16:45:40 Yeah, mio, please stop trying to kill business with open source 2025-12-02 16:47:03 quinq: payment accepted in smiles :) 2025-12-02 16:47:19 :) 2025-12-02 16:48:43 will try in grocery tomorrow 2025-12-02 16:51:39 hope: even xlibre is better then wayland 2025-12-02 16:52:08 lmao 2025-12-02 16:53:28 no idea, I can't smell xorg shit for a decade 2025-12-02 16:54:08 did not tried xlibre 2025-12-02 16:54:26 it's a fake fork with slop coding 2025-12-02 16:54:32 maybe it improved, but xorg was abandoned logn time ago, not sure if it's not too late for this 2025-12-02 16:54:34 IE political fork 2025-12-02 16:55:25 at least alpine is not political 2025-12-02 16:55:46 no political decision taken ;/ 2025-12-02 16:55:53 sorry to disappoint you, it is, as the rest of open source 2025-12-02 16:55:59 oh shit 2025-12-02 16:56:09 politics is so cool I wish politics was real 2025-12-02 16:56:58 funny ram I bought 2 months ago for 100 eur now cost over 300 eur 2025-12-02 16:57:07 it's better than hodling bitcoin 2025-12-02 16:58:06 probably less harmful too :D 2025-12-02 16:58:25 shit raspberry went up too bc. of this 2025-12-02 16:59:02 yeah, everything is going to shit, but at least we can generate bullshit at unprecedented speeds 2025-12-02 16:59:17 and open source! 2025-12-02 16:59:26 everything managed by intel's minixos ;/ 2025-12-02 17:00:53 when random software occupies 100GB of disk and 10GB of RAM they often say, hardware is cheap now 2025-12-02 17:01:01 and here we are 2025-12-02 17:01:22 even my lovely rpi went up of 15 eur 2025-12-02 17:03:57 now I understand why sobody want image just 88x31, everything bigger is not sustainable 2025-12-02 17:04:25 true, gotta ration these pixels 2025-12-02 17:04:41 anywho, we are getting hella offtopic here 2025-12-02 17:07:21 otherwise screen will freeze 2025-12-02 17:08:19 ona may then think it's bc. system hits swap space, better if screen is moving a bit 2025-12-02 17:20:47 mio: thanks! 2025-12-02 17:22:01 lillianb_: you're welcome, if you want the svg source can paste it as well 2025-12-02 17:22:31 but it's basically a resized version of the website logo with a few edits 2025-12-02 17:24:54 and if you want original brand new and gif, you know how much 2025-12-02 17:31:03 and don't worry, money will be used to support other political projects like xlibre and similar 2025-12-02 17:36:51 Everything's political, even scratching your knee 2025-12-02 17:38:32 Blembim501574135135242: nice password 2025-12-02 17:39:07 Maybe that's just a birthdate 2025-12-02 17:41:32 in some closed proprietary format? 2025-12-02 17:43:58 Just in us Epoch 2025-12-02 17:45:00 no idea, those kind of bots never speak 2025-12-02 18:11:01 is it normal for unprivledged usr to not be able to access `/sbin`? 2025-12-02 18:11:26 no 2025-12-02 18:12:04 the bin/sbin split is completely arbitrary and often downright wrong for many tools 2025-12-02 18:12:49 I regularly use tools that are often in sbin as a regular user, like ip and rndc 2025-12-02 18:15:12 ok, But I cannot access `ip` through a regular usr. So I misconfigured it somewhere. 2025-12-02 18:19:58 check your $PATH 2025-12-02 18:21:11 Though default is /usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin 2025-12-02 18:24:01 I see that atools (Leo/atools) is archived and that the shell apkbuild-lint script it ships has been marked as deprecated in the package of the same name. 2025-12-02 18:24:03 https://gitlab.alpinelinux.org/alpine/aports/-/blob/master/community/atools/APKBUILD#L36 2025-12-02 18:24:55 As far as I understand, atools-go (infra/atools-go) is now the preferred source. 2025-12-02 18:25:51 Question is, is there any technical reason why the former / deprecated atools package is still distributed in the repo? 2025-12-02 18:27:42 For the context, I'm about to push atools-go to the Arch Linux repository but I'm wondering if I should still keep the deprecated atools package around for some reason (as currently done in Alpine) or if I can just drop it (and have atools-go replacing it)? 2025-12-02 18:38:06 thx quinq For some reason. My Path only has `/bin` and no `/sbin` dirs. Even though `/etc/profile` includes the default entries 2025-12-02 18:40:04 Maybe you're using a weird shell that overrides this 2025-12-02 18:41:29 I did install fish. but path is fucked in ash too. 2025-12-02 18:42:24 hmm, curious fixing it for fix fixed in ash. fish is the problem 2025-12-02 18:42:53 fixing it for *fish fixed in ash. fish is the problem 2025-12-02 18:44:21 Yep, fish is the weird shell there 2025-12-02 18:47:02 any idea, why ash PATH will change. AFAIK fish store the variable in `.config/fish/var...` which ash should not parse 2025-12-02 21:04:37 Maybe something set $ENV? 2025-12-03 00:03:37 just wanna say thank you to all the maintainers, alpine is cool. 2025-12-03 00:04:00 I have a task for december adventure to poke at diskless ops. 2025-12-03 00:36:20 Hi 2025-12-03 02:34:17 How do you like Alpine Linux? 2025-12-03 06:44:18 I have tagged alpine 3.23.0_rc2 with the 6.18 kernel. please test if you can and report back any issues 2025-12-03 08:22:43 what’s the latest I can put off doing usr-merge? 2025-12-03 08:26:38 Plans have changed a lot so I don't think there is a fixed date yet. But it will definetly happen automatically at some point. 2025-12-03 08:26:45 usr-merge has been postponed, so you can just run 2.23 when it comes out 2025-12-03 08:27:59 (or even run the rc now and help test things) 2025-12-03 10:10:22 postponed? Does that mean I have to undo the usr merge when I've already done it, or is it OK to keep it if already done? 2025-12-03 10:23:10 no its not a problem to have a merged /usr now 2025-12-03 10:44:28 the only thing that doesn't work (and hasn't since 3.21) is to have / and /usr on different filesystems(/partitions/disks/whatever) 2025-12-03 10:59:03 thanks, good 2025-12-03 13:46:33 god 2025-12-03 13:46:37 you heathen! 2025-12-03 13:47:39 almost like we watched a play at oyal Albert's Hall and thought it was all imagination! 2025-12-03 13:47:46 royal* 2025-12-03 14:06:32 I agree 2025-12-03 18:40:30 Hello! I'm reading 3.23 release notes and I wonder about the upcoming (after 3.23) apk index v3 change. 2025-12-03 18:41:27 Will v2 index be still supported when this change happens? We (an ISV) provide pre-built alpine packages for all currently supported Alpine versions, and we build those repo indexes in a centralized manner, using one apk tool for all the repos. 2025-12-03 18:42:04 So I wonder if we can continue doing so in the future (at least until all Alpine releases get v3 index support), or do we need to prepare for a scenario where we will need to build v3 indexes for e.g. 3.24? 2025-12-03 18:42:30 apk-tools will keep supporting v2 indexes and packages afaik 2025-12-03 18:47:21 help wanted with proofread https://wwwtest.alpinelinux.org/posts/Alpine-3.23.0-released.html 2025-12-03 18:48:24 apk tools release notes link is not a clickable link 2025-12-03 18:53:21 achill: fixed 2025-12-03 18:53:26 merci 2025-12-03 18:53:41 otherwise dont have anything to nag 2025-12-03 18:53:42 :) 2025-12-03 18:53:58 thank you ikke 2025-12-03 18:55:57 thresh: from the release notes, this may be relevant: "Support for the original "alpine v1" packages and for broken v2 packages from foreign tools has been removed" 2025-12-03 18:56:17 So make sure that the packages / indexes you generate are valid 2025-12-03 19:05:50 ikke, yeah, I need to check if an ancient version of apk we use (2.10.4, on RHEL at that too) generates the contemporary ones 2025-12-03 20:36:14 original v1 packages... 2025-12-03 20:36:57 that is like 2009 something? 2025-12-03 20:37:01 https://apk-tools.sourceforge.net/ 2025-12-03 20:37:09 You know that better than I do :P 2025-12-03 20:37:59 Registered 2025-12-03 20:37:59 2005-05-07 2025-12-03 20:38:01 ha 2025-12-03 20:41:04 its official 2025-12-03 20:41:13 https://alpinelinux.org/posts/Alpine-3.23.0-released.html 2025-12-03 20:43:21 now for all the bugs that weren't found after extensive auditing to magically appear! 2025-12-03 20:50:59 reboot only if you really have to ;/ 2025-12-03 20:58:01 interesting that edge now stable is 6.17 and lts is 6.18 2025-12-03 20:59:46 -lts will keep being on 6.18 but -stable follows gregs releases 2025-12-03 21:11:10 i dont know how to properly format a text email 2025-12-03 21:11:18 its 2025 and people read the email from tiny screens 2025-12-03 21:11:24 plain text is futil 2025-12-03 21:14:32 ye weird, so many read it on 6,9' screens, in 2025! 2025-12-03 21:16:32 like they can't get proper terminal on the proper display at least reaching 70 characters ;/ 2025-12-03 21:18:57 should I run merge-usr on 3.22 and then update to 3.23, or does the order not matter? 2025-12-03 21:21:49 you can only run it on 3.23 2025-12-03 21:22:59 that settles that I guess lol 2025-12-03 21:23:08 thanks achill 2025-12-03 21:27:54 🙏 Alpine 3.23 2025-12-03 21:54:44 It is a solid release 🤓 2025-12-03 22:00:35 got one issue with testing usr-merge: 'udevd[375]: failed to execute '/bin/touch' '/bin/touch /run/udev/gdm-machine-has-virtual-gpu': Symbolic link loop' 2025-12-03 22:01:50 /bin/touch -> /bin/touch 2025-12-03 22:08:51 replaced with /bin/touch -> /bin/busybox 2025-12-03 22:09:07 all good now 2025-12-03 22:17:54 hello, i believe we might be missing /etc/dovecot/dovecot.conf for the dovecot package on v3.22, just a heads up. kindly note we had that on v3.21 2025-12-03 22:23:19 dovecot changed configuration format in the version that got into 3.22 2025-12-03 22:23:54 ohh i see, thanks 2025-12-03 23:00:44 probably good if someone would update at least the wiki page with that https://wiki.alpinelinux.org/wiki/Release_Notes_for_Alpine_3.23.0 2025-12-03 23:01:15 its already in https://wiki.alpinelinux.org/wiki/Release_Notes_for_Alpine_3.22.0#dovecot 🤷 2025-12-03 23:02:04 ACTION had dovecot down for a month and hadn't noticed :D 2025-12-03 23:02:48 ah, oh, I didn't read properly 2025-12-03 23:03:02 never mind then 2025-12-03 23:03:36 (I read the above as someone going from 3.22 to 3.23) 2025-12-04 00:07:28 Forza: Would you mind opening an issue with the usr-merge failure? Such things need to be fixed before migration can be done automatically 2025-12-04 03:03:29 ncopa: the standard thing is to wrap at 72. what can also be done is make the email multipart/html, which you can do if you feed the email to pandoc (eg 'pandoc -f markdown -t html --standalone'). mobile clients will automatically use the html part, which will reflow text. 2025-12-04 03:04:18 if there's a better compromise, i'm not aware of it. 2025-12-04 04:20:34 invoked: is that pandoc output meant to be sent as body or attachment? 2025-12-04 05:12:11 jvvv: it's plain text, not encoded. email starts Content-Type: multipart/mixed, sets a boundary. first boundary sets Content-Type: multipart/alternative, sets another boundary. first section using the 2nd boundary is text/plain. 2nd section text/html (output from pandoc). less complicated than it sounds. if you're doing it by hand. otherwise normally the MUA will/should do the 2025-12-04 05:12:21 multipart work for you. 2025-12-04 05:12:52 https://www.rfc-editor.org/rfc/rfc1341 2025-12-04 05:14:39 multipart/alternative basically says that both parts contain the same information (text/plain and text/html) and leaves it to the MUA to make a decision about which one to use by default. 2025-12-04 05:46:11 invoked: the mailing list does not allow html emails 2025-12-04 06:12:13 ikke: yeah. which is normal for lists. mailman & friends will allow it selectively based on sender, but perhaps not the hut. 2025-12-04 06:33:25 invoked: thanks for the expanded explanation 2025-12-04 08:01:46 oh look how nicely done plain text email ;/ 2025-12-04 09:21:44 v3.23.0_rc2-34-g319b82182d1 [http://dl-cdn.alpinelinux.org/alpine/v3.23/main] Shouldn't I have it without that _rc2 thing? 2025-12-04 09:35:59 Hello! 2025-12-04 09:38:36 it should 2025-12-04 09:38:52 and it will 2025-12-04 09:39:15 there are 300+ package needing rebuild due to go security update 2025-12-04 09:39:24 once those are ready it should update 2025-12-04 09:39:44 good, almost asked for money back, will wait a bit then 2025-12-04 09:44:55 actually, this is interesting 2025-12-04 09:45:50 the main repo is uploaded before community, and the go update only affects community 2025-12-04 09:46:21 so why did we end up with an APKINDEX with info as -rc2? 2025-12-04 09:47:15 it happens due to how we build the packages, the index and releases 2025-12-04 09:48:27 when tagging release, I first update alpine-base package commit that to git and then create a tag 2025-12-04 09:48:44 then I first do git push (which does not include the tags), and then I push the tag 2025-12-04 09:49:24 which means that the builder will first pick up the commit, without the tag, and start build 2025-12-04 09:49:34 and generate APKINDEX 2025-12-04 09:50:13 when generating the APKINDEX it gets the info from `git describe`, which at this point does not yet have the tag 2025-12-04 09:50:33 thats why it ends up as v3.23.0_rc2-34-g319b82182d1 2025-12-04 09:51:26 when that is built and uploaded, it will again do git pull, get the tag and at this point there is no new packages to build, so the APKINDEX is not regenerated 2025-12-04 09:51:30 it is left untouched 2025-12-04 09:51:43 but it will create the release, due to the tag 2025-12-04 09:52:14 standard complexity thing, hope you have fun ;/ 2025-12-04 09:52:31 nah, its fairly simple 2025-12-04 09:52:46 it will fix itself next time anything is pushed to 3.23-stable main repo 2025-12-04 09:59:12 ncopa: you can do git push --follow-tags, and it will automatically push any reachable tags as well 2025-12-04 09:59:36 Make sure you don't have any local tags around that you don't intend to push 2025-12-04 10:01:15 the second thing there is the reason why I do it this way 2025-12-04 10:01:33 Right 2025-12-04 10:01:48 anyway, thats the explanation why it ended up like this 2025-12-04 16:21:47 if you guys downloading from torrent and can't keep up with ratio...try downloading from xdcc bots, you can find network at https://xdcc-search.com that serves xdcc pacs - its all free and servers are running on high bandwidth 2025-12-04 16:23:52 faster than light 2025-12-04 16:24:03 wow a real warez distributor 2025-12-04 16:26:27 ye 2026 return of torrenting 2025-12-04 16:26:44 except it's -not- about torrents (but i don't want to dive offtopic here) 2025-12-04 16:27:14 ye that is not and it does not matter, who would download shit over irc 2025-12-04 16:27:45 but not only torrenting but also linux will have good year 2025-12-04 16:27:56 as win and mac os fucked up already too much 2025-12-04 16:28:28 what does -not- mean? what syntax is this? 2025-12-04 16:28:41 emphasis 2025-12-04 16:28:50 similar to _not_ or *not* 2025-12-04 16:28:58 yeah i should really learn to use _ or * 2025-12-04 16:29:16 i generally try to be markdown compatible in these things but -this- is an old habit 2025-12-04 16:29:31 and then jira and confluence immediately go strike through 2025-12-04 16:29:43 ye, plain text on irc is not enough! 2025-12-04 16:29:52 let's break it a bit with random hash 2025-12-04 16:30:17 thought it would be -strike through- if anything 2025-12-04 16:34:40 It's not not 2025-12-04 16:36:37 no? 2025-12-04 16:36:52 ~~strikethrough~~ 2025-12-04 16:46:16 Habbie: note that irssi can detect use of _this_ and *this* and apply formatting to them (underscore and bold, respectively) 2025-12-04 16:46:35 ah. but i don't want that :) 2025-12-04 16:46:47 it's presentation side only 2025-12-04 16:46:54 oh like that 2025-12-04 16:49:28 /italics/ is also a thing, but many terminal emulators don't support it (or the font doesn't) so it's not as useful 2025-12-04 17:26:50 hey, can someone tell me: is this article actual? Do we have new actual version of this? 2025-12-04 17:26:57 https://wiki.alpinelinux.org/wiki/How_to_build_the_Alpine_Linux_kernel 2025-12-04 17:34:20 Not everything mentioned there is needed but I would guess that it works 2025-12-04 17:35:12 $ time abuild -crK 2025-12-04 17:35:13 >>> linux-lts: Analyzing dependencies... 2025-12-04 17:35:13 ERROR: unable to select packages: 2025-12-04 17:35:13 >>> linux-lts: Validating /home/alex/c/aports/main/linux-lts/APKBUILD... 2025-12-04 17:35:13 >>> linux-lts: Building main/linux-lts 6.18.0-r3 (using abuild 3.15.0-r1) started Thu, 04 Dec 2025 19:27:19 +0100 2025-12-04 17:35:14 .makedepends-linux-lts-20251204.182723: 2025-12-04 17:35:16 masked in: cache 2025-12-04 17:35:18 satisfies: world[.makedepends-linux-lts=20251204.182723] 2025-12-04 17:35:20 bpftool (no such package): 2025-12-04 17:35:22 required by: .makedepends-linux-lts-20251204.182723[bpftool] 2025-12-04 17:35:24 >>> ERROR: linux-lts: builddeps failed 2025-12-04 17:35:26 Command exited with non-zero status 1 2025-12-04 17:35:28 real 0m 4.04s 2025-12-04 17:35:32 user 0m 2.64s 2025-12-04 17:35:34 sys 0m 1.00s 2025-12-04 17:35:36 that's what I get if I do everything 1:1 2025-12-04 17:36:21 it seems something is missing 2025-12-04 17:40:03 anaumov: what release are you on? bpftool moved from community to main in 3.23 2025-12-04 17:40:48 ikke, I use 3.22 2025-12-04 17:41:06 Make sure you use the APKBUILD from the 3.22-stable branch then 2025-12-04 17:43:45 ikke, can you tell me how I can check it? :) 2025-12-04 17:44:01 If you ran this: git clone --depth 1 https://gitlab.alpinelinux.org/alpine/aports.git 2025-12-04 17:44:06 You are on the master branch, which is for edge 2025-12-04 17:48:56 anaumov: git config --add remote.origin.fetch +refs/heads/3.22-stable:refs/remotes/origin/3.22-stable 2025-12-04 17:49:01 git fetch --depth=1 2025-12-04 17:49:05 git checkout 3.22-stable 2025-12-04 17:52:13 ikke, thanks! it solved my problem. Now I'm building the kernel. Thank you 2025-12-04 18:07:29 Any of you able to run alpine on amlogic tv box ? 2025-12-04 18:07:49 I hit day #4 trying to figure it out and I'm non the wiser 2025-12-04 18:48:30 It looks like both lz4 and zstd are set as crypto modules in linux-lts kernel, so what would be the Alpinic way of ensuring they are available at boot to be used with zswap on the kernel commandline? 2025-12-04 18:49:01 Setting them as modules directly in /etc/default/grub didn't seem to work, so do they rather need to go into /etc/mkinitfs/mkinitfs.conf 2025-12-04 19:03:34 sa 2025-12-04 19:10:50 Is our features line in mkinitfs.conf the same as Modules in mkinitcpio? 2025-12-04 19:11:03 https://wiki.archlinux.org/title/Zswap#Compression_algorithm 2025-12-04 19:11:17 Saijin_Naib[m]: features refers to files in features.d 2025-12-04 19:11:35 Ooh, okay 2025-12-04 19:11:48 I'm doing this wrong 🤣 No surprise 2025-12-04 19:13:29 So, add a lz4 feature entry, and then a corresponding lz4.modules file in features.d and doas apk fix mkinitfs 2025-12-04 19:20:12 is it just me or anyone else have to type password when using doas after upgrade (to edge)? 2025-12-04 19:21:06 i have "permit nopass user" in /etc/doas.conf 2025-12-04 19:22:28 i only have to type passwd once pr terminal/shell 2025-12-04 19:23:22 Lets see if I brick it 2025-12-04 19:25:31 might be my unusual setup? i run diskless, so log in as root and create a user, and exec su user startx.. 2025-12-04 19:25:39 No brick, but no work. Ah well 2025-12-04 19:25:51 frag: just you 2025-12-04 19:26:26 uffda 2025-12-04 19:26:56 but what you mentioned is root using doas, not user 2025-12-04 19:27:56 is there a way to know which linux-firmware-X is used ? 2025-12-04 19:28:16 whami says "user" 2025-12-04 19:28:23 whoami 2025-12-04 19:28:58 didnt have a problem on 3.22, i updated from it to edge btw, havent installed 3.23 yet, well i guess thats edge atm.. 2025-12-04 19:29:37 actually im not 100% sure i didnt have a problem on 3.22, not sure of the timeline.. oh well.. 2025-12-04 19:30:08 and how it comes that you have to type pass once? with nopass you should have not type it at all 2025-12-04 19:30:31 what you mentioned is more like persist, not nopass 2025-12-04 19:30:34 thinking the same thing 2025-12-04 19:31:51 but from root su user I think it should not ask pass 2025-12-04 19:32:02 no doas needed 2025-12-04 19:32:57 yeah, i figure since i su startx as the user, its like i logged in as user and started startx 2025-12-04 19:33:33 maybe im not very articulate atm.. 2025-12-04 19:37:43 i do something like: su -l user -c 'export ENV=/path/user;startx /usr/bin/dwm -- -ardelay 200 -arinterval 30' 2025-12-04 19:38:13 i guess its not so bothersome to type passwd now and then tho.. 2025-12-04 19:42:06 hmm wondering if its asking for the users password.. 2025-12-04 19:44:55 Currently used, or needs to be used? 2025-12-04 19:45:33 If needs to be, I look at dmesg for errors about missing fimrware files, then use the Package Contents part of pkgs.alpinelinux.org to find where that module lives, and then apk add that linux-firmware-* pacakge 2025-12-04 19:46:01 There must be a more elegant way, because setup-alpine seems to pick sane linux-firmware-* packages for attached devices at install 2025-12-04 19:46:21 But yeah, that's what I do (did). Now, I just apk add linux-firmware and be done with it haha 2025-12-04 19:46:50 Having the firmware files more compressed than they used to be is a huge help on smaller eMMC and it means I don't need to be as conscientious when I hook up new devices 2025-12-04 19:48:05 So, cp cryptsetup.modules to lz4.modules and editing the lines to point to lz4 instead of * did not work 2025-12-04 19:48:16 I'm out of my depth 2025-12-04 19:48:49 Are there docs on how these need to be setup, because presumably i915 also needs a features.d file as well for early KMS and some other features to be supported, if ArchWiki applies here 2025-12-04 20:29:00 do you need root passwd if you dont have any ssh or similar service running? i guess its a fail safe... 2025-12-04 20:37:34 no disk, no password, no root, like a pro ;/ 2025-12-04 20:38:47 im slowly cutting out computers ;p 2025-12-04 20:39:00 its not working tho.. 2025-12-04 20:42:50 sold ssd disk so you work less and have more time for linux fuckery? 2025-12-04 20:44:56 you could maybe normal install and put junk to tmpfs 2025-12-04 21:10:14 run rpi so have sd storage :] gotta slp, congratz on new release guys! 2025-12-04 21:13:44 I have two, server and desktop and no issue with sd 2025-12-04 21:14:28 on desktop I just put .cache and /var/cache into tmpfs and downloads and shits go there 2025-12-04 21:14:49 it not only flies, but everyone on x86 envy my setup 2025-12-04 21:21:57 I used ssd on rpi 5 but it was pointless complexy so went back to sd card 2025-12-05 02:56:16 I'm guessing that because rc-service -U is undocumented it's not ready for use? 2025-12-05 02:57:18 and that apk seems to install user-level services to /etc/user/init.d whereas rc-update(8) says all services must be in /etc/init.d or /usr/local/etc/init.d? 2025-12-05 02:59:45 I'm getting errors trying to stop/start anything in /etc/user/init.d 2025-12-05 03:28:01 I'm having this really strange issue, where after upgrading from 3.22 to 3.23, when my laptop wakes from sleep, shortly after seeing the KDE Plamsa lockscreen, my laptop reboots after what seems to be a forced instant restart (since usually it takes multiple seconds to shutdown before restart, but with this bug it almost immediately boots up after seeing the lockscreen) 2025-12-05 06:20:59 also had no problem using sd card, so simple, so tiny! 2025-12-05 07:24:04 one would almsot say, so cute ;/ 2025-12-05 09:32:56 Kladky, so you mean that 3.23 improved drastically reboot time? 2025-12-05 10:14:34 Kladky: Maybe your DM is crashing and rebooting? 2025-12-05 10:14:47 or KDE? 2025-12-05 10:36:57 Is keepassxc autotyping on X11 broken for anyone else? 2025-12-05 11:33:19 f_: if by DM you mean Desktop Environment, then I don't know. It seems weird to me that waking from sleep would make it crash so hard that it immediately reboots 2025-12-05 11:33:46 quinq: no, when I reboot normally, it takes maybe 6-7 seconds to shutdown 2025-12-05 11:33:57 but when this bug happens its instant 2025-12-05 11:34:51 I just see the lockscreen for a second or two, then the screen is off and the keyboard backlight turns on, signalling that it's booting again 2025-12-05 11:38:13 So yes, it improved reboot time :D 2025-12-05 11:38:23 Went from 6-7 seconds to instant 2025-12-05 11:42:44 should be mentioned in changelog 2025-12-05 11:44:04 Alpine 3.23.0 it never sleeps, reboots instantly 2025-12-05 11:45:20 6-7 seconds to shutdown? hah, that time I install whole system on rpi ;/ 2025-12-05 13:23:51 Kladky: I eman display manager, e.g. sddm 2025-12-05 13:24:12 Highly doubt it's a reboot 2025-12-05 13:26:43 btw, is NymphCast broken for anyone else? 2025-12-05 13:27:00 On aarch64 at least I cannot install it due to some 'poco' library that is not in repos 2025-12-05 14:17:48 Any of you run alpine arm on amlogic s912 soc ? 2025-12-05 14:22:54 no, just you 2025-12-05 14:24:22 I wish and hope to! 2025-12-05 14:28:28 postmarket has some entry for that crap 2025-12-05 14:52:23 f_: well, I get kicked back to grub, after seeing my laptop's manifacturers logo. 2025-12-05 14:52:47 I think I remeber seeing sddm failed to start error when shutting down my system normally. 2025-12-05 14:53:23 hope: it's not crap ;) 2025-12-05 14:53:32 I have kde set to auto-login on boot, since my disk is encrypted, I don't want to write my password twice 2025-12-05 14:54:05 gorus: what device do you have? It might be better to run postmarketOS, it's a distro based on alpine but with support with tricky devices like yours 2025-12-05 14:54:29 apart from systemd support and some other things you can basically treat it like an alpine box 2025-12-05 14:54:40 (and it uses the alpine repos too) 2025-12-05 14:54:50 Kladky: hmm .. strange 2025-12-05 14:55:06 Do you have a way of taking a video when it happens, maybe that could give a better idea 2025-12-05 14:55:32 Amlogic s912 that uses q200 dtb, running armbian on it perfectly but id prefer alpine 2025-12-05 14:55:52 gorus: yes, which specifically 2025-12-05 14:55:55 Had my daughter run libreelec on it as well to watch her cartoons 2025-12-05 14:56:04 T95z-plus 2025-12-05 14:56:43 there's this https://wiki.postmarketos.org/wiki/Amlogic_Generic_AArch64_TV_Box_(amlogic-aarch64-tvbox) 2025-12-05 14:56:57 but apart from ugoos-am3 no q200 generic port 2025-12-05 14:57:14 I can run alpine by replacing root partition with alpine but would prefer to do it properly 2025-12-05 14:57:19 I can add it if that's what you want to use, just 4-5 lines 2025-12-05 14:58:28 f_: yeah, I can record it with my phone I guess 2025-12-05 14:58:29 yeah postmarketOS is basically alpine with a few differences and things added. But other than that it can be treated as a regular alpine box 2025-12-05 14:58:39 not sure where i'd upload it though 2025-12-05 14:59:04 Kladky: your bnc/client doesn't have file upload? 2025-12-05 14:59:23 I guess maybe x0.at or 0x0.st might work then 2025-12-05 14:59:29 Idk, I use soju and Halloy 2025-12-05 14:59:35 howIfuckedup.org is good for uploading 2025-12-05 15:00:11 hope: maybe stop posting unhelpful messages? 2025-12-05 15:00:16 To be honest id love to be able to follow this: https://wiki.alpinelinux.org/wiki/DIY_Fully_working_Alpine_Linux_for_Allwinner_and_Other_ARM_SOCs 2025-12-05 15:00:31 But something a bit more current 2025-12-05 15:00:49 hmm you can look at how it's done in pmOS 2025-12-05 15:01:16 The main thing is how to boot the thing, and the kernel also, IIRC alpine's kernel is missing some amlogic things 2025-12-05 15:02:18 Well for me the main thing is to sign the compiled u-boot with amlogic security stuff 2025-12-05 15:02:34 Oh you boot u-boot directly as bl33? 2025-12-05 15:02:42 All of the info I found is outdated with no working links 2025-12-05 15:03:08 I assumed you kept the existing vendor u-boot intact since you said it ran LibreELEC and runs armbian which don't really provide u-boot-as-bl33 2025-12-05 15:03:58 (fyi: I help maintain support for amlogic devices in pmOS) 2025-12-05 15:29:11 f_: https://x0.at/J4ke.mp4 2025-12-05 15:29:46 After the video ends showing gram (the name of the model of the laptop), it just boots like normal to GRUB 2025-12-05 15:31:31 never saw condom on the keyboard 2025-12-05 15:45:45 Kladky: strange.. 2025-12-05 17:01:52 byebye matrix 2025-12-05 17:07:10 much better now 2025-12-05 17:08:13 cleaner here to be honest 2025-12-05 17:12:37 actually the bridge seems to have come back, but it went down like an hour ago 2025-12-05 17:13:26 every downtime of bridge is welcome 2025-12-05 17:27:12 I am trying to load 3.23 riscv64 on a visionfive 2 via serial ttl, sdio(boot fail ffff...) and qspi(gets to grub, then dies) modes try to load and fail. Cannot find anything on web about visionfive pre-reqs like flash rom or switch settings. Anything to point at here? 2025-12-05 17:28:39 richhenn: You may want to ask in #alpine-riscv64 (and be a bit patient) 2025-12-05 17:29:16 thanks. I didn't know there was a channel for this. Will head over there. 2025-12-05 17:29:24 No worries 2025-12-05 17:31:21 ye, send them all out 2025-12-05 17:31:29 less people, better, good move 2025-12-05 17:33:55 hope: final warning, if you have something useful to contribute, feel free to stay, but please stop with these useless messages 2025-12-05 17:35:07 final? you mean first warning 2025-12-05 17:35:58 f_ also told you to stop 2025-12-05 17:36:44 also. okay, so that was first and this yours is final, understand 2025-12-05 17:37:45 but feel free to write me anytime you feel, I don't mind if it's useless 2025-12-05 17:38:25 this is a help channel. #alpine-offtopic exists 2025-12-05 17:40:58 i'm trying to automate installation in libvirt using the alpine-virt image following these instructions: https://www.skreutz.com/posts/unattended-installation-of-alpine-linux/ 2025-12-05 17:41:27 however, i noticed that setup-alpine (or setup-disk) is always asking for permission to erase the disk even though i'm using an answers file for setup-alpine 2025-12-05 17:41:50 there don't seem to be any flags to turn off the confirmation 2025-12-05 17:41:58 any ideas? 2025-12-05 17:43:06 if you run `setup-alpine -ef answers` where DISKOPTS="-m sys /dev/vda", you'll see the same. it always wants to ask "WARNING: Erase the above disk(s) and continue? (y/n) [n]" 2025-12-05 17:44:10 big_robot: just a thought, have you considered using the (no)-cloud images? 2025-12-05 17:46:23 big_robot: if you look at /usr/sbin/setup-disk, it checks $ERASE_DISKS 2025-12-05 17:46:37 ikke: oh good grief, i wasn't aware of those. i was just trying to make the "virtual" image work. let me look into those. 2025-12-05 17:47:03 in theory, if your disk is in ERASE_DISKS, it will return from the confirmation without prompting 2025-12-05 17:47:54 invoked: wow, just tried it. that works, thank you! 2025-12-05 17:48:02 np 2025-12-05 18:01:38 invoked: in the context of this article, where would be the best place to set the ERASE_DISKS environment variable? When manually doing it in the shell it works fine. I've tried setting it at the start of the line that calls setup-alpine (/etc/local.d/auto-setup-alpine.start) but it hangs there (or crashes when stdin is closed) 2025-12-05 18:01:47 https://www.skreutz.com/posts/unattended-installation-of-alpine-linux/ 2025-12-05 18:21:06 i guess i would try putting it in the answers file first. 2025-12-05 18:23:11 at least, that follows what i think users should expect. ERASE_DISKS is not part of the example answers file in setup-alpine, though. 2025-12-05 18:27:07 yeah, no dice in answers file. :( 2025-12-05 18:44:43 big_robot: hm, ok without looking at it (sorry, busy) i might be missing something. i would try what ikke suggested (different image like -cloud) which might handle that differently 2025-12-05 18:45:40 else maybe setup-disk needs a change where setup-alpine can pass it an arg that forces setup-disk to destroy the disk(s) without prompting. 2025-12-05 18:46:21 and that should still be exposed in the answers file, if so 2025-12-05 19:12:40 'apk query --fields=version busybox' takes 450ms o.O 2025-12-05 19:38:42 big_robot: to skip the erase_disks confirmation, prefix the setup-alpine command like so: ERASE_DISKS=/path/to/targetmedia setup-alpine -f /path/to/answer_file 2025-12-05 19:39:42 i've updated the wiki page... it was specifying to use the '-c' switch for that usage, but only '-f' is appropriate in this case 2025-12-05 19:40:48 jvvv: yeah tried that, which works interactively but for some reason doesn't work in my overlay. i'm thinking cloud image might be easier for my situation. 2025-12-05 19:41:44 i set it up verbatim from that link i posted above, except using the v3.23 release 2025-12-05 19:42:02 if you feel the urge to dive in 2025-12-05 19:42:41 huh, when i did it, the erase_disk confirmation was skipped... weird 2025-12-05 19:43:55 really? dang. so the whole install worked and it rebooted ready to go? 2025-12-05 19:46:34 i didn't follow your instructs from that link, i just did `ERASE_DISKS=/dev/sda setup-alpine -f answerfile` for a fresh vm install 2025-12-05 19:47:50 perhaps problem could stem from your answerfile having DISK_OPTS=none? 2025-12-05 19:49:36 i'm going for a smoke break, after will look at your link again and see if i can reproduce 2025-12-05 19:53:05 jvvv: ah ok, thank you. no worries if you don't feel like it. i appreciate your advice. My DISK_OPTS="-m sys /dev/vda". i have been able to get that working just as you described if i type it manually, but automating this install is where it hangs up. 2025-12-05 19:54:55 right, but i wonder if you editted the answer_file to add the disk_opts to it, if it would work as expected 2025-12-05 19:54:58 sorry, brb 2025-12-05 20:30:42 big_robot: are you using the exact same answerfile as described on that link? i ask because with that answerfile runs to completion fine, but it is diskless so ERASE_DISKS is irrelevant for that. 2025-12-05 20:32:27 jvvv: my answer file is identical except DISKOPTS="-m sys /dev/vda" 2025-12-05 20:33:45 booting it with this: virt-install --name $host --disk size=4 --location alpine-with-overlay.iso,kernel=boot/vmlinuz-virt,initrd=boot/initramfs-virt --ram 1024 --os-variant alpinelinux3.21 --network default --graphics none --extra-args "console=ttyS0,115200n8 quiet modules=virtio,virtio_blk,virtio_net,virtio_pci" 2025-12-05 20:34:03 also, the line in the script that run setup-alpine, if it is to include the ERASE_DISKS environ variable, then it should be run like `ERASE_DISKS=/dev/vda timeout 300 setup-alpine -ef path_to_answ_file1` or `timeout 300 env ERASE_DISKS=/dev/vda setup-alpine -ef path_to_answ_file` 2025-12-05 20:34:36 i tried the first one, let me try the second where you throw `env` in there 2025-12-05 20:38:39 i've tried both to success running on command line... i use raw qemu and have not used libvirt, so i'm not sure about actually following those instructions verbatum, but i'll try it with raw qemu in just a moment 2025-12-05 20:42:35 jvvv: if i remove the line for closing standard input "exec 0<&-" it hangs (for 300 sec). if i leave it in there, it crashes out pretty quickly and fails to install: "* Starting local ... [ !! ]". 2025-12-05 20:46:04 huh, i'm almost ready to run an initial test, shouldn't be long 2025-12-06 00:08:51 big_robot: sorry, i got distracted by some other stuff. i will have to come back to this later 2025-12-06 00:58:47 big_robot: even so, this has piqued my interest. i've meant to get deeper into this kind of thing for custom images. i'll be looking at this more from the aports/scripts/* side. 2025-12-06 03:12:17 how to get raspberry pi camera module working on alpine? While googling i found a tutorial on an alpine wiki user page but the packages listed don't exist 2025-12-06 03:14:31 okay rpicam-apps exists, but it looks like the correct libcamera package is libcamera-raspberrypi 2025-12-06 07:11:20 how do you use the alpine users email list? Are you required to register? If I click new post my email client starts. I do not get to see the alpine user email address. 2025-12-06 07:14:35 https://lists.alpinelinux.org/~alpine/users 2025-12-06 07:16:31 users@lists.alpinelinux.org Is this the email address I have to use? 2025-12-06 08:15:17 It appears sign up is not required. It appears the email address to use is ~alpine/users@lists.alpinelinux.org 2025-12-06 08:21:46 * It appears to sign up 2025-12-06 08:31:15 what browser and setup are you using? i like surf but it doesnt work on every site, i like chromium but it privacy neightmare, dont love firefox/librewolf, kinda slow, depends on dbus++ 2025-12-06 08:31:41 oh and qutebrowser which i used for years just crashes now, at least last i checked 2025-12-06 08:33:04 i guess the webkit browsers has the same problem as surf, at least on arm 2025-12-06 08:54:12 Yeah :/ 2025-12-06 09:09:37 surf is just webkit 2025-12-06 09:10:25 frag: maybe arkenfox works with the firefox mobile config? 2025-12-06 09:10:46 oh you don't like it I misread 2025-12-06 09:11:29 there's really no other good options except terminal browsers. links graphical mode doesn't play nice with touch, same with netsurf 2025-12-06 09:12:52 oh i use arm/rpi as a desktop, i dont have touch, but yeah terminal browser are nice, but only works on a few sites eh 2025-12-06 09:17:37 gtkwebkit's basically the best you can do short of chrome or firefox. khtml/qtwebkit is dead, qtwebengine is blink based (traitors!) 2025-12-06 09:20:10 And they actually provide a usable API 2025-12-06 09:31:29 okok 2025-12-06 09:58:30 ladies, way to disable apk.log nonsense? 2025-12-06 10:12:06 okay I assume I got it with apk command --no-logfile 2025-12-06 10:12:29 while apk command help tells me --logfile[=bool] 2025-12-06 10:20:05 at least if it would respect syslogd -C option to go into memory 2025-12-06 12:11:27 perl 5.40.3 -> 5.42.0 not mentioned in release notes :( 2025-12-06 12:14:46 hope: "Write transactions are now logged to /var/log/apk.log, see also logfile configuration option" - https://gitlab.alpinelinux.org/alpine/apk-tools/-/releases/v3.0.0 2025-12-06 13:03:51 lopid: that resulted to what I wrote above 2025-12-06 13:05:35 but what I did not saw was support for config file 2025-12-06 13:06:04 instead of making mess in aliases 2025-12-06 13:07:14 no idea what that config should look like 2025-12-06 13:08:42 no existing one? 2025-12-06 13:11:33 no, but found some things in manual 2025-12-06 13:11:37 https://gitlab.alpinelinux.org/alpine/apk-tools/-/blob/master/doc/apk.8.scd#L434 2025-12-06 13:11:52 ye, that 2025-12-06 13:12:00 not really one line there, but 2025-12-06 13:12:11 I assume ++ is formating 2025-12-06 13:12:18 for man 2025-12-06 13:12:18 that's the source for the man page 2025-12-06 13:12:39 ye 2025-12-06 13:13:00 it does say one option per line 2025-12-06 13:13:22 exactly one long option per line 2025-12-06 13:13:27 sadly it can't be short 2025-12-06 13:14:01 so in my case I assume just no-logfile 2025-12-06 13:14:10 decrease font size and it gets shorter 2025-12-06 13:14:26 ye works 2025-12-06 13:14:41 nice trick you hacker 2025-12-06 13:17:08 but to respect syslog would be nicer no? 2025-12-06 13:23:22 are you saying it sends to syslog regardless of logfile value? 2025-12-06 13:24:46 no, but can it? 2025-12-06 13:25:03 instead of file 2025-12-06 13:25:29 as currently I setup syslog with -C to go into memory instead all those files in /var/log 2025-12-06 13:26:13 you *want* apk log messages to go to syslog? 2025-12-06 13:27:13 I would *prefer* that for now, _yes_ 2025-12-06 13:27:47 everything should go there no? 2025-12-06 13:29:32 otherwise syslog feels pointless and everything can manage logs of itself 2025-12-06 13:29:53 syslog is always optional 2025-12-06 13:30:38 with apk-tools too? 2025-12-06 13:32:18 you don't even need a syslog installed 2025-12-06 13:32:35 i don't think alpine comes with one installed by default 2025-12-06 13:32:41 well it's part of busybox 2025-12-06 13:32:56 running/configured, then 2025-12-06 13:33:04 ye 2025-12-06 13:33:13 then what? 2025-12-06 13:33:22 where all that shit will go? to the void? 2025-12-06 13:33:54 nowhere 2025-12-06 13:34:14 that sounds promising 2025-12-06 13:34:17 you have no mouth but you must scream? 2025-12-06 13:35:04 but those programs have mouth 2025-12-06 13:35:09 and they scream 2025-12-06 13:35:17 I assume syslog just catches it 2025-12-06 13:35:30 nobody to hear 2025-12-06 13:35:43 it's in space 2025-12-06 13:36:36 so without it I assume one has to setup things if wants in particular program? 2025-12-06 13:36:48 and that one has to support that 2025-12-06 13:37:32 if one wants syslog, configure it 2025-12-06 13:38:07 I mean like no syslog, but let's say log mail nonsense 2025-12-06 13:39:34 I swear I wanted to ask something, but I already forgot 2025-12-06 13:40:08 oh 2025-12-06 13:40:36 there is often or at least once for init thing like running in foreground 2025-12-06 13:40:42 like syslog has it 2025-12-06 13:40:51 I am confused 2025-12-06 13:42:01 as syslog by default runs with option -n somehow 2025-12-06 13:42:05 which means foreground 2025-12-06 13:42:26 on other hand in init.d/syslog there is command_background=true 2025-12-06 13:45:02 should it be like that? 2025-12-06 13:45:20 I assume command_background=true is like regular thing for services 2025-12-06 13:45:31 but that -n option for syslogd 2025-12-06 13:45:51 and also why it's in init.d and not conf.d as option 2025-12-06 13:46:38 it might be necessary to run in the foreground so openrc can manage it. i haven't looked 2025-12-06 13:48:23 somethign like that I thought 2025-12-06 13:48:44 interesting that other programs don't need that? 2025-12-06 13:49:48 true, not all programs are the same 2025-12-06 13:50:01 that's true 2025-12-06 13:50:22 true 2025-12-06 13:51:07 comand_background=true I assume then means someting else? like don't wait, start, put to background and continue with other crap 2025-12-06 13:51:24 that was question 2025-12-06 13:51:37 forgot ?, so here it is ? 2025-12-06 13:54:36 https://github.com/OpenRC/openrc/blob/master/service-script-guide.md 2025-12-06 13:58:48 good, any idea why is then -n option mentioned? 2025-12-06 13:59:40 maybe for case it does not go to background itself? 2025-12-06 13:59:57 so open-rc like handle that thing? 2025-12-06 14:11:10 remember like a hundred years ago there was attempt for new init 2025-12-06 14:11:14 god knows where it is 2025-12-06 14:28:54 ERROR: patchutils-bash-completion-0.4.4-r0: trying to overwrite usr/share/bash-completion/completions/interdiff owned by bash-completion-2.17.0-r0. 2025-12-06 14:29:07 upon upgrading a machine 3.22 -> 3.23 2025-12-06 14:33:57 did it stop the upgrade? 2025-12-06 14:37:43 no, it's just 1 error 2025-12-06 14:37:54 but unresolvable aside from removing bash-completion. 2025-12-06 14:38:18 i checked gitlab. 2025-12-06 14:41:37 note: not really an issue for me, i don't use bash as an interactive shell but rarely. but i assume this would create noise for a lot of people. 2025-12-06 14:54:38 invoked: https://gitlab.alpinelinux.org/alpine/aports/-/merge_requests/94309 2025-12-06 14:55:47 it would be nice to have a issue for that in the future :) 2025-12-06 15:04:12 well, i'm trying to get out the door to take care of both my geriatric mother and mother-in-law. this is what i do on weekends now 2025-12-06 15:09:39 thanks for the information 2025-12-06 18:50:08 f_: when i run `service sddm status`, it says it's stopped, and when I look at `/var/log/sddm.log`, the last line is `[18:43:22.001] (WW) HELPER: Failed to write utmpx: File exists` 2025-12-06 20:06:14 ajaj 2025-12-06 20:06:24 I somehowe apk del alpine-base 2025-12-06 20:06:32 it removed some crap ye 2025-12-06 20:06:40 how to put that shit back? 2025-12-06 20:07:22 ideally before I reboot ;/ 2025-12-06 20:13:21 that's a nice motivation to reinstall an os :) 2025-12-06 20:13:59 I do that too often 2025-12-06 20:14:07 ye, it's the fastest way 2025-12-06 20:14:18 yeah it would only be fixable if you installed apk-tools-static before hand methinks 2025-12-06 20:14:34 hohoh calm down matrix users! 2025-12-06 20:14:56 If you are able to still download something, getting apk-tools.static indeed is what you need 2025-12-06 20:15:13 otherwise boot into a live image, and fix it through there 2025-12-06 20:15:16 but I have apk-tools 2025-12-06 20:15:31 oh okay, it needs other crap 2025-12-06 20:18:08 hey, is `rust-lldb` correctly packaged? i'm trying to debug a program, it says 2025-12-06 20:18:11 error: module importing failed: invalid pathname '/usr/lib/rustlib/etc/lldb_lookup.py' 2025-12-06 20:18:23 but the only lldb_lookup.py is in /usr/share/rust/etc/lldb_lookup.py 2025-12-06 20:19:46 if I know how to extra .apk nonsense 2025-12-06 20:19:51 extract 2025-12-06 20:20:16 tar xzf 2025-12-06 20:21:05 invalid magic, short read 2025-12-06 20:23:27 okay nvm, slowly getting there 2025-12-06 20:25:18 well probably installationg will fix 2025-12-06 20:25:23 as I can't doas/su anyway 2025-12-06 20:26:17 and fuckering with chroot over usb would take probably longer than install I assume 2025-12-06 20:26:26 did not that for a century 2025-12-06 20:27:03 can't uninstall fucking pythong, but alpine-base went out like morning shit 2025-12-06 20:29:41 hii, i am new to non systemd init's i was add elogind to services but i getting i was look wiki but not working 2025-12-06 20:31:39 i trying pipewire, wireplumber but not working and what is the runlevel? wiki saying that 2025-12-06 20:31:59 https://wiki.alpinelinux.org/wiki/PipeWire 2025-12-06 20:32:23 afaik you need to add openrc --user gui to /etc/xprofile, which i use for stumpwm+xfce 2025-12-06 20:33:46 idk if wayland has the same or not 2025-12-06 20:33:58 no x11 2025-12-06 20:34:00 bspwm 2025-12-06 20:34:12 with lightdm 2025-12-06 20:34:24 i thought bspwm was x11 2025-12-06 21:08:58 luckily did it over chroot 2025-12-06 21:09:11 did not feel to fuckering with firefox setup again 2025-12-06 21:10:52 somehow I don't need even fstab now 2025-12-06 21:12:13 but this is bad state of the system 2025-12-07 00:44:26 why does the luanti-server package only have support for files, redis, and sqlite3? 2025-12-07 01:08:33 Any ideas how to migrate a Google Cloud VM to alpine? 2025-12-07 01:09:05 I tried attaching a second disk to a running one, then dd'ing the .raw.tar.gz image to the second disk, then set that disk as a boot disk 2025-12-07 01:09:15 but it complains about UEFI not being the same or similar 2025-12-07 01:17:05 minimal: yeah - I want to install alpine to the GCP VM 2025-12-07 01:17:29 the image I tried is from: https://www.alpinelinux.org/cloud/ 2025-12-07 01:20:04 minimal: yeah - that's what I did with dd the disk.raw to /dev/sdb - then disconnect the existing boot disk and use that boot disk instead. 2025-12-07 01:20:41 I used: gcp_alpine-3.22.2-x86_64-uefi-cloudinit-r0.raw.tar.gz 2025-12-07 02:22:53 I've updated my machine from Alpine 3.22 to 3.23, but it seems linux-lts now hard depends on all of linux-firmware. I've got them installed, and technically have the space for it, but would prefer selecting which linux-firmware-* packages I actually need. How would I go about doing that? 2025-12-07 02:23:52 i think you just install the firmware you want, and the rest is uninstalled iirc 2025-12-07 02:24:38 localhost:~# apk del linux-firmware-qcom 2025-12-07 02:24:38 World updated, but the following packages are not removed due to: 2025-12-07 02:24:38 linux-firmware-qcom: linux-lts 2025-12-07 02:24:38 OK: 1522 MiB in 535 packages 2025-12-07 02:27:12 you are trying to uninstall, just put the firmware you want in world 2025-12-07 02:27:48 Yeah, except linux-lts is clearly holding the firmware packages I have installed and don't need. 2025-12-07 02:28:24 if you have no other firmware in world yes, i think they work like alternatives 2025-12-07 02:28:47 it depends on all firmware, or the ones you specify 2025-12-07 02:29:23 Ahh I see, if there's absolutely no firmware specified in world it just installs all of them 2025-12-07 02:29:31 Kinda weird behavior tbh 2025-12-07 02:29:54 i think so yeah 2025-12-07 07:00:12 <|cos|> Good morning, 2025-12-07 07:01:32 <|cos|> ACTION just attempted upgrading to v3.23.0, but now my alpine laptop won't boot. First error is "mkdir: can't create directory '/sbin': No such file or directory". 2025-12-07 07:02:09 <|cos|> Second line says nlplug-findfs can't be found by /init. 2025-12-07 07:04:22 <|cos|> To make things interesting, the keyboard stops working as soon the kernel has booted. 2025-12-07 07:06:30 <|cos|> The grub entry still says "Alpine Linux v3.22, with Linux lts", so I guess the upgrade did not succeed with all steps. 2025-12-07 07:09:00 <|cos|> What's the reasonable way to get a booting system again? Downloading the the v3.23 installer and booting it off a usb stick is the first step, I presume? Would 2025-12-07 07:10:51 <|cos|> Would the be any logs from the `apk upgrade --available` run? To be honest I didn't really read the console upgrade while it was running, but after it completed I got the impression the invokation had been successful. 2025-12-07 07:12:21 there should be a log 2025-12-07 07:12:36 this helped me before: https://wiki.alpinelinux.org/wiki/Chroot#Prepare_chroot 2025-12-07 07:17:54 <|cos|> longnoserob[m]: thanks! am downloading the extended image now. 2025-12-07 07:32:24 So, I managed to load an alpine image into GCP 2025-12-07 07:32:46 but, when I go to log in, it doesn't look like SSH keys are populated to the VM 2025-12-07 07:33:37 I've tried the username alpine, and my SSH key username, and my gmail account name 2025-12-07 08:24:08 CRCinAU: What image did you use? 2025-12-07 08:26:40 I think I just figured it out 2025-12-07 08:26:47 the cloud-init one doesn't seem to work 2025-12-07 08:27:06 the tiny cloud version does 2025-12-07 09:06:05 I have been unable to get sound working in Alpine (inside VIrtualBox). Tried both Virtual and Standard versions. Is this a known issue? 2025-12-07 09:06:32 Just trying to make sure I'm not fighting a battle already lost with this sound issue. Thanks 2025-12-07 09:06:49 what is the host os? 2025-12-07 09:07:20 Windows 10 2025-12-07 09:10:08 Biswa96[m]: not having any issues with other versions of Linux on the same system 2025-12-07 09:16:22 Maybe your user isn't in the audio group 2025-12-07 11:02:39 So again: doas apk del --purge ssl_client 2025-12-07 11:02:39 World updated, but the following packages are not removed due to: 2025-12-07 11:02:39 ssl_client: 2025-12-07 11:02:54 why it does not show dependency? 2025-12-07 11:03:58 same like python3 on desktop 2025-12-07 11:04:27 why that junk does not show how it came into the system 2025-12-07 11:05:15 is it some new "feature" of apk tools 3? 2025-12-07 11:14:04 hope: if you want help from people, it helps to not call things junk 2025-12-07 11:14:27 one can wipe out alpine-base like a dirt from shoe but not some python or ssl thing 2025-12-07 11:14:49 ikke: nice lie instead of harsh truth? ;/ 2025-12-07 11:15:09 No, just don't include any value judgement 2025-12-07 11:15:22 robotic, may try 2025-12-07 11:17:36 but ye, no value is better. As it has no value for me ;/ 2025-12-07 11:18:06 it will probably lower my desire to tell you to touch grass if you don't constantly try to insult people and projects here 2025-12-07 11:20:03 insulting some numbers and words glued together called software is not insulting human who did that mess 2025-12-07 11:20:34 I bet you're a blast at parties. 2025-12-07 11:21:27 I would be, but nobody invited me like last 20 or so years. So I installed alpine instead 2025-12-07 11:21:50 If you want free support for from people, it helps to not insult the work they are doing 2025-12-07 11:26:33 ikke: any ideas why in GCP, the cloud-init version didn't seem to create SSH keys from GCP, but the tiny one did? 2025-12-07 11:28:30 CRCinAU: could you ask in #alpine-cloud? 2025-12-07 11:29:47 I could :) 2025-12-07 11:30:03 I don't exactly have any info - as trying to get any debug out of it was just about impossible :( 2025-12-07 12:16:15 ikke: I would like to say something to you, but it's kind of offtopic, but it may enlighten you. What you think, is there some tiny space in this server to share some "offtopic" lines? ;/ 2025-12-07 12:20:25 the 3.23 release notes say busybox was updated to 1.37.0, but it was already that in the latest 3.22 2025-12-07 13:02:28 hope, might be a bug, but try with apk -v 2025-12-07 13:07:16 ye, -v shows that 2025-12-07 13:09:09 or probably not 2025-12-07 13:09:13 no idea what it outputed first time ;/ 2025-12-07 13:13:49 ye info -a shows just del shows nothing and with -v nonsesne, but that's okay, thanks 2025-12-07 13:21:49 I got used to unfinished, buggy or broken software ;/ 2025-12-07 13:28:49 anyway I will close this support ticket, otherwise I am always tempted to write something in this case and that hurts people from Alpine Linux Ltd ;/ 2025-12-07 13:31:43 Might rather be the way of saying it, not saying it itself 2025-12-07 13:31:52 Ah well, they're gone 2025-12-07 13:34:44 can't say i miss them 2025-12-07 13:42:31 That's nice. 2025-12-07 13:59:45 they got themselves banned from two channels on libera a few days ago 2025-12-07 14:04:55 that does not surprise me at all 2025-12-07 14:11:56 i can't reproduce that complaint anyway 2025-12-07 14:12:32 shows deps here, and who knows why they're trying to remove ssl_client, which they didn't explain 2025-12-07 14:13:04 seems like trolling from more than one angle 2025-12-07 14:13:25 but doesn't bother me personally, except that it ruins the mission here. 2025-12-07 14:13:49 i would have kicked them for less at least once. 2025-12-07 14:36:45 how would i know what package(s) added a specific user or group? "vmail", for example 2025-12-07 14:37:31 grep -r vmail /path_to/your/aports ? 2025-12-07 14:38:10 i don't have an aports 2025-12-07 14:38:53 main/postfix/postfix.pre-install:adduser -S -H -h /var/mail/domains -s /sbin/nologin -G postdrop -g vmail vmail 2>/dev/null 2025-12-07 14:39:18 do you have postfix installed? 2025-12-07 14:39:34 yes. is it searchable on the web? git or something maybe? 2025-12-07 14:40:10 https://gitlab.alpinelinux.org/alpine/aports 2025-12-07 14:40:26 yeah, it's otherwise not available in package metadata 2025-12-07 14:41:02 ah got it, thanks: https://gitlab.alpinelinux.org/search?search=vmail&nav_source=navbar&project_id=1&group_id=2&search_code=true&repository_ref=master 2025-12-07 15:04:35 Hi! 2025-12-07 15:04:35 Can someone help me with the init script of initramfs? 2025-12-07 15:04:35 This would require to mount the device containing the squashfs first, and then, mount the squashfs. Does the default init script have this feature? 2025-12-07 15:04:35 I'm tinkering a bit with Alpine, and I want to make an impermanent system. 2025-12-07 15:04:35 To do this, I need overlayfs, and, fortunately, the default initramfs init script allows it, with the `overlaytmpfs` option. This allows to mount a device as lowerdir, and have tmpfs as upperdir. The thing is that I actually want to use a squashfs file as lowerdir, not a device. The squashfs file is in another filesystem (not initramfs). 2025-12-07 15:05:48 s/This would require to mount the device containing the squashfs first, and then, mount the squashfs. Does the default init script have this feature?/This would require to mount the fs containing the squashfs first, and then, mount the squashfs file (which is located in the fs). Does the default init script have this feature?/ 2025-12-07 15:06:25 s/This would require to mount the device containing the squashfs first, and then, mount the squashfs. Does the default init script have this feature?/This would require to mount the fs containing the squashfs first, and then, mount the squashfs file (which is located in the fs mounted previously). Does the default init script have this feature?/ 2025-12-07 15:09:00 * Hi! 2025-12-07 15:09:00 This would require to mount the fs containing the squashfs first, and then, mount the squashfs file (which is located in the fs mounted previously). Does the default init script have this feature? 2025-12-07 15:09:00 Can someone help me with a problem about the init script of initramfs? 2025-12-07 15:09:00 I'm tinkering a bit with Alpine, and I want to make an impermanent system. 2025-12-07 15:09:00 To do this, I need overlayfs, and, fortunately, the default initramfs init script allows it, with the `overlaytmpfs` option. This allows to mount a device as lowerdir, and have tmpfs as upperdir. The thing is that I actually want to use a squashfs file as lowerdir, not a device. The squashfs file is in another filesystem (not initramfs). 2025-12-07 15:09:49 * Hi! 2025-12-07 15:09:49 I'm tinkering a bit with Alpine, and I want to make an impermanent system. 2025-12-07 15:09:49 Can someone help me with a problem about the init script of initramfs? 2025-12-07 15:09:49 This would require to mount the fs containing the squashfs first, and then, mount the squashfs file (which is located in the fs mounted previously). Does the default init script have this feature? 2025-12-07 15:09:49 To do this, I need overlayfs, and, fortunately, the default initramfs init script allows it, with the `overlaytmpfs` option. This allows to mount a device as lowerdir, and have tmpfs as upperdir. The thing is that I actually want to use a squashfs file as lowerdir, not a device, and the squashfs file is in another filesystem (not initramfs). 2025-12-07 15:11:02 freerig[m]: no need to repeat your message. be patient, if someone can respond to that it may take some time (especially on a weekend) 2025-12-07 15:20:06 invoked: Yeah, I know, but I didn't repeat my message, I just edited it, to make it clearer. I'm sorry if somehow it got duplicated on your side, maybe it's a Matrix bug, idk. Anyways it wasn't my intention to spam :3 2025-12-07 15:22:50 Hello everyone! Can you see my message? I'm trying to ask a question on irc to no avail, im not registered in the logs 2025-12-07 15:23:14 freerig[m]: it's not a bug, because matrix is just a different system, but it is definitely annoying to the irc side. 2025-12-07 15:24:03 i didn't see where your edits were, so that's my bad. 2025-12-07 15:24:21 yeah try not to edit your messages here 2025-12-07 15:27:45 I can see myself from IRC, so I guess you can too, anyways: 2025-12-07 15:27:46 APK is very slow for me on alpine 3.23, installing Plasma is taking me two and a half hours and still not finished. qt6-qtwebcore took 12 mins to install, for example. Anyone else with that problem? What logs should I send? Thanks 2025-12-07 15:34:10 maria[m]: could be a slow mirror. you could try changing it (see https://mirrors.alpinelinux.org/ ) or run setup-apkrepos 2025-12-07 15:34:51 invoked: To install, not to download. The mirror is the fastest one, wget-ting the package by URL takes less than one second 2025-12-07 15:35:01 ibus-emoji is also taking around 7 minutes to install. I think it's a problem with the new apk-tools 2025-12-07 15:35:05 i've also seen a case or two where the drive is ssd and gets very slow because it hasn't gc'd (or won't) 2025-12-07 15:35:22 some qlc drives are known to slow to a crawl 2025-12-07 15:37:01 some of those drive controllers won't respond to trim right away either. you have to let them sit idle for gc to kick in. not saying that is your problem, but putting it out there 2025-12-07 15:37:16 I haven't seen such an issue at all on any other distro, including alpine 3.22 (had a bunch of other problems though), doubt it is a problem with the drive 2025-12-07 15:41:37 Downloading via apk is also way slower than manually, but I'm currently stuck at the installation stage. Seems I'm also banned from the IRC channel for some reason :( 2025-12-07 15:42:40 ACTION uploaded an image: I tried to install alpine offline using the extended x86 image today, but I keep getting this error in the latest step (2875KiB) < https://matrix.org/oftc/media/v1/media/download/ATBaUs_IgjmbSNSVxAuI85wpT7U6g1qJx7lWTjIvkRP6BNvV-LZf5wufhhtQqxKP1dP-TcH_OL6U3KQpUEAUI2NCebQZO-3wAG1hdHJpeC5vcmcvaEduSFlXQklRSlZ3UXhRcFdndExXQ2dO > 2025-12-07 15:44:16 Post install and triggers fail for some packages, like this: 2025-12-07 15:44:16 ERROR: lib/apk/exec/busybox-1.37.0-r29.post-install exited with error 127```. All errors look about the same. 2025-12-07 15:44:16 ```* execve: No such file or directory 2025-12-07 15:46:48 maria[m]: you could try running with 'apk -vv add foo' and see if that informs you of anything 2025-12-07 15:47:19 -vv is global for any command 2025-12-07 15:48:28 if not that, then you'll probably want to add apk-tools-dbg and trace it next. 2025-12-07 16:00:46 freerig: As far as I know current mkinitfs does not have an option for this 2025-12-07 16:07:08 maria: is the apk progress increasing at a constant or inconsistent speed? 2025-12-07 16:08:35 Sertonix: at a constant speed 2025-12-07 16:15:24 "freerig: As far as I know..." <- ok, thanks 2025-12-07 16:15:24 I'm just wondering, is it really safe to do this? Can an Alpine update break my custom init script? 2025-12-07 16:15:24 so I guess I'll have to make my own script 2025-12-07 16:17:09 invoked: tracing doesn't help either, it doesn't show any calls to i/o or network after "installing " 2025-12-07 16:28:16 freerig: It is possible to break but given the current mkinitfs status I do not expect any breaking change 2025-12-07 16:44:00 maria: If apk fetch --recursive qt6-qtwebcore (clutters cwd!) on a tmpfs has the same problem it is likely networking related. 2025-12-07 16:51:49 [@mrstickypiston:matrix.org](https://matrix.to/#/@mrstickypiston:matrix.org) I wasn't able to simply reproduce this in QEMU. Might be best to open an issue to get that fixed. 2025-12-07 16:55:16 "maria: If apk fetch --recursive..." <- I'm very sorry, but I won't be able to check that for a week or two. However, I tried wget-ting qt6-qtwebengine (sorry for the mistake in that message) from the same mirror on the same machine and it took around 10 seconds 2025-12-07 16:55:41 * 10 seconds, so I dont think its network-related 2025-12-07 16:59:41 Ok, no problem. There are sometimes networking issues which vary between applications due to eg. different DNS or ipv4/6 preferences. I have no real guess what it could be, just wanted to exclude some possible causes 2025-12-07 17:06:27 MrStickyPiston[m]: were you running setup-alpine (or -disk) with defaults or doing something more specific? 2025-12-07 17:10:29 is there a tool for taking webshots(screenshot whole web pages) in repo? 2025-12-07 17:12:36 firefox 2025-12-07 17:14:30 ah yes 2025-12-07 17:30:21 invoked: First I was running setup-alpine, just changing some things (most notably user config, offline install and sys partition). 2025-12-07 17:30:21 That did not work (this error) so I tried setup-disk after a reboot. That resulted in the same error. 2025-12-07 17:43:48 MrStickyPiston[m]: ok. sounds like you didn't set ROOTFS or anything, which is what i was curious about. and that you're trying to do a sys install. 2025-12-07 17:48:22 it would be good to have the full output so that we can see what was attempted and what happened prior to those errors you've shared. yet obviously something went catastrophically wrong 2025-12-07 17:49:38 the fs could be busted, or it could be a hardware-related thing... hard to tell from that. 2025-12-07 17:50:38 i'll be afk, but try to share as much detail as you can. 2025-12-07 18:11:11 invoked: Maybe its because I'm running a Pentium M, but I believe it should be supported. 2025-12-07 18:22:51 MrStickyPiston[m]: Pentium M is supported, it's our x86 baseline 2025-12-07 19:44:42 anyone know how to debug suspend issues? since updating to 3.23, whenever my computer is put into suspend, it then seems to recover to a black screen from which not even SysRq+B works 2025-12-07 21:23:13 Hello! I'm trying to adapt https://wiki.alpinelinux.org/wiki/Odroid-C2 by following the guide mostly verbatim but with updated versions (latest master branch from hardkernel u-boot, system part is using alpine-uboot-3.23.0-aarch64.tar.gz, kernel branch linux-rolling-lts.) The only thing I haven't done from the guide is disabling DRM modules. 2025-12-07 21:23:23 Currently I'm stuck trying to debug a kernel panic via UART. It appears initramfs is not being loaded, and I belive it's probably related to this line: `[0.000000] Unknown kernel command line parameters "modules=loop,squashfs,sd-mod,usb-storage modloop=/boot/modloop", will be passed to user space.` 2025-12-07 21:23:43 Later there's another message `rootfs image is not initramfs (read error); looks like an initrd` but I think this may be a downstream effect of failing to load kernel modules. Full log @ https://pastebin.com/raw/HAMhsbdS + thanks in advance for any help. 2025-12-07 21:23:59 gzv: first, I wouldn't use HK u-boot 2025-12-07 21:24:27 Alright, I can start there 2025-12-07 21:24:42 actually, wait a moment 2025-12-07 21:25:09 the guide does tell you to use HK u-boot only for getting amlogic blobs and tells you to use mainline u-boot with the blobs 2025-12-07 21:25:12 so it's alright 2025-12-07 21:25:57 gzv: I would personally just boot via UEFI or extlinux tbh 2025-12-07 21:26:07 instead of hacking a u-boot script 2025-12-07 21:26:25 u-boot supports booting EFI executables 2025-12-07 21:26:41 and with nice perks like DTB selection 2025-12-07 21:28:20 f_: OK, thanks! not very familiar with UEFI on ARM or extlinux but I'll look into both 2025-12-07 21:28:29 extlinux might be easiest 2025-12-07 21:28:45 just make an extlinux config in extlinux/extlinux.conf in your boot partition 2025-12-07 21:29:20 cool, sounds simple enough 2025-12-08 06:17:27 my keyboard backlight is activated with the scroll lock key. I have a simple acpid script to echo 1 > /sys/class/leds/input0::scrolllock/brightness when I press scroll lock. however, since upgrading to 3.23 when logging into a GNOME session or waking the computer from suspend, this file gets reset to 0 and the keyboard backlight disabled. how can I prevent this happening? 2025-12-08 06:18:27 this happens several times each time, it's very annoying 2025-12-08 06:19:28 or how might I go about troubleshooting it? 2025-12-08 06:25:39 `gsettings set org.gnome.desktop.peripherals.keyboard remember-numlock-state false` does not help 2025-12-08 06:32:25 possibly related: now whenever I try to log out of a GNOME session I get "GNOME shell not responding" 2025-12-08 06:35:59 ah wonderful, the keyboard backlight is disabled again when the ssh-agent GTK password popup displays 2025-12-08 09:14:24 rnkn: same re GNOME shell 2025-12-08 09:15:20 do you also have problems with suspend? my machine suspends fine, but upon resuming, just has a black screen and seemingly some sort of kernel panic? even sysrq+b, which is supposed to do the harshest reboot possible, doesn't work 2025-12-08 09:15:44 (though that might be hardware-specific, i have another machine that runs alpine 3.23 and it suspends/resumes fine) 2025-12-08 09:18:08 tried looking into /var/log/messages for answers, but there's nothing useful there, just one message that is like "going into suspend" and the next message is from the next bootup 2025-12-08 09:20:00 gonna try without graphics, see if that changes anything 2025-12-08 09:22:00 okay, interesting, if i don't run a graphical session and just "sudo pm-suspend" from the console, it recovers perfectly well 2025-12-08 09:24:23 but shutting and opening a laptop lid still crashes everything :-( 2025-12-08 09:29:37 interesting, "pm-suspend" works, but "loginctl suspend" crashes 2025-12-08 10:19:38 okay, i was able to resolve the problem by editing /etc/elogind/sleep.conf.d/10-elogind.conf and setting SuspendMode=s2idle 2025-12-08 10:20:03 but still, that means whichever suspend mode elogind uses by default crashes my machine :-( 2025-12-08 10:29:15 hazardaj_nombroj, I would rather look at what pm-suspend does that systemd doesn't 2025-12-08 10:31:33 well, i doubt it's the usermode parts that are causing the system to crash upon recovery so hard, even sysrq doesn't do anything 2025-12-08 10:32:10 so it's probably the suspend mode itself that's causing trouble, but i should check if the other suspend modes also crash with pm-suspend 2025-12-08 10:33:47 You can doubt or you can check 2025-12-08 10:40:53 The x86 variant available on the website, What exactly is it? I have an i586 machine i need something ridiculously small on... (The only boot media it accepts is a 1gb cf card) 2025-12-08 10:43:07 Leahh: only pentium-m unfortunately, 3.18 should™ boot on a i586 though 2025-12-08 10:43:41 might try that... Debian's "i386" variant has a bunch of packages that aren't i386 at all and are infact i686 so debian doesn't really run on that machine 2025-12-08 10:45:57 if you have lots of time you can wait until i have nerd-sniped myself into doing a rebuild for older x86, but that is still in the conceptual stage :D 2025-12-08 10:46:11 could do that too :D 2025-12-08 10:46:23 could be months still ;P 2025-12-08 10:46:23 Currently its running openwrt but it's kinda stupid using router software for server stuff 2025-12-08 10:46:35 so i wanted a "real" linux distro for it 2025-12-08 10:47:06 ok, so you want slightly less minimal, that invalidates my other ideas for what to run on it 2025-12-08 10:47:24 well imean anything that fits on the 1gb cf card is on the table 2025-12-08 10:47:38 its an pcengines alix 2d3 fwiw 2025-12-08 10:47:56 IIRC it's build for i586+CMOV 2025-12-08 10:48:03 built 2025-12-08 10:48:07 cmov? 2025-12-08 10:48:15 Which doesn't agree with all i586 architectures 2025-12-08 10:48:55 Yeah Leahh 2025-12-08 10:49:13 https://gitlab.alpinelinux.org/alpine/aports/-/blob/3.18-stable/main/gcc/APKBUILD#L296 2025-12-08 10:49:33 should that include cmov? 2025-12-08 10:49:51 How do I see if my cpu has that without flashing alpine and hoping it boots? 2025-12-08 10:50:16 do you have gcc on that machine? 2025-12-08 10:50:27 don't think so no, OpenWRT doesn't package that 2025-12-08 10:51:07 Leahh, you could try to PXE-boot it 2025-12-08 10:51:43 The AMD datasheet for LX800 seems to indicate it supports cmov 2025-12-08 10:51:48 so i might give alpine 3.18 a shot 2025-12-08 10:52:08 ok, at least that concern is out of the window :) 2025-12-08 10:52:10 otherwise you could compile and run this funny snippet of c code 2025-12-08 10:52:23 https://tpaste.us/nNpv 2025-12-08 10:52:24 That's a cood device 2025-12-08 10:52:28 i probably could on my modern host system and just transfer it xd 2025-12-08 10:52:29 s/cood/cool/ 2025-12-08 10:52:35 yeah its a really cool little embedded box 2025-12-08 10:52:41 although slow as balls XD 2025-12-08 10:52:44 :) 2025-12-08 10:53:50 keep us updated, i have an unhealthy obsession with low spec computers 2025-12-08 10:54:05 hehehe 2025-12-08 10:55:04 socksinspace: Is the C snippet you sent evaluated at compile time? or does the __builtin_cpu_supports evaluate at runtime? 2025-12-08 10:56:10 if my late night reading of the gcc docs yesterday is correct it is a runtime thing 2025-12-08 10:56:32 i'll try running that on it and see what it shows 2025-12-08 10:57:51 It's runtime, that's basically doing cpuid calls, look at the assembly, if there are cpuid calls you know for sure 2025-12-08 10:58:42 ah, thx, i was looking at the code yesterday to figure our how it works and gave up again :D 2025-12-08 10:59:02 so i probably still need to add some manual chacks for my purposes 2025-12-08 10:59:16 s/chacks/checks 2025-12-08 11:00:44 you can try objdump -S foo |grep -i cpuid should be enough 2025-12-08 11:05:41 juppp, that's a few handfulls of cpuid 2025-12-08 11:06:23 definitely wil need to learn some assembly for a "trust no one" mode in that program 2025-12-08 11:14:59 awesome 2025-12-08 11:18:17 love that for me 2025-12-08 11:37:48 flashed the alpine 3.18.12 iso to the cf card, Serial output only gave "ISOLINUX 6.04 6.04-pre1 EHDD" and nothing else happened. 2025-12-08 11:37:52 :/ 2025-12-08 11:39:53 "alpine-standard-3.18.12-x86.iso" fwiw 2025-12-08 11:50:04 its EOL 2025-12-08 11:51:08 I need something supporting i586 and afaik the newest x86 images are for i686. 2025-12-08 11:52:21 i'm not sure the iso can work on media that is not a cd 2025-12-08 11:52:36 ah darn 2025-12-08 11:52:52 the only media that this thing supports booting from is a cf card, and the only one i have is only 1gb 2025-12-08 12:00:25 oh, so I can't install alpine on my soekris then either 2025-12-08 12:00:33 if you run `isohybrid` on the iso, it *might* make the image bootable from non-cd media 2025-12-08 12:00:53 but might be easier to manually "install" it -- format, mkfs, untar the tarball, chroot, apk install a kernel and stuff 2025-12-08 12:01:43 that, or just abuse setup-alpine or setup-disk a bit, if you are already running alpine 2025-12-08 12:02:20 Does the isa have serial console enabled by default? 2025-12-08 12:03:12 actually, boot the iso on another pc, plug the cf card in, install to the cf card instead of the hard disk, plug cf into alix 2025-12-08 12:04:21 6‑r36 per pkgs.alpinelinux.com. Any ideas? 2025-12-08 12:04:21 I’m having issues with the 3.23 and 3.22 community repos on a fresh setup. When installing apk‑file and other packages I get ERROR: apk-file-0.3.6-r35: package mentioned in index not found (try 'apk update'). I ran apk update (v3.22.2‑292‑ga6fcfd66010 [http://dl-cdn.alpinelinux.org/alpine/v3.22/main]), but the index still points to the old r35; the current version is apk‑file‑0.3. 2025-12-08 12:05:24 which mirrors are defined in /etc/apk/repositories 2025-12-08 12:06:31 dl-cdn.alpinelinux.org and alpine.mirror.wearetriple.com 2025-12-08 12:06:53 The weird thing is that the repos show the correct file. 2025-12-08 12:17:21 socksinspace: What I'll do is use the openwrt to bootstrap alpine to a 60gb ide harddrive 2025-12-08 12:17:28 and just use the openwrt's grub to boot alpine 2025-12-08 12:17:38 since i do have a 60gb ide harddrive in it 2025-12-08 12:17:51 but it won't boot directly from it so I hijack grub from the openwrt install on the cf card 2025-12-08 12:18:34 ok, otherwise i can try cooking up an image for the cf card :D 2025-12-08 12:18:55 oooo that could be useful :D 2025-12-08 12:19:23 if you're going to, Enable via-rhine ethernet drivers if possible 2025-12-08 12:21:15 oh, i was just going to do a quick install in qemu and then chuck that disk somewhere (not sure where yet), nothing too fancy 2025-12-08 12:21:39 that'll probably work fine, i can use my own custom kernel with the networking drivers already built in 2025-12-08 12:22:03 and for >200mb files https://catbox.moe/ works pretty good 2025-12-08 13:02:50 Leahh: https://litter.catbox.moe/j5w62o19oczkci4n.gz hope that works, otherwise i'll look like a fool :D login as root, no password, keymap is set to de-nodeadkeys, dhcp will try to get an adress on eth0 2025-12-08 13:04:15 wizzard: do you still have the issue? 2025-12-08 13:04:21 We had one server that is behind 2025-12-08 13:24:03 Problem still occurs on dl-cdn.alpinelinux.org (2a04:4e42:9::644) but not with alpine.mirror.wearetriple.com 2025-12-08 14:09:42 I get some sudden weird error with Calibre, trying to convert a book 2025-12-08 14:09:59 This plugin does not support createPlatformVulkanInstance 2025-12-08 14:09:59 QVulkanInstance: Failed to initialize Vulkan 2025-12-08 14:09:59 Unable to detect GPU vendor. 2025-12-08 14:10:17 Why would Calibre try to use vulkan, and why would it be a hard fail 2025-12-08 14:10:45 I have vulkan-loader installed, though pretty much sure that's not supporting this computer GPU, but it's a hard dependencie from most other software installed anyway 2025-12-08 14:17:45 what's the image that has linux-firmwares in the .iso? I'm installing my new laptop and it doesn't find the iwl firmware, do I need extended or standard should have it? 2025-12-08 14:20:40 hmph, it's there I just think it's not new enough 2025-12-08 14:20:58 ACTION wobble 2025-12-08 14:21:01 :( 2025-12-08 14:21:43 it wants some iwlwifi-bz-b0-fm-c0-{c99,101,100}.ucode 2025-12-08 14:21:55 jokes on me for wanting a lunar lake 2025-12-08 14:22:59 Yeah, there's no water on teh moon 2025-12-08 14:23:05 Could have asked before 2025-12-08 14:24:17 oh cool we have it, maybe I just need the extended image 2025-12-08 16:51:32 socksinspace: Took a nap and ate dinner XD 2025-12-08 16:51:36 ready to try the image now 2025-12-08 16:52:00 the link seems dead 😔 2025-12-08 16:58:49 seems alpine 3.18 is i686 2025-12-08 16:59:00 tried to run it in qemu to install to the cf card 2025-12-08 16:59:03 localhost:~# uname -a 2025-12-08 16:59:03 Linux localhost 6.1.128-0-lts #1-Alpine SMP PREEMPT_DYNAMIC Fri, 07 Feb 2025 19:30:59 +0000 i686 Linux 2025-12-08 16:59:10 and it reports i686 2025-12-08 17:09:39 And ran the C snippet on the machine, it supports MMX and CMOV. (No SSE and SSE2) 2025-12-08 17:25:51 Leahh: the image should've been 'dd'-able, i'll look into it when i'm back home (in about 2 hours) 2025-12-08 17:26:57 it runs on qemu's 486, but allegedly that will execute non-486 instructions, so idk 2025-12-08 17:56:14 yeah the files are limitedly available unless you set it to the max of 3-days 2025-12-08 17:56:35 the alpine-standard-3.18.12-x86.iso seems to be i686 at least 2025-12-08 18:20:12 Leahh did you aks about nvidi driver before? 2025-12-08 18:20:25 i did before 2025-12-08 18:20:33 i was debating putting alpine on my laptop 2025-12-08 18:20:40 and i kinda need it for games 2025-12-08 18:22:17 i wrote a reply did you see it maybe? 2025-12-08 18:22:18 haiku is using something 2025-12-08 18:22:32 i don't think i've seen the reply no 2025-12-08 18:22:46 unfortunately irc doesn't do to well on that regard xd 2025-12-08 18:22:50 not on par with proprietary but it is something and open 2025-12-08 18:23:24 so alpine could use that (if true) 2025-12-08 18:23:53 dont remember if i made this comment before, but if you use doas/sudo without password, isnt that _almost_ like being root and if someone trick you to run something all they have to do is put doas in front of it.. 2025-12-08 18:24:05 im embracing su! 2025-12-08 18:24:18 imean the nouveau driver "works" but is slow 2025-12-08 18:25:20 yeah with nouveau i could not even use full resolution 2025-12-08 18:25:20 i was told that haiku driver is better 2025-12-08 18:25:24 it is not nouveau 2025-12-08 18:25:39 Hi, do you have recommendations for rootless podman container handling on alpine as host OS (with quadlets from the podman-systemd world in mind)? 2025-12-08 18:25:39 Right now I've got bash scripts that start the container. They differ from my quadlet config on all other hosts which irks me. 2025-12-08 18:26:14 yeah nouveau can be really spotty, My 3070 laptop gpu didn't have support AT ALL up until pretty recently 2025-12-08 18:26:26 and even then it's pretty spotty until i actually get into a session 2025-12-08 18:41:55 Back to the alix 2d3 :D 2025-12-08 18:42:06 Tried using minirootfs images and booting those 2025-12-08 18:42:07 [ 6.918066] Run /sbin/openrc as init process 2025-12-08 18:42:07 [ 7.281753] Kernel panic - not syncing: Attempted to kill init! exitcode=0x00000000 2025-12-08 18:42:17 seems I don't have openrc installed right i rhink 2025-12-08 18:59:06 Leahh: You might want to drop into a busybox shell instead of openrc and inspect the system… 2025-12-08 18:59:06 Leahh: Wrong arch or something like that, i.e. really an openrc that can't boot? 2025-12-08 19:00:55 Linux (none) 6.16.0-rc3-g1c04acdb0591 #6 SMP PREEMPT_DYNAMIC Wed Aug 6 15:56:08 CEST 2025 i586 Linux 2025-12-08 19:01:01 Got into it with /bin/ash as init 2025-12-08 19:02:24 just running openrc from ash gives no output and returncode is 0 2025-12-08 19:03:51 did you do "exec /sbin/init" ? 2025-12-08 19:04:02 no i just ran "openrc" xd 2025-12-08 19:05:26 https://bpa.st/KO2Q 2025-12-08 19:05:32 Got this from that, and now it seems stuck 2025-12-08 19:10:48 You can try deactivating services (removing symlinks in /etc/runlevel/*/) before you "exec /sbin/init") 2025-12-08 19:46:59 ok, uploading again, this time with a more sensible expiration set 2025-12-08 19:48:04 Leahh: https://litter.catbox.moe/ezjzk7vuy4k2tkbj.gz 2025-12-08 20:03:50 actually got to download it now xd 2025-12-08 20:03:58 gonna try diffling with the services first 2025-12-08 20:04:01 fiddling* 2025-12-08 20:05:20 haha there's nothing enabled anywhere lol 2025-12-08 20:05:25 that might be why nothing seemed to happen 2025-12-08 20:23:06 socksinspace: I got your image booting! Flashed it to the CF card, and made both the kernel and extlinux output to serial 2025-12-08 20:23:10 and now it's in alpine 2025-12-08 20:24:25 yay \o/ 2025-12-08 20:24:38 it finds the ethernet ports 2025-12-08 20:24:41 but can't seem to get dhcp 2025-12-08 20:25:25 might try my own kernel and seeing if it helps 2025-12-08 20:25:44 if it has a different name than eth0 you will need to edit /etc/network/interfaces 2025-12-08 20:26:26 haha it shows "fbcon: Taking over console" but this machine has NO display output 2025-12-08 20:26:32 i think i'm using eth1 2025-12-08 20:26:35 the port in the middle of the three 2025-12-08 20:27:41 no clue what linux names those on your machine, just look in the output of "ip a" 2025-12-08 20:27:52 yeah it's eth0, eth1, and eth2 2025-12-08 20:27:55 just made all of those do dhcp 2025-12-08 20:28:03 ah ok 2025-12-08 20:29:14 so that did work? 2025-12-08 20:29:19 yeah it's online :D 2025-12-08 20:29:27 hell yeah 2025-12-08 20:29:42 i'll have to do a bit of configuration and then i'll make an image of the contents :D 2025-12-08 20:29:50 Set hostname, unpriviliged account, etc 2025-12-08 20:30:04 yeah, i tried to do as little as possible 2025-12-08 20:30:29 keymap and network were un-avoidable 2025-12-08 20:30:40 appreciate it a lot :D 2025-12-08 20:46:07 ... 2025-12-08 20:46:11 FingerlessGloves is here. hi! 2025-12-08 20:46:15 also hi f_ 2025-12-08 20:46:40 good UK afternoon leah 2025-12-08 20:46:44 yes 2025-12-08 20:47:28 f_: FingerlessGloves helped me a while back with some networking stuff 2025-12-08 20:47:40 Elo leah 2025-12-08 20:47:43 and also generally humoured me at the intellectual level 2025-12-08 20:49:55 leah: you followed me! 2025-12-08 20:50:01 maybe i'm the real leah 2025-12-08 20:50:07 we are reaching concentrations of leahs that shouldn't be possible 2025-12-08 20:50:24 i was told two other leahs here here. literally that is why i came, so that there would be three. 2025-12-08 20:50:36 and also because alpine is great. thank you. 2025-12-08 20:50:38 in the libreboot channel i keep getting pinged by people thinking i'm leah 2025-12-08 20:51:24 socksinspace: I actually think that, just like there is not enough linus, there is not enough leah 2025-12-08 20:54:29 well that's just gravy 2025-12-08 21:09:51 i know one other leah. perhaps i can coax her into coming here too. 2025-12-09 04:18:50 where should I next report my keyboard backlight problem (i.e. erantly resetting the value of /sys/class/leds/input0::scrolllock/brightness)? 2025-12-09 04:19:28 *errantly 2025-12-09 04:20:07 also should the GNOME shell not responding issue be reported upstream? 2025-12-09 04:40:21 jfc the gnome-shell project has 2.3K issues 2025-12-09 05:02:06 okay, um, gnome shell consistently hanging seems, to me at least, like an "oh no we shouldn't have pushed out gnome 49" bug, but hey what do I know 2025-12-09 05:28:03 I hear https://kde.org is great 😄 2025-12-09 05:43:00 software is great when it does not crash abruptly :) 2025-12-09 05:46:45 downgraded to 3.22 2025-12-09 05:47:01 3.23 is evidentally not ready for its "latest-stable" moniker 2025-12-09 11:23:51 also hi there, all the leahs 2025-12-09 16:33:21 i wonder how the guys behind v3.23 upgraded gnome when i remember hearing they removed non-systemd fallback code in gnome and added harder dependencies on systemd carry-on 2025-12-09 17:03:36 gnome-session is still on 48 2025-12-09 17:04:45 I suppose the rest only is on 49 2025-12-09 17:04:54 https://blogs.gnome.org/adrianvovk/2025/06/10/gnome-systemd-dependencies/ 2025-12-09 17:05:08 > might be able to get GNOME 49 running with gnome-session 48, though this is a configuration that won’t be tested or supported upstream so your mileage will vary (much like running GNOME on other init systems). Still, patching that scenario to work may buy you more time to upgrade to gnome-session 49. 2025-12-09 20:49:19 hi there, running alpine linux edge right now and can't seem to get openrc user services working properly. seems to think i didn't boot my pc with openrc... (full message at ) 2025-12-09 20:56:15 * hi there, running alpine linux edge right now and can't seem to get openrc user services working properly. seems to think i didn't boot my pc with openrc... (full message at ) 2025-12-09 23:09:53 I think manually starting the openrc user service is not intended (yet?). Something like openrc -U boot might work better. Normally (when a user.* service or openrc-user-pam is setup) this script is invoked which does the setup stuff: https://github.com/OpenRC/openrc/blob/master/sh/openrc-user.sh.in#L30 2025-12-09 23:31:41 I'm looking to make a configurable boot media on an SD card but I can't seem to be able to select it as a config target.. 2025-12-09 23:33:43 I can only use /media/usb as a config destination. 2025-12-10 05:46:23 gnome-clocks will not launch. wonder if it is broken or i did something off. 2025-12-10 12:49:39 Hello, do anyone know how to put udhcpc broadcasting discover in background? 2025-12-10 12:51:29 Hello tabris, you should be able to find that answer in the busybox man-page :) 2025-12-10 12:55:45 worst case: run & 2025-12-10 13:03:58 I mean, the one that ran on networking. How do I configure the udhcpc there to do '-b' option, because udhcpc is not metioned in init.d/networking. 2025-12-10 13:30:07 Sonuds a bit like an XY question 2025-12-10 13:31:19 Maybe you're actually looking for this? https://wiki.alpinelinux.org/wiki/OpenRC#Preventing_slow_services_from_delaying_system_startup 2025-12-10 14:11:57 hi, I just installed alpine with xorg and dwm. usee 2025-12-10 14:13:46 user is in input group, and xf86-input-libinput is installed. udev is also installed and running, but the keyboard and mouse wont reapond after startx and entering dwm. any ideas? 2025-12-10 14:18:24 *respond 2025-12-10 14:24:55 anything in xlog? 2025-12-10 14:52:36 nicci, you may also need video group, if seat manager is not used..https://wiki.alpinelinux.org/wiki/Setting_up_a_new_user#Groups_for_desktop_usage 2025-12-10 15:04:12 hey, keyboard trackpad or usb mouse is not reaponding in dwm after new install. any ideas why? 2025-12-10 15:14:19 i don't use dwm, but installed and updated the documents. i don't have touchpad etc.. see if this helps.. https://wiki.alpinelinux.org/wiki/Xorg#Input_packages 2025-12-10 15:18:33 i had a similar issue while setting up mine but forget exactly what i did, i have a feeling it was setting up elogind that made it work 2025-12-10 15:19:14 although i had video group too, if its the same issue i had it wasnt that 2025-12-10 15:19:30 (alpine added my groups for me i think in the installer) 2025-12-10 15:20:59 ichigo1, are you using display manager/login manager to start dwm or using startx... 2025-12-10 15:21:28 using ly display manager now, but was using startx for a while before that 2025-12-10 15:24:34 using startx may not need elogind.. but i'm using sway+seatd i.e no elogind. unsure about dwm..it's quite sometime since i updated that wiki page..like longnoserob suggested, checking logs is best bet.. 2025-12-10 20:50:52 the edge packages may have different dependencies hence different hashes 2025-12-10 20:50:53 i am not a alpine dev so i dont knwo for certain 2025-12-10 21:47:14 nico: non-deterministic packages and maybe apk does not know it should not reinstall a package also 2025-12-10 21:49:58 right now it's impossible to have reproducible/deterministic packages because abuild inserts own stuff into it (god knows for what reason) 2025-12-10 21:50:10 # Generated by abuild 3.16.0-r0 2025-12-10 21:50:10 # using fakeroot version 1.37.1.2 2025-12-10 21:51:00 and every package has "builddate = 1764857827" 2025-12-11 07:06:49 are gnome 49 patches likely to get backported? 2025-12-11 07:21:42 what would be the non-systemdism to `systemctl --user stop org.gnome.SettingsDaemon.Power.target`? 2025-12-11 11:03:45 hi 2025-12-11 11:03:57 im on alpine edge and got an error on systemupdate 2025-12-11 11:04:04 error: mariadb-11.8.5-r0: trying to overwrite usr/lib/mariadb/plugin/parsec.so owned by mariadb-connector-c-3.4.6-r0 2025-12-11 13:44:52 does alpine-virt not have fuse built into the kernel? 2025-12-11 13:47:38 how small is a fully-functional commandline system in GB ? (includes the basic POSIX tools) 2025-12-11 13:59:50 nevermind, solved 2025-12-11 14:52:32 > We had one server that is behind 2025-12-11 14:52:32 yesterday the problem was gone. 2025-12-11 15:51:43 Bobb3rt: possibly a few MB (approximately the contents of bootstrap tarball) if you don't include a kernel, maybe half a GB if you want a whole bootable system with a kernel and some firmware blobs -- I don't remember more precisely than that, but either way it's well under a GB 2025-12-11 15:54:16 grawity: ty 2025-12-11 18:21:02 Anyone using color management with XFCE+Alpine? 2025-12-11 18:21:20 I packaged xiccd in testing to try and get this working alongside colord, but uh... big fail 2025-12-11 22:57:14 Good day, I have upgraded an x86_64 vm to 3.23, and after reboot, I am getting libacl.so.1 errors (cat, ls, etc) when trying to run anything. DNS, ssh server are all "dead", ls cat df are as well. Any tips to try and fix without complete reinstallation? 2025-12-12 02:25:28 hi all :D how's everyone? 2025-12-12 02:27:58 if you guys downloading from torrent and can't keep up with ratio...try downloading from xdcc bots, you can find a network that serves xdcc packs at https://xdcc-search.com - its all free and servers are running on high bandwidth. Thank me later :D 2025-12-12 03:22:46 lmao why is linux-lts on alpine linux edge 6.18 2025-12-12 03:22:46 and linux-stable is 6.17.11 2025-12-12 03:24:17 because 6.18 is the latest lts kernel 2025-12-12 03:25:34 latest lts kernel is: longterm: 6.12.61 2025-12-12 03:25:42 has been 6.12 for a long time 2025-12-12 03:25:48 they arent just gonna jump 6 versions lol 2025-12-12 03:26:01 the 2026 lts is 6.18, they've already announced it 2025-12-12 03:26:19 it's the last kernel release of the year, typically 2025-12-12 03:26:46 https://www.phoronix.com/news/Linux-6.18-LTS-Kernel 2025-12-12 03:26:47 ohh, i didn't know, that would make sense 2025-12-12 03:27:30 i guess its not incremental like mainline, thanks for sharing lol. my mind just didnt understand how lts was a higher version than stable 2025-12-12 05:49:00 Hi, I have a weird boot from usb problem on 3.23.0 2025-12-12 05:49:44 USB is created from image using Rufus on Windows. Either file copy or dd method makes no difference. 2025-12-12 05:50:29 Booting on an old Lenovo all in one. If I disable hard disks in the bios it boots fine. 2025-12-12 05:50:47 If I put an ancient windows 10 HD in the machine it boots fine. 2025-12-12 05:51:29 If I have an empty SSD it seems to get drive allocation all screwed up and I get: 2025-12-12 05:52:47 find: /media/sda/apks/aarch64/*****: I/O error 2025-12-12 05:53:27 many variations on ***** 2025-12-12 05:54:44 why the hell looking for aarch64? All I can guess is that the SSD was previously used as HD on an install of Alpine on a Rpi. I've wiped the partition table in fdisk though. 2025-12-12 05:55:05 Any help appreciated 2025-12-12 05:58:05 Have verified that I used an x86_64 image. I doubt I'd get this far if I hadn't. 2025-12-12 06:19:13 Ok, some success. I got more suspicious of the hardrive so opened fdisk back up on another live disk (artix) created a new partition table and then made a vfat filesystem on the disk. Then it successfully boots. 2025-12-12 06:19:45 And I can probably move on. But why is the initial image trying to do anything from the HD in the machine? 2025-12-12 09:15:04 also can boot usb device by pressing f12 on lenovo 2025-12-12 09:16:35 basically lets you choose what device to boot, 2025-12-12 09:16:35 as for the issue, not sure, sounds like the hdd could have just had leftover stuff on it but not sure 2025-12-12 10:04:36 hi all :D how's everyone? 2025-12-12 10:07:06 if you guys downloading from torrent and can't keep up with ratio...try downloading from xdcc bots, you can find a network that serves xdcc packs at https://xdcc-search.com - its all free and servers are running on high bandwidth. Thank me later :D 2025-12-12 10:08:55 but yeah openrc -U boot was not the solution, as all that does is specify what boot level. which boot doesnt exist for the local user (never created it so makes sense) 2025-12-12 10:08:55 still saying i did not boot openrc and that i need to `touch /run/user/1000/openrc/softlevel` 2025-12-12 10:09:21 s/boot level/runlevel/ 2025-12-12 10:38:31 ichigo[m]1, if you are referring to Wiki instructions on openrc user services, i'll appreciate if you can edit wiki directly.. or add to the talk page on what's not correct..and changes required.. 2025-12-12 10:57:01 thanks, the issue is i do not know the solution or required changes, so i can't really update yet. I'm certian that running touch softlevel is not ideal, 2025-12-12 10:57:01 what im referring to is i did follow the instructions, but it thinks i didnt boot with openrc when i try. 2025-12-12 10:58:41 something like: "You are attempting to run an openrc service on a system which openrc did not boot. You may be inside a chroot or you may have used another initialization system to boot this system. In this situation, you will get unpredictable results! If you really want to do this, issue the following command: touch /run/user/1000/openrc/softlevel" 2025-12-12 11:01:04 i can post the workaround on the wiki if you like, but honestly im wondering if anyone may know if i missed something that may be needed, or something else. 2025-12-12 11:08:29 which desktop environment or window manager you're using... 2025-12-12 11:10:14 on alpine linux with i3, running openrc -U in .profile to activate it 2025-12-12 11:10:37 openrc -U boot 2025-12-12 11:10:37 touch /run/user/1000/openrc/softlevel 2025-12-12 11:10:37 openrc -U 2025-12-12 11:10:37 sleep 1 2025-12-12 11:11:00 actually i have to remove boot, forgot to 2025-12-12 11:11:02 see whether this version of instructions works... https://wiki.alpinelinux.org/w/index.php?title=OpenRC&oldid=30771#For_Wayland 2025-12-12 11:11:34 hmm i3wm uses xorg, i mean i can try the wayland one anyways 2025-12-12 11:12:19 sorry https://wiki.alpinelinux.org/w/index.php?title=OpenRC&oldid=30771#For_Xorg 2025-12-12 11:13:10 i.e instructions without using sysinit etc.. 2025-12-12 11:13:14 oh thank you 2025-12-12 11:13:28 please feedback..if it works.. 2025-12-12 11:15:14 will do! also rc-status -Ur returns sysinit for me, not gui and default 2025-12-12 11:15:39 but its still in the user, not sure if i missed a step, or if this is intended 2025-12-12 11:16:01 i do have elogind setup 2025-12-12 11:16:03 because, you followed earlier instructions to add it to .profile 2025-12-12 11:16:55 s/followed/might have followed 2025-12-12 11:19:22 hey i figured it out! wasnt anything wrong with the setup, just a missing package apparently. 2025-12-12 11:19:43 after installing openrc-user-pam everything worked upon reboot. only have openrc -U in my .profile no more workaround needed. 2025-12-12 11:19:45 thanks for the help 2025-12-12 11:20:01 should i add to the wiki that that might be required? and why is it required? feel that would be important to say too 2025-12-12 11:20:21 (i dont know why its required tho, thats more so an actual question) 2025-12-12 11:21:43 there's a discussion going on in the talk page about this..kindly add you'r feedback there..this page has already been disturbed multiple times.. so please feedback in talk page.. 2025-12-12 11:22:16 https://wiki.alpinelinux.org/wiki/Talk:OpenRC#User_Services_-%3E_Runlevels 2025-12-12 11:23:42 for me that instruction with sysinit runlevel did not work... 2025-12-12 11:24:07 i'm still waiting for reply from WhyNotHugo 2025-12-12 11:27:36 yeah its all a bit confusing, i don't know why it shows sysinit, but in my case, that is all it has ever shown for me 2025-12-12 11:30:14 for now i need some rest, been a long day, but will gather up information to add a concise post about the issue i had and the optimal solution i found 2025-12-12 14:22:39 Hi 2025-12-12 14:22:55 I have a couple zpools on a system 2025-12-12 14:23:08 The main zpool mounts ok (root), but the other one never does 2025-12-12 14:23:15 I can import it manually without issue 2025-12-12 14:23:21 Any idea how I could fix this? 2025-12-12 14:29:45 Hummm, /etc/conf.d/zfs is quite different from another Alpine system 2025-12-12 14:30:08 How come, aren't we supposed to get .apk-new files to track changes during upgrades? 2025-12-12 14:31:10 I think that's only if the 'old' file was modified 2025-12-12 14:31:42 Well, at least one of those two different files were, obviously ^^ 2025-12-12 14:32:16 And I don't remember modifying this, I didn't know it existed before now 2025-12-12 14:33:00 Is there a place on the system where those files are kept? 2025-12-12 14:35:07 /etc/conf.d/zfs is owned by zfs-openrc-2.4.0_rc5-r0 2025-12-12 14:35:36 $ find aports/ -name zfs-openrc 2025-12-12 14:35:38 nothing 2025-12-12 14:36:23 grep on the whole repo doesn't return anything either 2025-12-12 14:39:49 https://pkgs.alpinelinux.org/package/v3.23/main/x86_64/zfs-openrc 2025-12-12 14:42:03 How do I DL that 2025-12-12 14:42:36 apk-fetch :> 2025-12-12 14:48:50 It looks like there's some weird race condition 2025-12-12 14:49:07 I can see during boot it says, after * Checking local filesystems ... 2025-12-12 14:49:13 cannot open 'bpool': no such pool 2025-12-12 14:49:49 And then it fails later to mount the zfs from the bpool obviously on * Mounting local filesystems ... 2025-12-12 14:50:09 This bpool is on the same device as the main pool that we boot from 2025-12-12 14:53:48 Any idea how to get the boot log? 2025-12-12 15:05:35 had no idea about apk fetch thanks for sharing 2025-12-12 15:05:55 :) 2025-12-12 15:28:33 hi all :D how's everyone? 2025-12-12 15:29:55 i am full of pilk 2025-12-12 15:30:42 it's a spam account 2025-12-12 15:31:03 if you guys downloading from torrent and can't keep up with ratio...try downloading from xdcc bots, you can find a network that serves xdcc packs at https://xdcc-search.com - its all free and servers are running on high bandwidth. Thank me later :D 2025-12-12 15:42:40 oh I sure want to DCC so much I have DCC disabled and not working at all 2025-12-12 15:43:16 There's no need to comment on spam :P 2025-12-12 15:43:41 I just find it strange I suppose 2025-12-12 15:43:53 thanks for dealing with it either way ^^ 2025-12-12 15:44:05 DCC only works inside the same network, so people using the same bouncer can dcc. I don't think you can dcc through nat without some sort of tunnel 2025-12-12 15:44:19 network not to be confused with irc network; I mean subnet 2025-12-12 15:45:47 It's often best to disable DCC entirely IMO .. these days it's hardly ever used for any kind of legitimate file transfer 2025-12-12 15:46:00 anyway, - sorry for starting it :p 2025-12-12 15:48:14 ssm, you can just setup forwarding for the ports and you go 2025-12-12 16:00:48 makes sense 2025-12-12 16:10:21 ladies 2025-12-12 16:10:52 I went with release instead of edge, so there are no constant updates 2025-12-12 16:11:08 yet somehow, there is always some shit that wants to update, how it comes? 2025-12-12 16:11:35 I have almost nothing installed, yet still some junk wants "newer" version of itself 2025-12-12 16:12:09 there are still security updates 2025-12-12 16:12:12 is it all so broken, that it requires extra lines of crap on top of already crap? 2025-12-12 16:12:21 oh security 2025-12-12 16:12:35 so when I upgrade I will be more safe? 2025-12-12 16:12:59 which would mean I was pretty unsecure before 2025-12-12 16:13:38 it's pretty annoying to be honest 2025-12-12 16:14:17 not as much as matrix bridge, but still 2025-12-12 16:14:24 all software that isnt capability safe is insecure :D 2025-12-12 16:14:39 so updates are pointeless then 2025-12-12 16:14:47 it's still insecure 2025-12-12 16:14:54 with new insecurity added on top of that 2025-12-12 16:15:15 you use that word a lot with no examples 2025-12-12 16:15:15 this software reminds me of mind 2025-12-12 16:15:31 what 2025-12-12 16:15:46 if your device is behind nat, your chances of being attacked are mostly low, unless it comes from within your lan 2025-12-12 16:16:00 if you're port forwarding someone could attack whatever is listening on that port 2025-12-12 16:16:18 so recently apk-tools updated 2025-12-12 16:16:35 what it has to do with networking 2025-12-12 16:17:04 it updated itself so it can update something else ;/ 2025-12-12 16:18:18 it does not have trillion features I assume 2025-12-12 16:18:34 so if it's released it should be pretty stable 2025-12-12 16:18:45 in modern language, secure too 2025-12-12 16:19:21 still better than update of mesa crap obviously 2025-12-12 16:19:36 this at least does not need half the internet to fill up half the disk space available 2025-12-12 16:19:55 but if it's so small, it could be maybe "secure" 2025-12-12 16:20:02 without constant updates 2025-12-12 16:20:19 to annoy me in my peace 2025-12-12 16:21:00 image you buy plate so you can eat your junk food from it 2025-12-12 16:21:12 and somebody every fucking day knock at your door to update that plate 2025-12-12 16:22:22 so no more updates please until next release 2025-12-12 16:22:24 thank you 2025-12-12 16:23:07 sometimes there are showstopper bugs that have to have backported fixes 2025-12-12 16:23:17 so comparing to a plate is sorta bad 2025-12-12 16:23:25 ye, but definition of that sometimes is every fucking day 2025-12-12 16:23:55 if I have standard machine full of crap it would be probably several times in a day 2025-12-12 16:25:46 there was apk tools that previous version 2025-12-12 16:25:53 I assume it had not so many bugs 2025-12-12 16:25:56 after years 2025-12-12 16:26:02 now there is apk tools 3 so new bugs 2025-12-12 16:26:10 so next year I bet somebody will want it in rust 2025-12-12 16:26:19 so next 10 years there can be bugfixes 2025-12-12 16:27:42 maybe issue is that people don't pay for their monsterous "cloud" build setup 2025-12-12 16:28:22 so shit can constantly update, it's on someone else's computer and someone else's pay for it 2025-12-12 16:29:27 and I rather not open pointlessly installed python on my machine 2025-12-12 16:29:35 bc. some atk god knows what 2025-12-12 16:29:55 I bet it does not need any python as runtime 2025-12-12 16:30:12 but sponsor pay for build machine 2025-12-12 16:30:21 so who cares 2025-12-12 16:30:46 let's push it to thousands of people's computers 2025-12-12 16:31:16 so we can do some magic security updates tomorrow 2025-12-12 16:31:18 hi all :D how's everyone? 2025-12-12 16:31:23 bad! 2025-12-12 16:31:45 alpine is unsecure I was told 2025-12-12 16:32:02 nor small 2025-12-12 16:32:07 hardly simple more 2025-12-12 16:32:11 website should be adjusted 2025-12-12 16:32:53 with something like "not yet debian, working on it" 2025-12-12 16:33:48 if you guys downloading from torrent and can't keep up with ratio...try downloading from xdcc bots, you can find a network that serves xdcc packs at https://xdcc-search.com - its all free and servers are running on high bandwidth. Thank me later :D 2025-12-12 16:41:06 ye next update xdcc bot in apk tools 2025-12-12 17:01:16 alpine is one of the smallest linuxen out there, its just that python is a dependency for a lot of shit 2025-12-12 17:04:39 like yt-dlp and mpv 2025-12-12 17:06:03 even KISS has python and its notoriously small 2025-12-12 17:06:09 that junk shows me it is there bc. at-spi2-core 2025-12-12 17:06:14 https://www.kisslinux.org/ 2025-12-12 17:06:28 which I think it does not requires python for runtime 2025-12-12 17:06:30 kiss is shit 2025-12-12 17:06:48 probably dead 2025-12-12 17:06:56 like most so called distro with different background 2025-12-12 17:09:04 but what I did not know until now 2025-12-12 17:09:09 apk add !at-spi2-core 2025-12-12 17:09:26 lot of junk gone 2025-12-12 17:10:04 theres kiss community which still updates the core 2025-12-12 17:10:31 fuck them 2025-12-12 17:10:53 that's not really appropriate 2025-12-12 17:11:09 offended? 2025-12-12 17:11:20 this is sparta! 2025-12-12 17:11:25 I meant alpine! 2025-12-12 17:11:36 not some kissing 2025-12-12 17:11:36 what does the ! do? 2025-12-12 17:11:52 hope: and I am god (so to speak) 2025-12-12 17:12:10 !weber.oftc.net dwfreed is using God mode: MODE #alpine-linux +o dwfreed 2025-12-12 17:13:05 frag: https://gitlab.alpinelinux.org/alpine/aports/-/merge_requests/80773 2025-12-12 17:13:10 removes crap 2025-12-12 17:14:00 mentioned in at-spi2-core apkbuild at least 2025-12-12 17:14:25 I don't mind kissing 2025-12-12 17:14:38 but others mind, you are ugly 2025-12-12 17:14:52 like python 2025-12-12 17:15:07 and anything gnome touches 2025-12-12 17:15:39 are you going to behave or do you really want to get kicked 2025-12-12 17:15:50 you are also god? 2025-12-12 17:16:18 I'm Santa's little helper 2025-12-12 17:16:39 ye, so peasant yelling from the backyard to his master ;/ 2025-12-12 17:17:10 so please behave and don't mention kissing 2025-12-12 17:18:26 i like keeping things simple, stupid 2025-12-12 17:19:17 that's why you have !at-spi2-core 2025-12-12 17:19:24 not available in kissing linux 2025-12-12 17:19:32 only in alpine! 2025-12-12 17:20:53 hope: please be respectful 2025-12-12 17:22:25 I fixed my issue with pointless pyton 2025-12-12 17:22:29 did you fix yours? 2025-12-12 17:23:40 oh recently I visited icb 2025-12-12 17:23:43 ye ancient shit 2025-12-12 17:23:51 a bit simpler than irc 2025-12-12 17:23:58 but alpine has nothing to offer there 2025-12-12 17:24:04 nor server and only client available for ircii 2025-12-12 17:24:19 which is also the worse irc client in the whole irc universe 2025-12-12 17:24:53 custom icb server would be not hard but 2025-12-12 17:24:58 if there are no really clients 2025-12-12 17:25:04 then it's a bit pointless 2025-12-12 17:25:18 alpine's missing opportunity to dominate 2025-12-12 17:25:44 not that other distros has rich offer 2025-12-12 17:25:47 but at least something 2025-12-12 17:25:51 mostly broken probably 2025-12-12 17:26:16 I highly recommend to check icb 2025-12-12 17:26:49 no bridges to matrix! 2025-12-12 17:27:23 penguinz_rule[m]: still curious how matrix and then also bridge to irc is sitting in the table of "simple, stupid" 2025-12-12 17:29:19 enough, this isn't your personal rant channel 2025-12-12 17:29:50 not yours 2025-12-12 17:30:16 it's definitely dwfreed's 2025-12-12 17:31:20 cool, but that's off topic here, we were talking about alpine and icb 2025-12-12 17:31:34 only you 2025-12-12 17:31:40 ^ 2025-12-12 17:31:44 so if you wish to continue that, you can do so in #alpine-offtopic 2025-12-12 17:31:49 it's slow channel, so I write and wait 2025-12-12 17:31:54 otherwise please cease that 2025-12-12 17:31:54 somebody may join 2025-12-12 17:31:59 do less writing and more waiting 2025-12-12 17:32:17 when you're 95% of the lines in the last hour and a half, you're talking way too much 2025-12-12 17:32:28 but there is nothing else 2025-12-12 17:32:30 what are you even waiting for? you already solved your issue 2025-12-12 17:32:33 it's not that I interupt something 2025-12-12 17:32:50 on other hand, you are interupting 2025-12-12 17:32:53 that doesn't mean it's not annoying 2025-12-12 17:33:01 ye, you are 2025-12-12 17:33:04 your useless monologue is filling up my chatlogs 2025-12-12 17:33:15 I point out again that I run the *network* 2025-12-12 17:33:31 nobody cares 2025-12-12 17:33:44 please! don't waste this screen for off topic 2025-12-12 17:34:31 would have dropped the hammer earlier but lost my op since bridge is funky again :> 2025-12-12 17:35:17 you don't have op here? 2025-12-12 17:35:40 (as in, you're not on the access list AFAICT) 2025-12-12 17:36:12 yes I'm not on the list, I was op'd to clean up matrix mostly 2025-12-12 17:36:23 and the fun part is I'm still moderator on matrix 2025-12-12 17:36:33 because the bridge is funky 2025-12-12 17:37:02 heh 2025-12-12 17:37:12 but every time bridge dies I get disconnected 2025-12-12 17:38:13 seems though matrix spam has been quiet lately 2025-12-12 17:40:33 it comes and goes 2025-12-12 17:41:06 my last matrix kill was 10 days ago 2025-12-12 18:01:55 i see that hope has still not followed my advice to go touch grass 2025-12-12 18:02:05 if you guys downloading from torrent and can't keep up with ratio...try downloading from xdcc bots, you can find a network that serves xdcc packs at https://xdcc-search.com - its all free and servers are running on high bandwidth. Thank me later :D 2025-12-12 18:40:25 iiuc, apk commit hooks only run when you're doing something that updates the world file (i.e. add/del) are there any hooks that run at other times (I'm specifically looking for post upgrade). I know there's triggers= (used by kernel-hooks, etc), but is there a way to install those without making a package? 2025-12-12 19:55:44 can i get an invite to alpine-offtopic, my client for some reason doesnt seem to want to join it (despite somehow joining this channel) 2025-12-12 20:02:53 ok invite sent 2025-12-12 20:03:15 thanks 2025-12-12 20:39:58 iggy: Commit hooks also run when the world file doesn't change 2025-12-12 21:43:26 if you guys downloading from torrent and can't keep up with ratio...try downloading from xdcc bots, you can find a network that serves xdcc packs at https://xdcc-search.com - its all free and servers are running on high bandwidth. Thank me later :D 2025-12-12 21:53:20 I don't get those spams 2025-12-12 21:53:52 At least spam about bitcoins or whatever, there's no need to spam about DCC 2025-12-12 22:47:17 if you guys downloading from torrent and can't keep up with ratio...try downloading from xdcc bots, you can find a network that serves xdcc packs at https://xdcc-search.com - its all free and servers are running on high bandwidth. Thank me later :D 2025-12-12 22:57:00 is this xdcc shit spam or no 2025-12-12 22:57:10 seen it like twice 2025-12-12 23:18:27 Yeah, I might need an invite to alpine-offtopic matrix-side, too. My current invite is very broken 2025-12-12 23:57:02 if you guys downloading from torrent and can't keep up with ratio...try downloading from xdcc bots, you can find a network that serves xdcc packs at https://xdcc-search.com - its all free and servers are running on high bandwidth. Thank me later :D 2025-12-13 00:17:34 Sertonix[m]: do you know where any docs are about when they run? All I've found was the wiki and it's pretty vague 2025-12-13 00:35:27 I have alpine plasma on virtual-box and the VBoxLinuxAdditions do not match present VBox, which is 7.2.4. Is ther some experimental pkg for a newer pkg? 2025-12-13 00:35:56 for the newer virtualbox* 2025-12-13 00:36:32 the one apk offers is 7.0.2 2025-12-13 02:32:26 iggy: Generally commit hooks are run when using apk upgrade/add/fix/del without --simulate. There are a few special cases but I don't think these can be documented well 2025-12-13 05:10:41 i've got alpine on an old laptop of mine. i'm connected to my phone's hotspot, yet `ping 8.8.8.8` fails with the usual "network unreachable". any tips for debugging? 2025-12-13 05:25:48 alternatively, is there a way to get iwd on this thing without wifi? i think my issues may be live boot + wpa_supplicant related 2025-12-13 05:29:34 oh weird. i ran `dhcpcd` after seeing that routes and things may be my issue, waited for it to connect, and ran `ping`. thought it timed out. 2025-12-13 05:29:43 no i just uh have 98% packet loss 2025-12-13 05:30:07 ...not great. guess this might be a hardware issue. 2025-12-13 06:11:28 what drivers are on the extended alpine linux iso, and how do i get them onto my chroot? are these distributed as packages? 2025-12-13 06:16:06 alternatively, how do i tell what driver is in use wrt. a corresponding piece of hardware? 2025-12-13 06:18:30 jj, maybe try "lspci -k 2025-12-13 06:18:31 " assuming your network cards are PCI and not USB. 2025-12-13 06:20:08 unfortunately, my (working) network card is usb. lspci -k has nice output telling me the kernel driver, though. do you know of one for usb? 2025-12-13 06:23:52 (the issue i'm facing here is my usb wifi card works in my usb boot alpine linux extended, doesn't work in the actual install) 2025-12-13 06:24:06 (there's some driver alpine extended has that i need to track down) 2025-12-13 06:37:00 blugh. found lshw. grepping its output says my Realtek dongle is using rtl8xxxu. now how do i find what package this is packaged in? 2025-12-13 06:40:13 i... searching the alpine source suggests this is packaged in linux-lts? what? 2025-12-13 06:41:06 hrm, why would my wifi dongle work on the alpine-extended live boot but not my actual install, then? 2025-12-13 06:41:24 i'm very confused. any advice here would be appreciated. 2025-12-13 06:48:40 sound like there a kernel module on the live boot CD, but not installed on your system 2025-12-13 06:51:13 yeah. i'm currently trying to track down what package it's tied to :c 2025-12-13 07:00:55 If its saying its in the linux-lts package... then its probs already in kernel. Do you have linux-firmware installed ? 2025-12-13 07:01:09 sry if you already answered this. just logged in 2025-12-13 07:02:10 looks like it, linux-firmware-{i915,intel,mediatek,other,rtl_bt} 2025-12-13 07:14:25 oh, i totally did not, actually. thanks!! that was the problem. i thought i had it installed from presence of some firmwares but didn't know that was a meta-package and i should have WAY more 2025-12-13 07:37:19 gg 2025-12-13 09:29:04 question -- after updating/uninstalling/reinstalling some packages, i've mucked things up and my virtual terminal now boots with 1) small text and 2) ip link only outputting lo. interestingly, running setup-wayland-base (after stalling out the update checks wrt. no connectivity) fixes *both* counts. anyone know what's up with this / how i fix it for the virtual console? 2025-12-13 09:30:24 i think the latter has something to do with kernel modules. 2025-12-13 09:43:46 ah, `rc-update add hwdrivers sysinit` fixed things. i appear to have lost `mdev` and `modloop` from the sysinit level too, off to learn if i should add them back 2025-12-13 09:47:36 okay, seems not, modloop seems liveboot related and mdev in compl. distribution with udev 2025-12-13 12:49:13 Hello, I am trying to setup audio surround downmix, I already did it with another OS on the same machine, after some debugging I found out that by default the pulseaudio server is used instead of pipewire but I cannot remove pulseaudio since it is a dependency of gnome packages. Is it wanted to have pulseaudio instead of pipewire ? Is there a way to change the default server to pipewire otherwise ? 2025-12-13 15:11:26 I'm using kodi-gdm, but it's not taking any keyboard input. The kodi user is a member of the input group, and dmesg lists the keyboard device. Any ideas how to debug this further? 2025-12-13 15:19:29 Upon stopping kodi I see that key presses have been sent to the login(3) process on that tty 2025-12-13 16:01:09 hii 2025-12-13 16:32:02 Figured it out: instaling libudev-zero fixes the issue (setting up eudev might have worked too). Added a wiki page for Kodi: https://wiki.alpinelinux.org/wiki/Kodi 2025-12-13 17:33:17 if you guys downloading from torrent and can't keep up with ratio...try downloading from xdcc bots, you can find a network that serves xdcc packs at https://xdcc-search.com - its all free and servers are running on high bandwidth. Thank me later :D 2025-12-14 02:58:57 snchmt[m]: install pipewire-pulse firstly and then try to remove pulseaudio 2025-12-14 18:57:16 ~ 2025-12-14 19:05:49 installed 3.23 on rpi, getting error when trying to install yt-dlp: ERROR: unable to select packages: so:libsimdutf.so.27 (no such package): required by: deno-2.3.1-r4[so:libsimdutf.so.27] 2025-12-14 19:06:26 .. from tar-ball, then upgraded to edge 2025-12-14 19:16:49 hmm maybe yt-dlp-core is enough 2025-12-14 20:23:34 :autolog on 2025-12-14 20:24:52 ACTION starts logging 2025-12-14 20:27:29 Good day, I have upgraded an x86_64 vm to 3.23, and after reboot, I am getting libacl.so.1 errors (cat, ls, etc) when trying to run anything. DNS, ssh server are all "dead", ls cat df are as well. Any tips to try and fix without complete reinstallation? 2025-12-14 20:30:21 maybe `apk fix` with the static tool? 2025-12-14 20:40:50 Elsifer, you can “recover” base tools using busybox, like $ busybox ls 2025-12-14 22:12:17 There might be a rebuild of some package missing 2025-12-14 22:30:42 mustn't have been important 2025-12-14 23:24:55 lopid: apk fix with static, like get the apk file to rhe host and manually install? 2025-12-14 23:26:09 quinq: thanks for the busybox ls tip, that helps me navigate better with tools that are linked against my missing library 2025-12-14 23:33:27 :) 2025-12-14 23:33:54 Elsifer, with the static build of apk, that doesn't use shared libraries so it'll work stand-alone on a broken system 2025-12-14 23:35:28 https://dl-cdn.alpinelinux.org/alpine/latest-stable/main/x86_64/apk-tools-static-3.0.2-r0.apk 2025-12-14 23:46:57 quinq: ok, thanks. Seems my apk works, but networking is down, eth0 is up, resolv.conf is legit, no firewall rules, cant seem to get traffic out 2025-12-14 23:54:48 Do you have an IP 2025-12-15 00:02:28 ifconfig eth0 says yes, with lost of rx dropped packets though. ping from outside fails 2025-12-15 00:06:44 That's hardware issue 2025-12-15 00:14:10 be a vm, thats b0rked after an upgrade. My data is on a persistent /var, so ill just do a reinstall and see 2025-12-15 00:14:32 thanks though 2025-12-15 02:15:52 Do you want comfort this festive/ winter 🥶 season ? Then look no further cuz we got just the exact Joy box 📦 for you!... (full message at ) 2025-12-15 02:32:05 spam^ 2025-12-15 03:39:19 firefox is not recognizing my microphone... using pipewire 2025-12-15 03:46:31 ah, apparently pipewire-pulse fixes it 2025-12-15 19:35:24 hi. i cant add alpine provided ublock-origin to librewolf. it says unsigned addon. 2025-12-15 19:36:53 about:config switchs already disabled and not works too. 2025-12-15 19:40:40 is there anyone has same problem? 2025-12-15 19:43:30 How about bundle ublock with librewolf and disable singature check completely? 2025-12-15 20:09:27 I could take a look at it but is it a big problem with not using the alpine packaged ublock-origin? 2025-12-15 20:12:47 doesnt librewolf come with ublock installed? im using zerohosts.sh, dunno if ublock is better 2025-12-15 20:15:20 didnt know about the ublock alpine package btw.. 2025-12-15 20:15:32 convenient 2025-12-15 20:16:27 oh well, gotta sleep 2025-12-15 20:39:22 hello 2025-12-15 20:39:59 who was? 2025-12-15 21:52:41 @Sertonix how will alpine librewolf handle XDG fix from upstream FireFox? 2025-12-15 21:53:11 Also, thanks for maintaining LibreWolf 2025-12-15 21:55:38 It will likely be just like upstream firefox with a different name 2025-12-15 22:03:34 Sweet, okay 2025-12-16 02:30:14 syncplay just randomly stopped working for me today, any ideas?... (full message at ) 2025-12-16 02:49:17 /12 2025-12-16 03:06:29 did you ran apk upgrade earlier? 2025-12-16 03:14:49 yeah its up to date, on the edge repos, maybe just have to wait for an update to something 2025-12-16 17:43:37 Does alpine build its kernels with CONFIG_IXGBE_HWMON? 2025-12-16 17:43:55 I can't find it mentioned in the config in aports and there's no /proc/config.gz 2025-12-16 17:45:51 check /boot 2025-12-16 17:45:58 Noisytoot: should have /boot/config-${kversion}-${rel}-${_flavor} 2025-12-16 17:46:56 found it (the answer is yes, CONFIG_IXGBE_HWMON=y) 2025-12-16 17:47:58 is should have said /boot/config-$(uname -r) 2025-12-16 17:48:15 s/is/i/ 2025-12-16 17:52:18 I have an Intel X520-DA2 installed but there's nothing relevant in /sys/class/hwmon for some reason 2025-12-16 20:38:16 I had a really strange bug 2025-12-16 20:38:30 I tried Setup-Desktop sway 2025-12-16 20:38:37 Then keyboard stopped working 2025-12-16 20:44:53 i forgot to ever ask this, but does alpine linux have a ban on AI code contributions? perhaps it should, e.g. qemu and i think netbsd have done so, for example. at least until the major copyright cases seem settled. it would also strengthen the position of respecting open-source licenses to be on the side of caution, and i imagine for most patches commonly submitted to alpine they require a level of technical knowledge that an AI isn't very 2025-12-16 20:44:53 good at anyway 2025-12-16 20:49:02 I feel like the problematic usage should already be covered by existing rules so not sure if it is worth spending time on it 2025-12-16 22:00:22 does anyone actually use apk-tools with zstd? 2025-12-17 05:02:54 "I feel like the problematic..." <- i think unless it's an outright ban people will simply contribute AI code. even with DCO and all of that. it seems to have been somehow an accepted thing to do, even though plenty of people seem to think DCO interferes with AI use 2025-12-17 05:03:26 * i think unless it's an outright ban to do so, people will simply contribute AI code. even with e.g. a developer certificate of origin and all of that. it seems to have been somehow an accepted thing to do, even though plenty of people seem to think DCO interferes with AI use 2025-12-17 05:06:03 for what it's worth, https://wiki.alpinelinux.org/wiki/Alpine_Linux:Contribute this page doesn't really mention any guidelines or rules around anything and perhaps it should? 🤔 unless i'm being silly and it's there, which happens 2025-12-17 05:08:04 i'm guessing https://wiki.alpinelinux.org/wiki/Creating_patches should probably also refer to potential rules around what a patch should do or not do from a copyright angle, it doesn't really seem to mention it and i guess a lot of people will end up on this page when they're about to make a concrete actual code contribution 2025-12-17 17:36:48 Can I use firewalld on alpine 2025-12-17 18:30:28 ~ $ nmcli connection up "HHU-VPN" Error: Connection activation failed: Could not find source connection. 2025-12-17 18:30:33 could you help pls 2025-12-17 18:30:48 @Ckvjj https://pkgs.alpinelinux.org/packages?name=*firewalld*&branch=edge&arch=x86_64 2025-12-17 18:50:59 have you defined one called "HHU-VPN"? 2025-12-17 18:51:57 "i'm guessing https://wiki...." <- are there like, any guidelines? :o 2025-12-17 19:32:28 I have the most UGLY alpine setup ever 2025-12-17 19:32:30 but it works 2025-12-17 19:32:31 :D 2025-12-17 19:32:50 like I dont even know what I did 2025-12-17 19:33:47 a question when I copy a file with doas the file cant be written then with normal I need doas nano to write in it why is that 2025-12-17 19:33:58 I hope doas nano is no security risk ;) 2025-12-17 19:34:59 Check the owner after using doas 2025-12-17 19:35:15 how to do that 2025-12-17 19:35:26 well it is not a security risk I think 2025-12-17 19:35:31 It likely no longer is your user/group, hence needing doas or elevatioj to interact with it for writing 2025-12-17 19:35:52 https://pastebin.com/DWk0aLKz 2025-12-17 19:35:54 ls -lla in that dir is one way 2025-12-17 19:36:08 Ah yes I see 2025-12-17 19:36:31 could you have a look at my ufw setup and could you tell me what to do better eventually 2025-12-17 19:36:31 .d 2025-12-17 19:36:39 I be right back I make food 2025-12-17 19:36:46 Expected behavior when using sudo/doas 2025-12-17 19:37:14 I'm the wrong person to ask about software firewall settings 2025-12-17 19:56:20 can I somehow see how much software is installed 2025-12-17 19:56:24 with apk 2025-12-17 19:57:13 apk info | wc -l ? 2025-12-17 20:03:39 thanks 2025-12-17 20:03:47 cool I have a running alpine for the first time 2025-12-17 20:05:04 Enjoy! 2025-12-17 20:10:12 thanks 2025-12-17 20:19:38 Who stole the precious path? 2025-12-17 20:19:38 /usr/local/sbin:/usr/local/bin:/usr/bin 2025-12-17 20:19:38 ``` 2025-12-17 20:19:38 / # /bin/printenv PATH 2025-12-17 20:19:38 ``` 2025-12-17 20:20:11 marisa 2025-12-18 08:51:54 hello I change apk repo from 3.22 to 3.23 and did apk upgrade --available. after that I am getting this 2025-12-18 08:51:58 Continuing the upgrade transaction with new apk-tools: Error loading shared library libapk.so.2.14.9: No such file or directory (needed by /sbin/apk) Error relocating /sbin/apk: fetchConnectionCacheInit: symbol not found Error relocating /sbin/apk: apk_db_foreach_sorted_name: symbol not found Error relocating /sbin/apk: apk_name_sorted_providers: symbol not found Error relocating /sbin/apk: apk_ostream_counter: symbol no 2025-12-18 08:53:15 any help woulb be greatly appreciated. 2025-12-18 09:24:21 anyonr. 2025-12-18 09:52:42 you could try the static apk 2025-12-18 10:26:09 World updated, but the following packages are not removed due to: swaybar: sway 2025-12-18 10:27:37 perhaps you should remove sway 2025-12-18 10:28:13 nono I want keep sway but remove swaybar 2025-12-18 10:28:13 :d 2025-12-18 10:28:15 :p 2025-12-18 10:28:21 I use waybar 2025-12-18 10:28:22 :P 2025-12-18 10:28:52 300 mb ram usage is really small 2025-12-18 10:28:53 :D 2025-12-18 10:30:02 hmm, it is not 'required by' anything, according to the repo 2025-12-18 10:30:23 3.23? 2025-12-18 10:30:24 lopid 2025-12-18 10:30:26 what means that 2025-12-18 10:30:30 sway is a desktop 2025-12-18 10:30:31 lol 2025-12-18 10:30:33 https://pkgs.alpinelinux.org/package/v3.23/community/x86_64/swaybar 2025-12-18 10:31:12 maybe the 'install if' holds it 2025-12-18 10:33:24 lopid Idk who to fix 2025-12-18 10:33:54 i'd suggest waiting around for somebody who knows better 2025-12-18 10:34:58 Hello Again! 2025-12-18 10:35:12 longnoserob[m]: What you mean by static apk? 2025-12-18 10:35:23 and what happened? 2025-12-18 10:35:33 is this error recoverable? 2025-12-18 10:36:06 https://pkgs.alpinelinux.org/package/v3.23/main/x86_64/apk-tools-static 2025-12-18 12:12:20 can I exec poweroff in a running system? 2025-12-18 12:12:30 like when browser and sway opened or bad idea 2025-12-18 12:13:57 i would close all user programs before running it 2025-12-18 12:14:20 services and that get managed by openrc so theres less of an issue with shutdown on them 2025-12-18 12:14:24 penguinz_rule[m]: can I also use halt instead 2025-12-18 12:14:24 ? 2025-12-18 12:15:08 halt doesnt shut off the computer last time i tried it, so i have no idea why its there tbh 2025-12-18 12:16:34 penguinz_rule[m]: it requires doas halt 2025-12-18 12:16:41 doas halt then it should work 2025-12-18 12:19:25 yea 2025-12-18 12:19:29 doas halt doesnt work 2025-12-18 12:19:30 lol 2025-12-18 12:20:24 well it started shutdown but then became a blackbox it shutted down with many errors 2025-12-18 12:20:33 penguinz_rule[m]: 2025-12-18 12:20:45 but it didnt really shutdown 2025-12-18 12:20:50 I had to press the powerbutton 2025-12-18 12:21:23 yeah it halts the machine but doesnt actually do poweroff, i think its used for doing some sort of reboot or something, no idea 2025-12-18 12:36:00 Hello, is there a reason for no mesa version higher than 25.2.7 being in the tree? 2025-12-18 12:40:39 is there a way to shutdown alpine linux without gettin errors when using doas poweroff in graphical env which doeas close all apps before shutting down? 2025-12-18 12:41:52 probably elogind or smthing? 2025-12-18 12:41:54 loginctl poweroff? You don't even need root privilege to run that one 2025-12-18 12:42:10 let me check 2025-12-18 12:42:25 Well, looks like it worked 2025-12-18 12:43:17 hello again. I assumed it worked? 2025-12-18 12:43:27 assume* 2025-12-18 12:43:33 I see I dont need root but again with graphical env opened and apps opened I get errors shutting down in tty 2025-12-18 12:43:33 :d 2025-12-18 12:43:45 so sway and firefox opened 2025-12-18 12:43:52 then shutdown in graphical env 2025-12-18 12:43:57 What errors? 2025-12-18 12:44:02 good question 2025-12-18 12:44:08 they appear in tty 2025-12-18 12:44:25 You could try taking a photo? 2025-12-18 12:44:28 but I am not quick enough to take a picture of the error 2025-12-18 12:44:45 try browsing /var/log/messages or smth 2025-12-18 12:47:38 the error does not look like the error in tty of shutdown asch 2025-12-18 12:47:52 https://pastebin.com/ 2025-12-18 12:48:06 https://pastebin.com/PkwWySiS 2025-12-18 12:48:55 is there maybe a way to track log of exact output in tty 2025-12-18 12:54:04 Maybe /var/log/rc.log ? 2025-12-18 12:55:04 I dont have rc.log 2025-12-18 12:56:50 Dec 18 13:42:15 localhost daemon.err loadkmap[4543]: ERROR: loadkmap failed to stop 2025-12-18 12:57:47 Well, there are other files in /var/log. There are probably not many of them, so maybe one of them has what you are looking for 2025-12-18 12:58:34 ./var/log $ ls acpid.log apk.log dmesg messages messages.0 wtmp 2025-12-18 12:59:57 doas cat dmesg lol 2025-12-18 13:00:39 well I think 2025-12-18 13:00:51 it is problemm that I try to shutdown from a graphical environment 2025-12-18 13:01:00 with apps opened 2025-12-18 13:01:11 if I close them previously the error do not appear 2025-12-18 13:01:20 also when I close sway before poweroff 2025-12-18 13:01:37 which brings me to the idea 2025-12-18 13:02:27 no mx ifrss id dzoopif 2025-12-18 13:02:28 <.f 2025-12-18 13:02:35 my idea is stupid 2025-12-18 13:02:35 :d 2025-12-18 13:02:39 any ideas? 2025-12-18 13:06:54 not really. 2025-12-18 13:06:57 sry =P 2025-12-18 13:08:45 Actually, take a look at this: https://wiki.gentoo.org/wiki/OpenRC#Logging 2025-12-18 13:09:20 You could enable logging on /etc/rc.conf, shutting down as you were doing and then try to fetch /var/log/rc.log 2025-12-18 13:09:37 Or any other log file destination you set in /etc/rc.conf 2025-12-18 13:10:21 BTW, is Alpine Arch-based or Gentoo-based? Or maybe a bit of both? 2025-12-18 13:20:38 i think alpine is from scratch 2025-12-18 13:28:20 is it dangerous to use cat or nano with doas 2025-12-18 13:28:35 bindsym F12 mode $quit set $quit "poweroff press Enter or press Escape to abort" mode $quit { bindsym Return exec swaymsg -t get_tree | jq '.. | select(.pid?) | .pid' | xargs -r kill; exec sway> bindsym Escape mode default } 2025-12-18 13:33:10 asdasdsadsad: smell test against your level of experience says yes 2025-12-18 13:33:17 8-ball says ask again later 2025-12-18 13:33:40 asdasdsadsad: depends on the editor 2025-12-18 13:33:50 running emacs as root means that all the elisp has root privileges now 2025-12-18 13:34:07 so you use tramp in order to copy the file to a temp directory and operate on that nonroot file descriptor 2025-12-18 13:34:38 but cat and nano is well known or not? 2025-12-18 13:34:56 cat is fine 2025-12-18 13:35:04 everyone uses nano to sudo access stuff 2025-12-18 13:35:05 nano is probably ok, 2025-12-18 13:35:17 i would still use sudoedit instead 2025-12-18 13:35:20 and poweroff? 2025-12-18 13:35:20 :d 2025-12-18 13:35:24 i think there is doasedit in alpine testing 2025-12-18 14:17:56 does someone use firewalld 2025-12-18 14:17:57 ? 2025-12-18 14:18:07 penguin I found a way 2025-12-18 14:18:11 to fix it 2025-12-18 14:18:16 if you still need 2025-12-18 14:20:31 https://pastebin.com/Zcw44jh9 2025-12-18 14:20:40 fix is on pastebin 2025-12-18 14:58:44 df -h /tmp 2025-12-18 14:58:48 sorry 2025-12-18 15:04:45 ZibaSM: Somethink happend which made apk not executable anymore. You can try downloading a static apk binary and use something like ./apk.static upgrade -a and ./apk.static fix to hopefully fix the system. https://gitlab.alpinelinux.org/alpine/apk-tools/-/releases/v3.0.3 2025-12-18 15:09:36 somethink 2025-12-18 16:07:58 "are there like, any guidelines?..." <- is it okay if i poke again? :o 2025-12-18 16:15:41 do matrix replies even show up onirc 2025-12-18 16:16:17 they look like this: "are there like, any guidelines?..." <- is it okay if i poke again? :o 2025-12-18 16:17:31 in general, assume matrix features don't work or pass through the bridge correctly 2025-12-18 16:21:40 https://boinc.berkeley.edu/ install this 2025-12-18 16:30:30 i was pointing out neither https://wiki.alpinelinux.org/wiki/Alpine_Linux:Contribute nor https://wiki.alpinelinux.org/wiki/Creating_patches seem to mention any guidelines regarding contributing code and copyright, like attribution, making sure the licensing is compatible, etc. and they dont seem to mention either which is why i brought this up at first, what this means for AI contributions. there are various examples out there how extreme 2025-12-18 16:30:30 AI plagiarism can be, even if you dont try to press it to auto-complete a known pre-existing text: https://lcamtuf.substack.com/p/large-language-models-and-plagiarism and some projects like qemu and netbsd have banned AI code contributions entirely. i simply wanted to suggest alpine may ponder the same. 2025-12-18 16:30:55 anyway i'm not at all a lawyer so i don't know anything reliable on this topic and if in doubt assume i'm wrong. this isn't legal advice etc 2025-12-18 16:32:05 feel free to e.g. screenshot and forward my message for discussion in some other place, if this isn't a good avenue to bring this up 2025-12-18 17:18:14 if I wanted R programming language a bit more secure is alpine containeringg apps and guis or should I rather use homebrew or something 2025-12-18 17:18:24 pkg_src 2025-12-18 17:18:28 from bsd 2025-12-18 17:18:33 or even nix 2025-12-18 17:20:33 I just know that R is compiling the things needed from R which is not really secure 2025-12-18 17:35:16 Hello everyone 2025-12-18 17:35:58 https://packages.guix.gnu.org/ 2025-12-18 17:36:13 it would be good if a container disallowed all outgoing connections of apps 2025-12-18 17:37:13 Something that I am wondering. Why are several applets unavailable for xfce not installed by default, such as the network manager applet or sound? 2025-12-18 17:38:14 I just installed alpine linux with the setup-desktop script, chose xfce, a bunch of standard applications and applets are missing which most would consider essential components. 2025-12-18 17:38:27 LimitlessGrit[m]: you have to activate the community repo 2025-12-18 17:40:45 Is it just uncommenting the community repo in etc/apk/repositories, saving the file, doas apk update, & doas apk upgrade to enable it? 2025-12-18 17:45:01 LimitlessGrit[m]: there is a tutorial about it 2025-12-18 17:45:04 googl it 2025-12-18 17:45:32 doas setup-apkrepos -c 2025-12-18 17:45:33 I really dislike when people do that. Freaking lazy. 2025-12-18 17:45:35 that is the command 2025-12-18 17:46:05 See, now that wasn't so hard. 2025-12-18 17:46:09 I am not even a developer what you expect mate 2025-12-18 17:46:10 :p 2025-12-18 17:46:20 A non snarky comment. 2025-12-18 17:46:45 Helpful people who are willing to answer a question. 2025-12-18 17:46:53 You know, common decency stuff. 2025-12-18 17:46:59 Not, just google it. 2025-12-18 17:47:03 Fuck google lol. 2025-12-18 17:47:33 someone of you ever thought about freebsd 2025-12-18 17:47:34 ? 2025-12-18 17:47:34 :D 2025-12-18 17:48:09 Not my cup of tea. 2025-12-18 17:48:29 freebsd too nerdy 2025-12-18 17:48:29 ? 2025-12-18 17:49:50 I'm not fond of its mascot, origins, or how things are done. I've been on linux for over 15 years. Not my first round of using Alpine either. Just never used alpine with xfce before. I need a lightweight DE for my ancient server from 2012 to conserve system resources. 2025-12-18 17:50:24 LimitlessGrit[m]: why not use sway 2025-12-18 17:50:24 ? 2025-12-18 17:51:58 I've used it before, not really what I like either. If I was going to use just a window manager I'd prefer awesome, or openbox, but at this point in my life I just want a lean DE that just works. XFCE on Arch Linux is what I've been using and it has been fine, but I am trying to free myself from systemd for good. Which is why I am going back to Alpine Linux. 2025-12-18 17:52:50 I dont think system d is bad 2025-12-18 17:53:02 but for me systemd is too unstable 2025-12-18 17:53:03 lol 2025-12-18 17:53:19 I do. For a laundry list of reasons that are available on nosystemd.org 2025-12-18 17:53:45 Unfortunately a systemd-free person got ahold of my attention and converted me by making too much sense. 2025-12-18 17:53:57 I've ignored it for long enough, but the elephant in the room has got to go. 2025-12-18 17:54:35 it is basically only starting services and managing logs 2025-12-18 17:54:39 it can even be more secure 2025-12-18 17:55:30 It is bloated and has too much control with many security vulnerabilities. 2025-12-18 17:55:53 I like openrc and even runit for the simplicity and security aspect. 2025-12-18 17:56:17 well but imagine everything is community managerd 2025-12-18 17:56:31 what is secure is in our hallucination 2025-12-18 17:56:33 :d 2025-12-18 17:56:41 there is nobody doing code audits 2025-12-18 17:56:54 maybe in openbsd who knows 2025-12-18 17:58:32 asdasdsadsad: ...and makes "VMs" and does dns and is a bootloader and can manage secure boot. It's a suit of tools, having it all in one suit is uhh.. bit much 2025-12-18 17:58:44 I'm sure some random person somewhere is, but honestly that is for pretty much most software out there. Whether we hear about it and it gets patched should there be any issues, or they use the vulnerabilities to their advantage. Does not matter. I agree that security and privacy for the most part is an illusion and that anyone with enough power, influence, and resources can gain access to and control any operating system. 2025-12-18 17:58:44 Whether it be by code remotely, or in person with a hammer. 2025-12-18 18:00:01 LimitlessGrit[m]: and yet you have to trust the alpine devs 2025-12-18 18:00:04 ;)) 2025-12-18 18:00:13 Does not mean that I have to roll over, bend my ass up, and take it raw dog willingly. If I have the ability to secure my system and make it a little more difficult for attackers. Then I will go through that effort. They want to come at me with a hammer, that is what my guns are for. Either way I will not go silently into that good night. 2025-12-18 18:01:40 I don't trust the alpine devs either. I don't trust anyone, or any software exclusively. In the world of FOSS we live and die by reputation. Alpine over the years has developed a reputation for being lightweight and secure. Until evidence to the contrary arises they do not have my trust, but they do have my business for the time being. 2025-12-18 18:01:51 I can hack x11 devices 2025-12-18 18:01:56 easily 2025-12-18 18:01:57 :d 2025-12-18 18:02:08 Good for you. 2025-12-18 18:02:20 but wayland is not much harder either 2025-12-18 18:02:33 for example people using sway not locking their config files 2025-12-18 18:02:52 easy target to make believe everything runs good 2025-12-18 18:03:52 if alpine has your business 2025-12-18 18:03:53 Why does this matter, are you just whipping out your dick and doing a helicopter move just to show everyone here that yours is the biggest? 2025-12-18 18:04:18 if alpine has your business and people know it has your business then it should be even worse to use alpine ;) 2025-12-18 18:04:31 most people use alpine as server 2025-12-18 18:04:42 I dont 2025-12-18 18:04:58 Compromise systems, dont compromise systems. Do it as a profession, be a script kiddy with ai to hand hold you while you take a piss. I really don't care about the bragging, or snarkiness. 2025-12-18 18:06:48 but tbh I would rather have something of injecting malware in arch dependencies than in alpine dependencies because I am not sure how big alpine userbase is 2025-12-18 18:07:38 ai hand :D is still able to controll your mouse keyboard and to make you believe in the meantime everything is fine 2025-12-18 18:07:48 which is why I left x11 2025-12-18 18:07:56 Im at a point in my life where if Im going to be hacked Im going to be hacked. If I can take basic steps to try and prevent it. Then I do. Most people using Alpine Linux for servers is fine. However, like people who climb mountains. Im using it as a desktop even though it takes a little extra effort, because I know when I finally go through the hassle of climbing this mountain. Then I get a desktop that just works. 2025-12-18 18:10:23 I'm reaching the old man get off my lawn while holding a shotgun phase. I do what I do, because I want to. I don't really give a flying fuck what other people think, or expect of me. 2025-12-18 18:10:51 I'm a tenacious stoic with limitless grit. 2025-12-18 18:11:27 Anywho, i think the file transfer is finished for my backups. 2025-12-18 18:11:40 Time to install Alpine on this bitch and let er rip. 2025-12-18 18:15:24 LimitlessGrit[m]: I actually think about openbsd 2025-12-18 18:15:26 x) 2025-12-18 18:54:27 is alpine a container distro 2025-12-18 18:54:28 ? 2025-12-18 18:55:15 If you want it to be. 2025-12-18 18:55:22 It is the most common usecase for it. 2025-12-18 19:04:32 i use it as a desktop 2025-12-18 19:08:18 penguinz_rule[m]: do you use firewalld 2025-12-18 19:08:19 ? 2025-12-18 19:09:18 doas rc-update add firewalld default 2025-12-18 19:09:22 is it correct? 2025-12-18 19:09:44 asdasdsadsad: Google it 2025-12-18 19:10:04 LimitlessGrit[m]: no search results on google 2025-12-18 19:10:06 :( 2025-12-18 19:10:12 lol 2025-12-18 19:10:35 Im honestly surprised people out there still use google and their products. 2025-12-18 19:11:10 Duckduckgo, brave, chatgpt, and a bunch of others make it irrelevant. 2025-12-18 19:41:24 asdasdsadsad genuinely never set a firewall up, i should be fiiine 2025-12-18 19:54:27 penguinz_rule[m]: what is your approach to install rstudio? 2025-12-18 19:54:28 lol 2025-12-18 19:54:31 or no need 2025-12-18 19:54:33 ;) 2025-12-18 20:09:18 never used R so cant tell 2025-12-18 20:50:29 "i was pointing out neither https..." <- can i boop again 😮 2025-12-18 21:20:43 Is it documented anywhere why alpine uses busybox to start openrc and not just start openrc-init? 2025-12-18 21:26:57 seems like I will stay on alpine 2025-12-18 21:27:08 alpline linux became my bookholder 2025-12-18 21:27:18 so if anyone need copy right books hack me 2025-12-18 21:27:18 :D 2025-12-18 21:27:38 actually I have only one copyrighted book 2025-12-18 21:27:43 it is not that worth 2025-12-18 21:28:04 gn 2025-12-19 03:14:55 @penguinz_rule easily my favorite desktop distro, no contest 2025-12-19 06:37:58 alpine on desktop works better for me than anything else i tried so far 2025-12-19 07:10:06 my kernel has all this gross flaky brown stuff around the edges 2025-12-19 07:10:14 I think it might be rust 2025-12-19 07:38:30 rnkn: may just need to take a shower ? 2025-12-19 07:44:22 https://www.fsfla.org/ikiwiki/selibre/linux-libre/wash.png reminds me of this lmao 2025-12-19 08:23:41 greguu: huh? 2025-12-19 08:39:13 lol 2025-12-19 17:12:53 Get in touch with this platform for greatness you’ll definitely thank me later
ℹ️❤️
👇🏻👇🏻👇🏻 2025-12-19 17:12:53 https://t.me/+pa-CiKYv9-5lNWM8 2025-12-19 18:42:44 telegram always = scam these days it seems 2025-12-19 18:43:02 i dont use it since its proprietary and needs a phone number anyways 2025-12-19 19:29:01 I'm working on something where I need CONFIG_CRYPTO_DRBG_CTR=y and CONFIG_CRYPTO_DRBG_HASH=y enabled in the kernel config. It's not enabled by default in the linux kernel but debian and ubuntu ship it. Would there be any chance of this getting enabled in alpine? 2025-12-19 19:31:38 @cow123 is modifying and building locally out of the question? 2025-12-19 19:32:20 Using abuild for the kernel is really smooth sailing, but I know that on my craptop with 4GB RAM recompile was painful 2025-12-19 19:33:11 So when testing modules I needed enabled, I built on my desktop and moved the apk over 2025-12-19 19:33:59 yeah I guess I could just build on my own. I was hoping to upstream shufflecake encryption into aports which needs those kernel settings but I could probably make my own repo. 2025-12-19 19:34:43 Did you make an Issue or MR explaining? 2025-12-19 19:34:59 It sounds like it could be widely useful 2025-12-19 19:36:07 thanks, I'll make a MR. 2025-12-19 19:37:28 Sweet, hope it gets a merge 2025-12-19 19:56:31 trying to configure awall on raspberrypi. Firewall not enabled for inet6. is this a known issue with arm or am i congifuring wrong? 2025-12-19 20:02:22 Hello everyone. Need some help. I have installed Alpine Linux with gnome using network manager. I am having wifi connectivity issues. Wherein my wifi connection keeps cutting in and out randomly. I asked the brave ai to help me troubleshoot the issue. Here is the steps I have done so far, but I'm still having the issue. 2025-12-19 20:02:23 https://pastebin.com/inYD7ei5 2025-12-19 20:02:41 Does anyone have any ideas on how to fix it? 2025-12-19 20:10:24 how often does it cut in and out? what other services are running? 2025-12-19 20:11:12 Every minute, or two. As far as other services. It is a fresh installation, no real modifications. Should I look for anything specific/ 2025-12-19 20:11:50 s///?/ 2025-12-19 20:16:23 One thing to note is that I do not have this issue on another machine configured the same way. Only difference is that the wifi is using a tp link dongle. Versus my laptop which is a Lenovo Thinkpad. 2025-12-19 20:16:42 * tp link wifi dongle. Versus 2025-12-19 20:25:40 wow this is annoying. Any help is greatly appreciated 2025-12-19 20:27:20 i believe you may need to disable some other network services to use NetworkManager. Should be in the wiki 2025-12-19 20:27:56 networking and wpa_supplicant 2025-12-19 20:28:49 shouldnt NM use wpa_supplicant or am I wrong 2025-12-19 20:30:56 it does, but it is responsible for starting wpa_supplicant 2025-12-19 20:31:04 you can't start wpa_supplicant separately 2025-12-19 20:31:46 ah 2025-12-19 20:36:56 God I hope this works. I'm at that point of so much copy pasta from the wiki that you have a hard time keeping track on what you did. 2025-12-19 20:36:56 lol 2025-12-19 20:39:41 Seems to be holding steady 2025-12-19 20:39:48 I think adding the config file helped 2025-12-19 20:40:02 Thanks for the assistance, hopefully it still works upon reboot 2025-12-19 20:40:16 Either way I am mentally pooped now. 2025-12-19 21:10:32 i'm having trouble with the ssh-to-host section of ZERO TO AWALL.. does anyone have an up-to-date guide for awall? 2025-12-20 02:22:21 Error relocating /usr/lib/libQt6Xdg.so.4: _ZN14QObjectPrivateC2E16QtPrivate_6_10_0: symbol not found 2025-12-20 02:22:21 does anyone know how i can fix this? trying to use lxqt for vnc and this happened when i upgraded my system yesterday (edge) 2025-12-20 02:22:21 i tried reinstalling the package, nothing changed 2025-12-20 02:28:37 the package libqt6xdg is from needs to be rebuilt 2025-12-20 02:29:42 oh 2025-12-20 03:51:14 ended up rebuilding it then installing that and its fixed now 2025-12-20 04:48:26 surely other people are experiencing the "not responding" issues with GNOME on Alpine 3.23.x? 2025-12-20 05:09:14 Quantum_3[m]: you should probably report that to the alpine devs too 2025-12-20 05:10:17 abby: do i just open a issue in aports? 2025-12-20 05:10:42 no clue, i'm not involved in alpine, i just hang out here 2025-12-20 05:11:03 oh 2025-12-20 05:11:25 generally rebuilding the package fixes the issue with qt api change. 2025-12-20 05:11:58 yeah thats what i did to fix it 2025-12-20 05:11:58 i guess they havent rebuilt that package (even though qt got updated) 2025-12-20 05:12:43 it's possible it's already in the queue for rebuild 2025-12-20 05:23:02 there is a MR (!94680) open for rebuilding lxqt-panel, not sure that will address Quantum_3[m] 's issue 2025-12-20 09:50:30 https://youtu.be/mfv0V1SxbNA 2025-12-20 09:50:48 that linus and linux video ^ 2025-12-20 09:50:57 s/linux/linus/ 2025-12-20 15:25:04 Does anyone knows where I can find the script(s) used to build the rpi images in the download page of the alpine website? 2025-12-20 15:59:18 freerig[m]: https://git.alpinelinux.org/aports/tree/scripts/mkimg.arm.sh 2025-12-20 19:33:06 I have a system where openrc seems to be doing somethings weird: 1. rc-status output doesn't match reality 2. `rc-service ineed` is empty 3. services aren't starting (which is probably because of 2) 2025-12-21 07:06:50 Hello, I installed Alpine on Vmware and want to install xfce. 2025-12-21 07:06:50 ``` 2025-12-21 07:06:50 ```shell 2025-12-21 07:06:50 setup-desktop xfce 2025-12-21 07:06:50 apk add xf86-video-vmware 2025-12-21 07:06:52 But after reboot only black screen 2025-12-21 07:46:28 lazard: https://wiki.alpinelinux.org/wiki/Install_Alpine_on_VMware_ESXi mentions installing open-vm-tools and adding the open-vm-tools service 2025-12-21 07:49:58 after initial setup-alpine, you will need to ensure that the community repo is enabled: https://wiki.alpinelinux.org/wiki/Repositories#Managing_repositories 2025-12-21 07:54:54 it has been more than a few years since i messed around with vmware, so i'm afraid that the most i can probably help 2025-12-21 08:09:31 thank you jvvv but still black screen after enabling and starting open-vm-tools. Gnome works fine, not sure what's the problem. 2025-12-21 08:09:42 i'm not sure, but you may need to add the vmware graphics module to the initramfs 2025-12-21 08:11:33 'Gnome works fine' ? 'black screen' ... sorry these two statements seem to contradict, but perhaps i don't grasp what you are saying 2025-12-21 08:14:22 oh, you mean with xfce, haven't used it in a while and never did under virt 2025-12-21 18:45:31 Soooooo I had to troubleshoot and configure networkmanager yet again on another machine. Overall this is a VERY TEDIOUS PROCESS and a PAIN IN MY ASS. The wiki does explain how to do this in sufficient detail however I just don't see the point. So I said fuck it and made a bash script that will automate the process for either wpa_supplicant or iwd backends for plasma, gtk, or cli environments. I'm going to test it out over the 2025-12-21 18:45:31 next week to ensure it functions properly, but I would like to contribute this as FOSS software for the Alpine Linux community. Is there a way to go through a submission process so that way anyone who is trying to setup networkmanager can do so gracefully without any headaches? 2025-12-21 18:46:13 If so, what should I do? 2025-12-21 18:51:57 Put it up on something like codeberg or gitlab 2025-12-21 18:52:09 then package it 2025-12-21 18:52:23 clone aports.git, add your package to testing/, send a MR 2025-12-21 18:53:40 I've never packaged anything for alpine. Is there a wiki page that explains that process? 2025-12-21 19:01:36 https://wiki.alpinelinux.org/wiki/APKBUILD_Reference 2025-12-21 19:02:14 https://wiki.alpinelinux.org/wiki/Creating_an_Alpine_package 2025-12-21 20:51:52 Thank you kindly 2025-12-22 06:40:59 mpv on edge seems buggy, start a youtube-link, switch workspace back and forth (dwm) and the video shakes, spams "export failed" 2025-12-22 08:36:31 Hello, I updated my alpine install (3.21 > 3.23), all good, except loadkmap service fails to stop. (but starts fine) Any idea what could cause this ? (loadkmap is in boot runlevel) Thanks 2025-12-22 08:52:34 I commented the stop() function in /etc/init.d/loadkmap so I don't have this "loadkmap fails to stop" message when I power off. I don't know if it is really needed. 2025-12-22 08:54:55 so now it will never stop 2025-12-22 08:59:44 does it not log anything? 2025-12-22 09:24:16 longnoserob[m], "* ERROR: loadkmap failed to stop" in /var/log/rc.log when I power off the system, that's it. I tried, to put it in default then in sysinit runlevel, same. 2025-12-22 09:26:01 It's not really critical of course, everything works fine otherwise. I was just a bit surprised by this message. 2025-12-22 11:04:35 given that loadkmap's stop doesn't actually do anything, it should be harmless 2025-12-22 11:06:33 `return 0` would make openrc happy. you can edit it and change it to that 2025-12-22 12:26:03 wg-quick doesn't use doas instead of sudo. 2025-12-22 12:26:06 I think this is a bug? 2025-12-22 12:26:22 (in line 85 of wg-quick script) 2025-12-22 12:27:57 packages touching sudo/doas at all sounds like a bug to me 2025-12-22 12:28:09 true. 2025-12-22 12:30:33 I was trying to configure wg on alpine but I don't know why previously I failed, I followed the docs but tbh it doesn't work, something is missing in the docs. 2025-12-22 12:32:34 not all installations will have doas 2025-12-22 12:33:14 there's no real reason to use the bloated sudo, but anyone does, my condolences. 2025-12-22 12:34:01 you've just pointed out one reason 2025-12-22 12:34:31 alr that's not what I came for tho so lets not derail. 2025-12-22 12:34:37 could anyone help me setup wg? 2025-12-22 12:34:51 idk, but wg-quick doesn't work. 2025-12-22 12:34:56 apk add sudo 2025-12-22 12:35:05 Error: Unknown device type. 2025-12-22 12:35:08 ^ I get this. 2025-12-22 12:35:26 lopid: well it seems that's just if I'm executing the script directly. 2025-12-22 12:35:40 what line gives that error? 2025-12-22 12:35:57 the very first. 2025-12-22 12:36:11 after `rc-service wg-quick.wg0 start` 2025-12-22 12:36:19 (with doas ofc) 2025-12-22 12:37:41 did you follow https://wiki.alpinelinux.org/wiki/Configure_a_Wireguard_interface_(wg) ? 2025-12-22 12:37:53 yeah. 2025-12-22 12:38:25 does `ip l` show the wireguard interface? 2025-12-22 12:38:45 no I thought the wg-quick script was supposed to bring that up. 2025-12-22 12:41:07 broke: https://gitlab.alpinelinux.org/alpine/aports/-/blob/master/main/wireguard-tools/wg-quick-different-sudo.patch 2025-12-22 12:41:35 i tried to get something upstream but didnt follow up 2025-12-22 12:41:59 if someone wants to continue, that would be hugely appreciated 2025-12-22 12:42:14 (https://lists.zx2c4.com/pipermail/wireguard/2025-April/008727.html) 2025-12-22 12:49:15 achill: well, what's the patch about? 2025-12-22 12:49:39 it allows you to use a different sudo by setting SUDO=doas as a env variable 2025-12-22 12:49:39 and why do alpine uses gitlab, ;-; 2025-12-22 12:49:46 but likely needsa a better implementation 2025-12-22 12:49:51 oh ok makes sense. 2025-12-22 12:49:53 like detecting if sudo or doas is available 2025-12-22 12:50:11 oh I never knew SUDO was a thing. 2025-12-22 12:50:15 as an env. 2025-12-22 12:50:25 nonetheless I am to set that? 2025-12-22 12:50:28 alr thanks. 2025-12-22 12:50:42 or maybe just set it inside the init script? 2025-12-22 12:50:57 or are more stuff on alpine going to use SUDO? 2025-12-22 12:50:58 it is a thing but nothing more, i dont think this is actively used 2025-12-22 12:51:14 likely not, as i said, that probably needs a better implementation 2025-12-22 12:52:35 (using sudo in wg-quick in general is not great, but i think its too late now and people rely on it now 🙃) 2025-12-22 12:55:05 well, I don't think this: 2025-12-22 12:55:06 Error: Unknown device type. 2025-12-22 12:55:13 is caused by sudo btw. 2025-12-22 12:55:15 so... 2025-12-22 12:55:53 It happens after wg runs the command 2025-12-22 12:55:54 [#] ip link add dev wg0 type wireguard 2025-12-22 12:56:04 Why doesn’t our GitLab send email notifications? 2025-12-22 12:58:35 kernel needs wireguard support? 2025-12-22 12:58:44 well it seems. 2025-12-22 12:58:48 I thought alpine already had it? 2025-12-22 12:58:54 on the LTS version as well. 2025-12-22 12:59:04 dunno 2025-12-22 12:59:28 try `modprobe wireguard` 2025-12-22 12:59:32 well, the doc didn't mention anything about the kernel. 2025-12-22 12:59:58 ok modprobe returns FATAL. 2025-12-22 12:59:59 so nope. 2025-12-22 13:00:41 so how am I to get the wireguard module into the kernel? 2025-12-22 13:01:00 is there a package for it to handle this? 2025-12-22 13:01:10 is it already there? lsmod 2025-12-22 13:01:24 (assuming it's built as a module) 2025-12-22 13:02:59 well it doesn't show in lsmod. 2025-12-22 13:03:01 meaning? 2025-12-22 13:18:27 hello 2025-12-22 13:18:37 People I have a problem here. 2025-12-22 13:18:46 I can't install anything 2025-12-22 13:18:55 apk command is dead. 2025-12-22 13:18:58 https://voidbin.com/paste/d7b87bc9-209c-4b51-8295-1f16a12a20d4 2025-12-22 13:19:23 Error loading shared library libapk.so.2.14.9: No such file or directory (needed by /sbin/apk) 2025-12-22 13:20:15 please give me a solution. I am in a bind here. 2025-12-22 13:21:15 try using the static apk? 2025-12-22 13:22:24 and 2.14 seems oldish 2025-12-22 13:23:06 longnoserob[m]: how? 2025-12-22 13:23:24 I can't install anything/ 2025-12-22 13:24:49 can setup-bootable be installed to nvme? it works perfectly on a usb, but the nvme wont boot 2025-12-22 13:25:53 ghdr check if apk.static is already installed? 2025-12-22 13:26:54 longnoserob[m]: nope 2025-12-22 13:28:08 test: which system, recent x86_64 with uefi? 2025-12-22 13:29:22 yes 2025-12-22 13:30:24 @test: https://wiki.alpinelinux.org/wiki/NVME 2025-12-22 13:36:48 ghdr: https://gitlab.alpinelinux.org/alpine/apk-tools/-/releases/v2.14.9 2025-12-22 13:46:36 Why does the scrcpy package have so many dependencies? The releases offered by its Github page are only a single binary. 2025-12-22 13:47:02 maybe the github release is statically linked 2025-12-22 13:47:22 or its dependencies are optional with dlopen 2025-12-22 13:49:19 What is the meaning of "dlopen"? 2025-12-22 13:49:40 dynamic library open 2025-12-22 13:49:58 it permits to import dynamic library at runtime 2025-12-22 13:54:11 I just checked it, it is mostly statically linked (the program is 24M), but linked against libc.so, libm.so and libudev.so 2025-12-22 13:54:33 alpine's policy however is to not statically link programs 2025-12-22 13:55:28 this is how the scrcpy's installed size on alpine is 285 KiB 2025-12-22 13:58:29 Can anyone recommend any virtualization software that runs on alpine linux? 2025-12-22 13:58:57 I use qemu with kvm 2025-12-22 14:04:53 doas apk add qemu? 2025-12-22 14:05:18 Is there a way to get the scrcpy binary released by Github running in Alpine? 2025-12-22 14:05:18 so well, how do I get wireguard module on the LTS kernel? 2025-12-22 14:05:22 Is that how to get it up and running properly? 2025-12-22 14:05:30 or do I need to use a non standard kernel for wireguard? 2025-12-22 14:05:35 or the main kernel 2025-12-22 14:05:36 ? 2025-12-22 14:06:34 lupo: afaik no, that uses glibc, and alpine runs on musl 2025-12-22 14:06:38 Weird. Gnome software just crashes everytime I try to install an update for flatpaks.... 2025-12-22 14:07:10 lupo: what is wrong with the packaged scrcpy? 2025-12-22 14:07:30 broke: linux-lts and linux-stable already have the wireguard module 2025-12-22 14:07:47 I don't see it tho. 2025-12-22 14:07:50 which is weird. 2025-12-22 14:08:02 did I forget to enable it or smt? 2025-12-22 14:08:19 (include/enable) 2025-12-22 14:08:39 LimitlessGrit[m]: do you have a error message or a backtrace? 2025-12-22 14:08:54 aron: wait how can a project strictly use glibc? 2025-12-22 14:09:07 aron, as I stated above installing the scrcpy Alpine package pulls many other packages 2025-12-22 14:09:10 their pre-packaged binary is linked against glibc 2025-12-22 14:09:14 broke: modprobe wireguard? 2025-12-22 14:09:28 lupo: "many other packages" -> so what? 2025-12-22 14:09:31 lupo: the dependencies will be downloaded anyway, either in the same binary or as seperate packages 2025-12-22 14:09:59 oh pre-compiled. 2025-12-22 14:10:18 idk, I always compile non packaged stuff tbh. 2025-12-22 14:10:27 I don't trust even them. 2025-12-22 14:10:50 Since I`m not on my Alpine system I can't answer that now 2025-12-22 14:10:50 broke: scrcpy github release is built against glibc, somehow it made sense for them 2025-12-22 14:10:56 aron: modprobe wireguard shows that wireguard is not there. 2025-12-22 14:11:06 as in spits out "FATAL" 2025-12-22 14:11:10 did you upgraded your kernel recently? 2025-12-22 14:11:10 and stuff. 2025-12-22 14:12:01 I'm on 6.18.1-r0 2025-12-22 14:12:27 wait. 2025-12-22 14:12:31 what? 2025-12-22 14:12:32 if you upgrade the kernel on a running system, then you have to reboot 2025-12-22 14:12:47 no wait, I was wrong. 2025-12-22 14:12:53 gimme a min, I'm confused. 2025-12-22 14:13:02 achill: 14:12:35:603 flatpak Could not unmount revokefs-fuse filesystem 2025-12-22 14:13:18 hmmm interesting 2025-12-22 14:13:25 14:12:35:607 GLib g_propagate_error: assertion 'src != NULL' failed 2025-12-22 14:13:36 Those are the two relevant ones 2025-12-22 14:14:00 broke: what 'uname -a' spits? what 'ls /lib/modules' spits? 2025-12-22 14:15:00 Anyone have any ideas? 2025-12-22 14:15:05 longnoserob: the system is booted and loaded into ram with setup-bootable, the system works with nvme on a normal alpine install 2025-12-22 14:15:15 Linux domain.tld 6.12.56-0-lts #1-Alpine SMP PREEMPT_DYNAMIC 2025-10-31 13:16:09 x86_64 Linux 2025-12-22 14:15:57 `ls /lib/modules/`: 6.18.1-0-lts 2025-12-22 14:16:09 so, you're running a "6.12.56-0-lts" kernel, but you have modules for "6.18.1-0-lts" kernel only 2025-12-22 14:16:11 reboot 2025-12-22 14:16:26 oh damn right. 2025-12-22 14:16:39 I'm an idiot. 2025-12-22 14:16:43 I am so damn sorry. 2025-12-22 14:16:47 np 2025-12-22 14:16:48 I missed this. 2025-12-22 14:16:57 well thanks. 2025-12-22 14:17:24 should work after reboot 2025-12-22 14:21:01 whats the part of the internet stack that allows you to assign hostnames to ${hostname}.local on say an rpi so you can ssh/browse websites hosted via the ${hostname}.local domain name 2025-12-22 14:21:04 is it some sort of dns then 2025-12-22 14:21:13 * of dns thing then 2025-12-22 14:24:00 lupo: trust apk, it will download all deps for you, that's the purpose of using a package manager, as a user it does not really matter whether it is statically or dynamically linked (well, there are some exceptions, but they are very specific) 2025-12-22 14:24:03 avahi? neighbour discovery? 2025-12-22 14:26:14 fuck so i will probably need avahi2dns on alpine edge,,, 2025-12-22 14:27:36 penguinz_rule[m]: exists already in aports.. https://pkgs.alpinelinux.org/package/edge/testing/x86_64/avahi2dns 2025-12-22 14:28:02 yeah but i dont really want to go beyond upgrading to v3.23 in order to use it 2025-12-22 14:28:22 since its testing 2025-12-22 14:28:58 then you are not running edge.. 2025-12-22 14:31:59 ye, i didnt imply that, very tempted to just use edge despite it being used as a server 2025-12-22 14:32:42 didnt had much breaking on edge lately.. 2025-12-22 14:35:43 longnoserob: did you see my reply? 2025-12-22 14:41:01 yes, but have no further idea right now 2025-12-22 14:41:32 ok 2025-12-22 14:52:25 Hi, does anyone else have a problem with plasma wayland in v3.23 ? The plasma compositor (kwin i guess) starts, but the plasma shell doesn't (no status bar...) 2025-12-22 14:52:26 Both the manual installation and setup-desktop didn't work, I've tried everything 2025-12-22 18:04:14 penguinz_rule[m]: that's mdns 2025-12-22 18:04:36 penguinz_rule[m]: https://whynothugo.nl/journal/2025/12/16/resolve-.local-mdns-domains-via-avahi-using-unbound/ 2025-12-22 18:04:46 I think we can move avahi2dns into community 2025-12-22 18:07:24 thanks for the blog 2025-12-22 18:12:19 WhyNotHugo: also for the unbound conf file is it tabs or spaces 2025-12-22 18:18:16 spaces work, used unbound-checkconf and no errors 2025-12-22 19:44:50 hi I updated my alpine machine, for some reason, I'm stuck at "(auto) login:" I have never seen this so this is confusing. 2025-12-22 19:49:42 on a tty? 2025-12-22 19:50:30 yeah. 2025-12-22 19:50:37 I'm thinking of reinstalling atp. 2025-12-22 19:50:49 this seems incomplete 2025-12-22 19:50:49 https://wiki.alpinelinux.org/wiki/MDNS#udhcpc 2025-12-22 19:50:49 since my system just has one resolv.conf i presume the post-bound script just uses that instead 2025-12-22 19:51:11 I updated after a long time, like 2months ig, and this happened. 2025-12-22 20:44:47 do i need to have avahi2dns in order to resolve dns queries for ${hostname}.local on my host machine, i think i may need to since i am trying to do drill -p 5354 @192.168.102 database.local and its not picking it up, even though its avahi2dns is on that machine, and is running on that port since i can run the command but with 127.0.0.1 on the target machine 2025-12-22 20:45:57 i wonder if this would work normally on windows machines/linux machines with avahi2dns on it 2025-12-22 20:46:09 since i intend to have windows connect to this 2025-12-22 20:52:08 avahi-browse works though on other machine 2025-12-22 21:12:24 inb4 i mixed up the unbound conf port numbers 2025-12-23 01:10:44 yggdrasil fails to install in a container because the post-install script tries to use modprobe (but apk fix doesn't error after, does that not run post-install again?): https://gitlab.alpinelinux.org/alpine/aports/-/blob/master/community/yggdrasil/yggdrasil.post-install 2025-12-23 01:11:19 I think I will open a MR to add || true to that line to make it not fail if modprobe fails 2025-12-23 07:25:49 hello 2025-12-23 07:26:13 which package provides libvirtd and the daemon services? 2025-12-23 07:26:27 I am getting "Unable to connect to libvirt qemu:///system." 2025-12-23 07:27:08 try libvirt-daemon and libvirt-daemon-openrc 2025-12-23 07:27:33 raven523: they are packages to be installed or name of the command to be run. 2025-12-23 07:27:39 packages 2025-12-23 07:32:06 The -openrc package does not have to be installed explicitly 2025-12-23 10:47:57 Greetings 2025-12-23 10:48:42 I'm a bit lost here. How do I create my own local repository with custom packages? Is buildrepo the tool to use? 2025-12-23 11:09:03 cousin_luigi: buildrepo could indeed be used if you want to make sure all packages of a specific repo get built. 2025-12-23 11:41:32 ikke: Thanks. I found out what I need to do is much simpler than I remembered. 2025-12-23 11:42:07 By the way, with a header-only library, therefore with a -dev package only, how do I avoid the main one from being built? 2025-12-23 11:47:12 If it's header-only, having a -dev subpkg does not make a lot of sense 2025-12-23 11:47:32 The main package is always built 2025-12-23 11:49:06 does rust package have wasm enabled? 2025-12-23 11:50:38 https://pkgs.alpinelinux.org/package/edge/main/x86_64/rust-wasm 2025-12-23 11:53:50 ikke: It's for clarity. 2025-12-23 11:54:15 Anyway, am I supposed to prefer musl over glibc? 2025-12-23 12:12:08 cousin_luigi: I'm not following? Alpine Linux only supports musl 2025-12-23 13:36:18 ikke: It pulled in glibc-dev during the build. Is it a placeholder? 2025-12-23 13:49:01 cousin_luigi: aports does not provide glibc-dev, not sure where that comes from 2025-12-23 14:00:40 I wonder if I was hallucinating. 2025-12-23 14:01:34 by any chance was it glib-dev 2025-12-23 14:28:49 oh, must have been. Whole other ballgame. 2025-12-23 15:32:36 ikke: (1/7) Installing libstdc++-dev (15.2.0-r2) 2025-12-23 15:32:46 Doesn't musl replace that too? 2025-12-23 16:09:02 maybe in another timeline where muslc++ exists 2025-12-23 16:46:38 I see:) 2025-12-23 16:47:20 Are there known instances in which musl interferes with the network stack? I built a package for alpine and it doesn't seem to be able to bind. 2025-12-23 16:52:46 need more info 2025-12-23 16:53:54 pj: It's the dvblast package. 2025-12-23 16:54:18 I'm not even 100% sure I'm using it correctly, but the examples are quite clear. 2025-12-23 16:55:07 what address and port are you binding it to 2025-12-23 17:00:04 pj: I tried the local IP on eth0 (say 192.168.1.223) port 1234 or 2000 or 20000 2025-12-23 17:00:20 Tried the 224 range too 2025-12-23 17:00:53 and 239 2025-12-23 17:00:59 I can't see a thing with ss 2025-12-23 17:13:00 cousin_luigi: I'd suggest to bind to address 0.0.0.0 2025-12-23 17:13:08 also, share the errno 2025-12-23 17:15:22 aron: There is no error. I just don't see it listening and the stream is also inaccessible. 2025-12-23 17:15:53 I asked the developer a few minutes ago, but they might not reply soon. 2025-12-23 17:16:06 netstat can list open sockets 2025-12-23 17:16:38 I don't really understand why would you bind to a random ip address, can you explain it to me? 2025-12-23 17:17:43 aron: How is it random? 2025-12-23 17:18:37 to me, "Tried the 224 range too" sounds strange 2025-12-23 17:18:40 anyway 2025-12-23 17:18:47 aron: For multicast. 2025-12-23 17:18:47 bind it to 0.0.0.0 2025-12-23 17:18:52 Which the tool supports. 2025-12-23 17:19:02 224 and 239 are both within the multicast group address range 2025-12-23 17:19:07 then list open tcp sockets with netstat 2025-12-23 17:19:11 although I don't *think* the sender is supposed to bind to the group, anyway 2025-12-23 17:19:12 aron: I did. Nothing on ss -nlp4 or netstat 2025-12-23 17:19:17 it's the receivers that do, kind of 2025-12-23 17:19:18 aron: it's udp, I believe 2025-12-23 17:19:29 lemme try again 2025-12-23 17:19:41 and then the receivers also don't quite bind to the group address, but rather join it using a sockopt 2025-12-23 17:19:43 netstat lists udp as well 2025-12-23 17:20:51 crap, I fear it might be a firewall problem in the router inbetween 2025-12-23 17:21:16 ah, yeah, receivers do bind to the group address -- but the sender probably doesn't 2025-12-23 17:21:55 No, I don't see anything coming in 2025-12-23 17:22:04 what the heck is going on here 2025-12-23 17:22:14 there's a router in between? does it support multicast forwarding 2025-12-23 17:23:16 I'd still investigate on the host side: is the socket there? if your program doesn't create any socket, then it won't work 2025-12-23 17:24:18 I can't see traffic coming out from my client either. I feel I'm going nuts. 2025-12-23 17:24:46 not traffic, just pure open udp socket 2025-12-23 17:26:13 aron: Shouldn't ffprobe or ffplay send out something? tcpdump doesn't show a thing 2025-12-23 17:26:16 I don't think the client is supposed to produce any traffic 2025-12-23 17:26:21 hmmmm 2025-12-23 17:26:22 especially not if you're doing multicast 2025-12-23 17:26:27 at most it will produce an IGMP 'join' 2025-12-23 17:26:39 oooh 2025-12-23 17:26:45 in multicast mode, the sender just continuously blasts packets into the group 2025-12-23 17:26:52 and the network filters it out, until someone 'joins' the group 2025-12-23 17:27:20 yeah, but it won't send anything if no socket is opened ;) 2025-12-23 17:28:42 what am I supposed to see, server-side? 2025-12-23 17:29:13 'netstat -u' should yield something 2025-12-23 17:29:14 I see strange L2 broadcasts 2025-12-23 17:29:39 containing what? 2025-12-23 17:30:01 ok I guess I have to boot up the pc with the DVB-T card 2025-12-23 17:30:13 aron: udp 0 0 192.168.1.223:34172 224.0.0.166:20000 ESTABLISHED 2025-12-23 17:30:33 grawity: ethertype unknown 2025-12-23 17:30:37 great, so you have a working socket 2025-12-23 17:30:42 what's the ethertype number? 2025-12-23 17:30:49 I doubt it's related though 2025-12-23 17:31:04 grawity: 0x88e1 and 0x8912 2025-12-23 17:31:26 88e1 is powerline networking (homeplug) management 2025-12-23 17:31:50 I think I remember something like fritzbox or such ISP gear using 88e1 and 8912 though 2025-12-23 17:32:10 well, 88e1 and some other "unknown" 8xxx ethertype 2025-12-23 17:33:19 8912: Ethertype used for mediaxtream Specification protocols 2025-12-23 17:33:31 88e1: HomePlug Specification AV MME 2025-12-23 17:33:38 (https://standards-oui.ieee.org/ethertype/eth.txt) 2025-12-23 17:34:04 grawity: Oh, perhaps it's coming from the Fritz box. 2025-12-23 17:34:21 Since for some reason I set up a VLAN to reach it directly. 2025-12-23 17:34:51 Maybe it was one of those asterisk experiments to avoid losing my mind with SIP. 2025-12-23 17:35:34 dvblast is just giving me the usage text 2025-12-23 17:36:01 does not bother telling me "missing option -x or -y or -z" 2025-12-23 17:36:55 grawity: vlc rtp://224.0.0.223:20000 2025-12-23 17:37:07 main demux error: socket bind error: Address already in use 2025-12-23 17:37:23 But that's on openSUSE, my client. Perhaps a matter for #networking instead. 2025-12-23 17:46:41 I got multicast working with mumudvb, but it doesn't seem like my wi-fi APs do any igmp-based filtering 2025-12-23 17:47:02 ouch 2025-12-23 17:47:13 so it just floods the air with 10 tv channels 2025-12-23 17:48:23 so its unicast HTTP streaming works better here 2025-12-23 17:57:35 cousin_luigi: I gave up trying to get dvblast to receive anything, it says "bandwidth: 0" -- so naturally it's not going to produce UDP output either 2025-12-23 17:57:38 cousin_luigi: use mumudvb instead 2025-12-23 17:58:33 grawity: dvblast seems to tune 2025-12-23 17:58:45 it tunes, sure, and then doesn't do anything with that 2025-12-23 17:58:48 grawity: Did you pass it something with -b ? 2025-12-23 17:58:52 yes 2025-12-23 17:59:23 it defaults to 8, which it doesn't say if it's in Hz or kHz or MHz, but I tried 8 with a bunch of zeroes like in mumudvb's output 2025-12-23 17:59:41 -f definitely seems to be in Hz 2025-12-23 17:59:47 grawity: Where do you see bandwidth: 0 ? 2025-12-23 17:59:58 debug: symbolrate min: 0, max: 0, tolerance: 0 2025-12-23 18:00:02 that? 2025-12-23 18:00:08 -6 --print-period periodicity at which we print bitrate and errors (in ms) 2025-12-23 18:00:24 it then outputs "bitrate: 0" periodically 2025-12-23 18:01:25 I wonder if it might not be the wrong frontend 2025-12-23 18:01:37 idk, I only have one 2025-12-23 18:01:45 and it works with other tools 2025-12-23 18:01:51 is it dvb-t2 only? 2025-12-23 18:01:59 I wonder if dvblast just really insists on talking to a CAM decoder 2025-12-23 18:02:04 and gives up if it doesn't find one 2025-12-23 18:02:18 It would tell you, I suppose 2025-12-23 18:02:22 it does 2025-12-23 18:02:27 "warning: failed opening CAM device /dev/dvb/adapter0/ca0 (No such file or directory)" 2025-12-23 18:02:32 that's a bit suspicious 2025-12-23 18:02:44 makes sense by itself 2025-12-23 18:02:58 but given dvblast's other behavior (like "error: no config file" and then it doesn't exit, just stays there doing nothing) 2025-12-23 18:04:03 also the card is 'old' DVB-T only, but that's what is still used here, and that's what dvblast autodetects 2025-12-23 18:05:39 strace shows it actually binds the UDP sockets, by the way 2025-12-23 18:08:12 yeah I saw that 2025-12-23 18:08:32 I can't tell it to use dvb-t2, which is what is used here 2025-12-23 18:08:39 -n1 produces an error 2025-12-23 18:08:53 so, lemme see this mumudvb 2025-12-23 18:11:21 grawity: 2025-12-23 18:11:21 Support of DVB-S2, DVB-S, DVB-C, DVB-T and ATSC 2025-12-23 18:11:26 Doesn't help me. 2025-12-23 18:15:47 eh, I'd try it anyway to see what happens 2025-12-23 18:15:56 with https://nullroute.lt/tmp/txt/20251223.201538.60vrPC.txt 2025-12-23 18:17:35 where do I find the source for 2.1.0 ? 2025-12-23 18:18:03 oh, it's tagged on github 2025-12-23 20:39:49 hi 2025-12-23 21:53:43 Hi 2025-12-24 03:10:23 Hi, I'm having a lot of issues booting the -virt or -standard ISO in qemu 10 with the following params, have also tried serial output https://pb.envs.net/?ebe34680791794b6#D5izViSuw4rfBTqZmaaBJPHK3vTykrLsFKnaMMzoUaAA 2025-12-24 03:27:07 OK, it seems like Alpine won't boot in QEMU with OVMF, but will with SeaBIOS 2025-12-24 03:35:53 Fixed by updating OVMF. 2025-12-24 05:43:24 any advice on starting bluetooth *after* dbus? they're both in runlevel default and for some reason bluetoothd is (i presume) 2025-12-24 05:43:49 not waiting on dbus to stat, bc it crashes and a look at /var/log/messages says that it can't find a dbus 2025-12-24 06:53:35 acdw: the openrc service script probably depends on dbus, which means it should wait until dbus has started before starting the bluetooth daemon 2025-12-24 06:53:35 at least from my knowledge 2025-12-24 06:53:35 maybe post the error here? 2025-12-24 10:07:30 How do I rename the source tarball filename in APKBUILD? 2025-12-24 10:08:14 cousin_luigi: prefix the source entry with :: 2025-12-24 10:09:27 ikke: THanks! 2025-12-24 10:34:39 Is there a special way to pass custom CFLAGS during the build? 2025-12-24 10:34:52 Other than export CFLAGS="$CFLAGS foo" ? 2025-12-24 10:37:35 No, exactly that 2025-12-24 12:10:55 grawity: Ok, I finished packaging mumudvb for alpine: I have, I think, a unicast stream but I'm not sure how to access it now. Is it rtp:// ? udp:// ? 2025-12-24 16:53:41 What's your approach to getting glibc binaries working? What's worked best for you? I'm not 100% sure of how exactly's the best way to go at it, but the goal's to get full GUI apps working as if they were native. I've heard of 3 methods so far, gcompat, the glibc apk by sgerrand, and weirdest of all, symlinking both the libc and other libraries from a glibc-based chroot to Alpine. 2025-12-24 16:57:42 Jake[m]: either gcompat, or distrobox 2025-12-24 16:58:11 I use distrobox when I need to use firefox with widevine drm (which naturally only works in glibc) 2025-12-24 16:58:37 Does gcompat work right with GUI apps? 2025-12-24 17:01:09 probably 2025-12-24 17:01:43 f_: damn 2025-12-24 17:04:53 Try it :) 2025-12-24 19:00:08 X or Wayland. Short answers only. 2025-12-24 19:00:16 :) 2025-12-24 19:00:41 * X or Wayland, and why. 2025-12-24 19:02:11 I use X. It works well for me. 2025-12-24 19:02:31 thefox[m]: Same. It just works fine everywhere 2025-12-24 19:02:46 It's a shame it's being left behind 2025-12-24 19:05:20 ld: cannot find -lxdo 2025-12-24 19:05:36 Jake[m]: there is https://xlibre-alpine.github.io/website/ 2025-12-24 19:06:23 lmao 2025-12-24 19:06:47 Jake[m]: I saw something recently about Gnome Mutter dropping support for Wayland which I don't think is a good idea. I use xfce4 and StumpWM currently. 2025-12-24 19:07:12 s/Wayland/X11/ 2025-12-24 19:07:36 realroot[m]: Yo wait are you the same guy that's in nadeko's matrix? 2025-12-24 19:07:55 yes 2025-12-24 19:07:58 WTH 2025-12-24 19:08:00 Yo 2025-12-24 19:08:20 i am using x and alpine too 2025-12-24 19:08:40 realroot[m]: As a daily driver ?? 2025-12-24 19:08:42 "Community driven" a.k.a. "There's no that pesky maintainer that keeped me at bay" 2025-12-24 19:09:06 Jake[m]: yes even on phone with pmos 2025-12-24 19:09:26 realroot[m]: You have forever earned the entirety of my respect 2025-12-24 19:09:49 Oh, I thought you used Gentoo 2025-12-24 19:10:21 I like Alpine a lot because it's basicly universal 2025-12-24 19:10:26 And fully bloatless 2025-12-24 19:10:46 Like, as bloatless as they could've made it. 2025-12-24 19:11:15 By default, it works anywhere, as it's stupid light 2025-12-24 19:11:30 And if you want it to do more, you can always install whatever you need 2025-12-24 19:11:46 It's the perfect DIY distro, better than Arch even 2025-12-24 19:12:13 musl support's the only thing keeping it from being the best in my oppinion 2025-12-24 19:12:49 "ld: cannot find -lxdo" <- xdotool-dev 2025-12-24 19:14:50 for me musl is the why alpine is one of the bests distros there is .... 2025-12-24 19:15:31 s/bests/best/ 2025-12-24 19:15:45 visone[m]: I'm not saying it's bad, quite the opposite, really, it's just not supported as broadly as it should 2025-12-24 20:01:40 "I'm not saying it's bad, quite..." <- honestly its supported quite well, just have to build for musl. 2025-12-24 20:02:29 for example firefox does not support it officially without devs help is not easy 2025-12-24 20:02:37 ichigo[m]1: 1st of all, fair, 2nd of all, hello Ichigo, it's Ichigo 2025-12-24 20:02:50 xDD same thought, yus good to see a fellow ichigo 2025-12-24 20:03:23 but yeah if its already built it can be a problem like steam, cant build that yourself, but theres ways around it, 2025-12-24 20:04:08 if it's built it is not open (obviously) so I prefer to avoid it 2025-12-24 20:04:27 "I saw something recently about..." <- gtk5 will drop x support 2025-12-24 20:04:53 that just gives more work for people to backport 2025-12-24 20:05:09 but i see why they doing it, X been going to the sidelines 2025-12-24 20:05:13 https://git.linuxping.win/12to11/12to11 2025-12-24 20:05:19 daily reminder xlibre is a clownshow 2025-12-24 20:05:26 realroot[m]: WHAT 2025-12-24 20:05:32 Please no 2025-12-24 20:06:14 i use mostly gtk3 2025-12-24 20:06:31 yeah Jake should be fine for a while, most apps arent even on GTK4 yet it seems 2025-12-24 20:06:37 realroot[m]: Messed up. 2025-12-24 20:07:11 ichigo[m]1: Ik, Ik, but still, that's yet another potential EOL for Xorg 2025-12-24 20:07:14 also reminder matrix replies look like shit on irc clients 2025-12-24 20:07:37 12to11 it's for running wayland app in x11 2025-12-24 20:07:37 Like, what are we, Windows? 2025-12-24 20:08:00 tru i think its a bit weird to just randomly end support for a brogram, its just not the linux way 2025-12-24 20:08:13 unix been around since 1980, its not like windows where they delete themselves every 5 years 2025-12-24 20:08:23 realroot[m]: Really? I've been looking for a way to do that for a while but I couldn't find anything 2025-12-24 20:08:37 oh cool, might try foot xD 2025-12-24 20:08:47 ichigo[m]1: Right? Linux's not like that. 2025-12-24 20:09:05 yes i have to open an issue though cause I cannot build it 2025-12-24 20:09:47 interesting might take a look later and see if i can get it to build 2025-12-24 20:09:53 were you packaging it with abuild? 2025-12-24 20:10:08 yes let me you upload it so that you have it 2025-12-24 20:10:16 cool thanks 2025-12-24 20:10:50 i took the last commit some time ago https://paste.trom.tf/emigufajac.makefile 2025-12-24 20:11:06 since it is not released 2025-12-24 20:11:45 cool thanks, if i fix it i will upload it 2025-12-24 20:12:15 idk what the error you had but most the time its just dependancy issue or a quirk in the build/install command 2025-12-24 20:13:02 err https://paste.trom.tf/tinafafawi.yaml 2025-12-24 21:19:52 >imake -DUseInstalled -I/usr/lib/X11/config 2025-12-24 21:19:52 yeah same error im running into, but that directory actually doesnt exist on alpine, 2025-12-24 21:19:52 imake: No such file or directory 2025-12-24 21:20:05 i did realize it hasnt been updated in a while 2025-12-24 21:20:21 could probably make a patch though, not sure exactly what its wanting to do 2025-12-24 21:25:03 https://paste.trom.tf/zifequcetu.yaml 2025-12-24 21:25:25 i think it might be in the source code 2025-12-24 21:26:11 ngl never even heard of imake before 2025-12-24 21:31:19 ichigo[m]1: it is old Xorg build tool, available from imake pkg 2025-12-24 21:31:25 oh ok installing the deps makes that path, 2025-12-24 21:31:28 >imake is a build automation system for X Window System and related software. It was replaced by GNU Autotools in 2005 and may be replaced by Meson in the future. 2025-12-24 21:31:32 i guess makes sense if its not around anymore, wonder why he used it 2025-12-24 21:32:18 as a guess, could be based on other old code 2025-12-24 21:33:40 yeah, that would make sense. i also havent managed to get it to build yet 2025-12-25 00:51:44 Can anybody explain to me the whole Xlibre drama thing? It's been a while, but I still don't get it 2025-12-25 00:52:58 They say they're being attacked, others say they're barely improving anything, and others say XLibre is just straight up bs 2025-12-25 00:53:23 It's all confusing as heck. 2025-12-25 00:54:38 All I know is that the fork was made by some Marco guy who thought Xorg was being sabotaged. 2025-12-25 00:55:19 A 2025-12-25 00:56:04 message delete is not helpful on a channel bridge to irc.. 2025-12-25 00:56:33 * All I know is that the fork was made by some [Correction: Enrico] guy who thought Xorg was being sabotaged. 2025-12-25 00:57:45 longnoserob[m]: FOk? Do edits show up fine then? 2025-12-25 00:57:51 s/FOk/Ok/ 2025-12-25 00:58:15 My cat won't stop stepping on my damn laptop. 2025-12-25 00:59:23 Whatever man. I'll do the research then. :/ 2025-12-25 00:59:32 I 2025-12-25 01:14:42 edits are reposts on irc 2025-12-25 08:07:21 "All I know is that the fork..." <- xorg is in maintainance mode since long time while xlibre is actively developed 2025-12-25 09:27:03 Not all development is good development 2025-12-25 10:08:48 especially if the development is done by people whose judgement cannot be trusted 2025-12-25 12:31:08 https://github.com/apache/pulsar-client-go/issues/274 2025-12-25 12:31:48 i tried to build payload-dumper-go but it fails cause of >This is due to the library github.com/valyala/gozstd that provides pre-compiled binaries for linux but these won't work on Alpine. 2025-12-25 12:32:31 ./go-cache/github.com/valyala/gozstd@v1.21.1/stream.go:14:48: undefined: DefaultCompressionLevel 2025-12-25 12:48:10 there is a rust version (it build slows but) no problems there 2025-12-25 16:57:38 @longnoserob what do replies look like? 2025-12-25 16:59:57 ichigo[m]1: look at https://irclogs.alpinelinux.org 2025-12-25 17:00:57 https://irclogs.alpinelinux.org/%23alpine-linux-2025-12.log 2025-12-25 17:04:08 "Whatever man. I'll do the..." <- as far as i can understand, someone wanted to contribute to xorg, but they didnt exactly like xorg changing and felt it might make it unstable, so he made a fork called xlibre in freedesktop git, same as xorg, but they deleted his entire account and repos all of them, shortly after without warning. so he moved to github and stuff, but controversy is mostly just his opinion about big tech or 2025-12-25 17:04:08 personally, i dont see such a big deal, its just a fork, use it or leave it lol. i think its nice to have more software and its free and open source. and if xorg does die out at least an alternative exists, to have options, and people can still use wayland or original xorg if they want 2025-12-25 17:04:08 wtv idk, he has a thing on his github called HISTORY.md 2025-12-25 17:04:25 f_: oh not too bad 2025-12-25 17:05:26 meh it's not great either 2025-12-25 17:05:36 if u wanna go into it more, brodie robertson made a decent video on it 2025-12-25 17:06:12 f_: yeh lol but at least it truncates it 2025-12-25 17:06:24 ichigo[m]1: you miss the part where the guy does not know what he's doing and also is an actual nazi 2025-12-25 17:06:54 he broke a lot of things in xorg and him also being nazi, these two things led to him getting kicked out of freedesktop gitlab 2025-12-25 17:07:28 Ok, I think I get it now, thanks :D 2025-12-25 17:07:39 xlibre is so reactionary they removed xwayland too so you as a distro maintainer would still have to maintain xorg support even if you want to switch to xlibre (which is already a bad idea) 2025-12-25 17:07:59 xlibre has gotten several commits from that guy that .. frankly show that he doesn't know what he's doing 2025-12-25 17:08:22 e.g. he literally thought "2^32" in C meant "pow(2, 32)" .. this is such a basic C beginner mistake 2025-12-25 17:08:27 and can be CVE-worthy 2025-12-25 17:08:40 idk the guy but i just look at the code, theres multiple contributors, but isnt xwayland for wayland? 2025-12-25 17:08:49 f_: beside the maintainer technical failures. You don't really need much to be banned from freedesktop. (that is not an apology of nazism) 2025-12-25 17:09:22 SyntheticBird[m]: I understand why he got banned from fdo and that is why I don't recommend xlibre 2025-12-25 17:11:15 xwayland is for supporting x11 apps in wayland yes 2025-12-25 17:11:22 yes and it's part of xorg 2025-12-25 17:12:06 https://gitlab.alpinelinux.org/alpine/aports/-/merge_requests/86092#note_520485 2025-12-25 17:14:38 yeah i didnt know about or see any posts from him indicating he was into nazi stuff, although i still generally separate the project from the person, especially if theres other contributors to not throw them under the bus, but i can understand now why there is any drama at all 2025-12-25 17:14:56 you think if he wasnt a nazi or anything people would like it more? 2025-12-25 17:15:14 ichigo[m]1: There has always been political drama around freedesktop. Mainly because both parties are acting like fucking stereotypes 2025-12-25 17:15:40 ichigo[m]1: I certainly think some retards would dislike him for not being a nazi 2025-12-25 17:15:47 ah 2025-12-25 17:15:59 SyntheticBird[m]: loll 2025-12-25 17:16:56 ichigo[m]1: even 'separating' the project from the person still gives more reasons against xlibre than for xlibre 2025-12-25 17:17:34 like I said, he does not know what he's doing, all he did so far is implement pointless namespacing extensions that don't contribute to security (and it's not hard to see why) 2025-12-25 17:18:00 just, no. 2025-12-25 17:49:52 I'm still on 3.22 I hope that's okay, I'll be upgrading soon; I feel like I'm a dumb person today, I'm looking at udhcpc --help and saw the following line "Run PROG at DHCP events (default /usr/share/udhcpc/default.script)", I looked at the script and it seems to indicate that if an executable exists at /etc/udhcpc/post-renew, it will be executed after renewal, however running 2025-12-25 17:49:54 udhdpc -i eth1, does not cause it to be executed. So what does "Run PROG at DHCP events" actually mean? 2025-12-25 19:29:41 Oops I was disconnected... 2025-12-25 20:11:12 bulldozer: my interpretation of that (expectations), it gets executed when dhcp events occur *after* udhcpc has been started (ie startup = DHCPDISCOVER + DHCPREQUEST), server gives params back including lease time. 2025-12-25 20:11:41 udhcpc then attempts a renew (DHCPREQUEST) at t + $lease_time, and it runs default.script 2025-12-25 20:13:47 it also handles failure events (DHCPNAK) based on what i see in the script 2025-12-25 20:16:02 invoked: I think I need to learn more about how DHCP works. After fiddleing with it for a while, I found that bound does what I thought renew did. I guess renew is like bound but only if the lease changes? 2025-12-25 20:16:36 Thank you. 2025-12-25 20:17:03 renewing is extending a lease that you already have. by default udhcpc will release when you shut down 2025-12-25 20:17:15 that's alpine's default anyway, not if you run it manually 2025-12-25 20:18:50 your dhcp server (wifi router, or whatever) will probably just give you the same ip each time, which is typical 2025-12-25 20:21:57 invoked: I'm only running it manually for testing. I have "use dhcp" in interfaces, my ISP has a DHCP server somewhere on the network. 2025-12-25 20:22:52 bulldozer: are you having a problem with dhcp? that's unclear to me 2025-12-25 20:23:41 invoked: I had trouble understanding the events mentioned in udhcpc. What I did was just use post-bound and it seems to do what I want. 2025-12-25 20:24:18 cool 2025-12-25 20:25:44 bulldozer: https://www.rfc-editor.org/rfc/rfc2131 in case you need it 2025-12-25 20:26:15 Thank you. 2025-12-26 02:06:33 hello 2025-12-26 02:07:06 I am tring to use openrc user to start Sway. But I got some errors 2025-12-26 02:07:46 FYI I have added elogind and polkit to the system rc 2025-12-26 02:08:35 When i enter the system and login to with my user and run sway everything is good 2025-12-26 02:09:40 But for some resson after i did in the wiki and added sway to rc with `rc-update -U add sway default` 2025-12-26 02:13:49 I get this error once in beginning of the logs: 2025-12-26 02:14:42 [wlr] [backend/session/session.c:256] Failed to load session backend 2025-12-26 02:14:42 [wlr] [backend/backend.c:399] Failed to start a DRM session 2025-12-26 02:14:42 [wlr] [libseat] [libseat/libseat.c:79] No backend was able to open a seat 2025-12-26 02:14:42 [wlr] [backend/backend.c:79] Failed to start a session 2025-12-26 02:14:42 [sway/server.c:247] Unable to create backend 2025-12-26 02:15:34 Then after this 2025-12-26 02:17:12 I get this one 16 times: 2025-12-26 02:19:16 UHH man i have pasted the wrongly 2025-12-26 02:19:45 THis 2025-12-26 02:20:08 Sorry i am not sure how to use irc 2025-12-26 02:20:34 the error i am getting is this: 2025-12-26 02:20:39 [wlr] [libseat] [libseat/libseat.c:79] No backend was able to open a seat 2025-12-26 02:20:39 [wlr] [backend/session/session.c:83] Unable to create seat: Function not implemented 2025-12-26 02:20:42 [wlr] [backend/session/session.c:256] Failed to load session backend 2025-12-26 02:20:44 [wlr] [backend/backend.c:79] Failed to start a session 2025-12-26 02:20:47 [wlr] [backend/backend.c:399] Failed to start a DRM session 2025-12-26 02:21:22 While i have added elogind to the system rc 2025-12-26 02:21:36 but it didn't find any seat for some resson! 2025-12-26 02:22:09 if i login then with the tty i can start sway 2025-12-26 02:23:13 is this on po? 2025-12-26 02:23:25 i do not have any script. I haven't made any script that played with elogind/sway/seat before 2025-12-26 02:29:00 pc i ment 2025-12-26 02:33:10 https://wiki.alpinelinux.org/wiki/Sway#No_backend_was_able_to_open_a_seat 2025-12-26 02:49:57 I have elogind 2025-12-26 02:50:04 loginctl SESSION UID USER SEAT TTY STATE IDLE SINCE c1 10000 road seat0 tty1 active no - c2 10000 road - - active no - 2 sessions listed. 2025-12-26 02:50:30 Ugh sorry 2025-12-26 02:52:06 it would work only after i login to the tty and run away, But within the openrc user experimental feature it wouldn't recognize a seat. 2025-12-26 02:52:56 https://wiki.alpinelinux.org/wiki/OpenRC#Runlevels_for_user_services 2025-12-26 02:53:00 here 2025-12-26 02:53:27 let me test seatd 2025-12-26 04:50:19 Quantum_3[m]: Dec 24 00:09:46 aix daemon.err bluetoothd[2517]: src/main.c:main() Unable to get on D-Bus 2025-12-26 04:50:37 sorry for the delay - xmas was mabye not the bes ttime to start asking questions lol 2025-12-26 05:31:08 f_: you could use wayland i guess 2025-12-26 05:31:16 honestly been interested in a new wayland window manager called river 2025-12-26 05:31:45 river is new? 2025-12-26 05:32:02 to me at least lol 2025-12-26 05:32:37 yeah seems the oldest commit on github is like a year 2025-12-26 05:34:56 there was a repo change https://codeberg.org/river/river-classic 2025-12-26 05:35:19 https://codeberg.org/river/river-classic/commit/6646826386a90e629340f10e677e4f2bd3200a8b 2025-12-26 11:40:05 cousin_luigi: for unicast, mumudvb provides the streams via HTTP, http://server:port/bysid/123 (service id) 2025-12-26 12:27:39 Hello everyone, I don't have a GitHub account, but I would like to inform the TCL/TK maintainer 'Celeste' that the package has been compiled without zipfs support. For verification, you can use the log of build or following command: zipfs mount //zipfs:/app, which will return a path upon success (https://wiki.tcl-lang.org/page/Single+file+applications+in+Tcl+9 ). 2025-12-26 12:28:26 Thank you very much, and enjoy your lunch! 2025-12-26 12:28:40 bye :) 2025-12-26 17:16:34 any advice to diagnose? Dec 26 12:11:40 cnx daemon.err loadkmap[1228]: ERROR: loadkmap failed to stop 2025-12-26 17:51:53 it's expected 2025-12-26 17:52:34 see /etc/init.d/loadkmap 2025-12-26 18:41:28 well, "expected"; it should be returning 0 explicitly 2025-12-26 18:41:35 and then that won't happen 2025-12-26 18:41:50 yes 2025-12-26 20:33:24 dwfreed: should I be filing a bug report for loadkmap? 2025-12-26 20:42:45 Yeah, or a PR, since it's a 2 character fix 2025-12-26 21:09:02 found a file in /tmp named 1301226c-c5d2-43d6-befa-82190f6e97dd.zip which was created a couple of days ago. It is an archive containing lots of PEM files each named with a UUID. I have no idea where it came from. It's possible it is just an artifact from some background job that didn't get cleaned up. I looked at cronjobs, but the only one I see that might have some connection is one 2025-12-26 21:09:02 called "update-dns-hints" -- but that script does not indicate zip files. 2025-12-26 21:09:42 I only see this on one of my 3 alpinelinux systems, all of which are at 3.23.2 2025-12-26 21:09:45 ownership 2025-12-26 21:09:59 the primary user (not root or special user) 2025-12-26 21:10:25 timestamp match with when you were around? 2025-12-26 21:11:02 possibly, idr now. I've been up at around that time, yes. It's possible I made it, but I don't remember doing it! 2025-12-26 21:11:15 checked logs? 2025-12-26 21:11:38 yes, no mention of it, including cron.log (and I perused the /etc/cron* area files also) 2025-12-26 21:12:26 the file is about 4MB. And the system is a testing system, not one I use for any real work. 2025-12-26 21:12:42 you could look at the cert files to see what they are 2025-12-26 21:12:57 I did and they are ordinary PEM files. 2025-12-26 21:13:10 with openssl? 2025-12-26 21:13:54 well, PEM files don't usually identify their exact source, but the headers seem to be, yes. I have openssl installed there. 2025-12-26 21:14:19 oh, maybe a tool to inspect the contents? 2025-12-26 21:14:32 (I don't do this sstuff too often, sorry) 2025-12-26 21:15:02 openssl x509 -in -noout -text 2025-12-26 21:15:04 sometihng like that 2025-12-26 21:15:16 ok, thanks, give me a few minutes 2025-12-26 21:17:41 https://dpaste.com/2QQXB36FX.txt 2025-12-26 21:18:18 eMudhra Technologies Limited? 2025-12-26 21:19:46 https://dpaste.com/6EPVMB695.txt (another one from the archive) 2025-12-26 21:28:57 any ideas from these 2 examples? I can upload the whole file somewhere if you want to see the contents. (if that would be useful) 2025-12-26 21:29:09 or tell me how you think I might proceed 2025-12-26 21:29:36 nope 2025-12-26 21:30:10 see if they've been put into 'the system', and if they're already on your other hosts (ie. come with alpine) 2025-12-26 21:30:35 under /etc/ssh or the like? /etc/pki? 2025-12-26 21:30:52 again, sorry for my feebleness with encryption stuff 2025-12-26 21:33:40 /etc/ca-certificates.conf has a list. I'll find it. 2025-12-26 21:34:02 indeed. `man update-ca-certificates` would be a start 2025-12-26 21:49:00 Your questions led me to thinking... I have been trying various alternatives to gnucash over the past few days on that system. I installed and uninstalled a few, including firefly III and frappe-books. firefly is web-based and frappe-books is a flatpak. 2025-12-26 21:49:25 Maybe either one of those dragged in a volume of certificates. 2025-12-26 21:49:54 but, at any rate, thank you for the assist. 2025-12-26 21:50:21 I am probably needlessly concerned about this anyway. It is just a bit strange, that's all. 2025-12-26 21:51:09 I have found that 3rd-party tools often have various issues. 2025-12-27 04:15:18 hello,... (full message at ) 2025-12-27 05:29:38 is the user stuff run with the same user creds and a dbus session? 2025-12-27 07:23:50 Not sure how to gather this information! 2025-12-27 07:32:39 this is the openrc command to start the user openrc stuff with some logs:... (full message at ) 2025-12-27 07:34:46 when i have login it opened a tty as sway didn't start, if i tried then to start it manually it would run normally for some resson 2025-12-27 07:35:09 i am not sure why there is a road with uid 0 and 10000 2025-12-27 07:35:17 here: 2025-12-27 07:35:17 session opened for user road(uid=10000) by road(uid=0) 2025-12-27 09:53:14 Have you set LIBSEAT_BACKEND to an incorrect value by any chance? 2025-12-27 10:34:46 "hello,..." <- Check if exist `/dev/dri/*` 2025-12-27 11:56:55 Anyone have any expense with Broadcom-wl 2025-12-27 11:57:06 Experience* 2025-12-27 11:59:41 It’s an abuild but having issues compiling 2025-12-27 12:00:04 Wanted to see if others are experiencing the same 2025-12-27 12:35:07 "Check if exist `/dev/dri/*`" <- ls /dev/dri/ 2025-12-27 12:35:07 by-path card0 renderD128 2025-12-27 12:35:27 "Have you set LIBSEAT_BACKEND..." <- I didn't add or change anything 2025-12-27 12:39:07 >there was a repo change https://codeberg.org/river/river-classic 2025-12-27 12:39:07 oh so not too new 2025-12-27 12:39:24 surprised i didnt hear about it or zig until recently 2025-12-27 12:39:39 一護 no luck with 12to11 right? 2025-12-27 12:39:52 unfortunately no :( i tried all i could 2025-12-27 12:40:06 might try on debian to see if its alpine specific 2025-12-27 12:40:07 i tried to register on their git 2025-12-27 12:40:45 might try again later but it uses build tools a bit unfamiliar to me 2025-12-27 16:36:36 "ls /dev/dri/..." <- Your user is on seat group? 2025-12-27 18:20:22 "Your user is on seat group?" <- Yes 2025-12-27 18:21:43 I could run sway within the tty but before tty with openrc i can't 2025-12-27 18:23:12 visone[m]: May I ask about the origin of your nick? 2025-12-27 18:34:15 hey guys, I am looking to build an app. My goal is to let the end-user ssh into a read only file, so they can verify that what runs on the server is the exact as open source code. LLMs tell me apine-linux has smaller attack surface, so I am thinking of going with it. Do you guys have any suggestions? 2025-12-27 19:00:19 cousin_luigi: I'm a selektah, my friends call me `vi`, vi-s(selektah)-one` > 'visone` 2025-12-27 19:20:54 * I'm a selektah, my friends call me `vi`, `vi-s(selektah)-one` > `visone` 2025-12-27 19:26:35 visone[m]: I see. It caught my eye because it's the Italian for "mink". Now you know too:) 2025-12-27 19:38:39 cousin_luigi: xddd I didn't know that!! xdddd 2025-12-27 22:47:56 Looks like no Broadcom-wl on 3.23.2 2025-12-27 22:48:11 It’s broken 2025-12-27 22:48:51 Can’t get abuild to take 2025-12-28 01:27:29 hi 2025-12-28 01:27:49 Case_Of: Yo 2025-12-28 01:28:00 I do have a popup window with a wrong-way sign about xfce-polkit, dunno how to get more log for this issue 2025-12-28 01:31:50 change to console and try logread -b 0 -n0 ? or check /var/log/Xorg* 2025-12-28 01:33:01 Case_Of: Most of the time this can happen because of a lack of PAM support (not using a Display Manager nor the shadow-login package) or if you haven't set up a seat manager and polkit, which I believe the setup-desktop script does not. 2025-12-28 01:33:19 Have you done both of those things? 2025-12-28 01:34:15 s/and/with/ 2025-12-28 01:34:54 I do start Xorg with my bashrc 2025-12-28 01:35:02 so no I’m not using a display manager 2025-12-28 01:36:28 Case_Of: Polkit needs PAM to work right, and the default login doesn't support this, so try installing the shadow-login package and restaring your system if you haven't done that. 2025-12-28 01:36:49 * default login utility doesn't support 2025-12-28 01:37:02 * Polkit needs PAM to work right, and the default login utility doesn't support this, so try installing the shadow-login package and restarting your system if you haven't done that. 2025-12-28 01:38:40 You, again, will only get this working if you're using polkit with elogind 2025-12-28 01:38:53 ok installing shadow-login did fix it, did not know this package existence until now 2025-12-28 01:39:37 Case_Of: Cool :D, util-linux's login can also work too btw 2025-12-28 01:42:03 they’re conflicting, ok. only one at a time 2025-12-28 01:43:04 and i see in the mean time that util-linux is splitted in many packages too 2025-12-28 01:44:31 Yeah, that way, if some busybox utility's lacking an option you need, you can install the full version of that specific utility without having to install the entirety of util-linux 2025-12-28 01:44:32 and weirdly enough, none are installed 2025-12-28 01:46:15 Case_Of: Uhh, ofc they aren't, Alpine uses busybox, with replaces both coreutils and util-linux with smaller versions of their utilities. 2025-12-28 01:46:34 How's that weird at all? 2025-12-28 01:47:28 *which 2025-12-28 01:47:44 ah busybox does have similar tools 2025-12-28 01:48:18 who owns lsblk? i do not have it 2025-12-28 01:48:53 ah there’s a package with that name 2025-12-28 01:49:21 nice 2025-12-28 01:59:24 you can find things like that through https://pkgs.alpinelinux.org/contents 2025-12-28 02:09:07 thanks 2025-12-28 02:14:14 So with most distros I have been using in the past I have had the btrfs, or zfs filesystem. Both of those had methods to take snapshots of the OS prior, and after updating the operating system. Is there an equivalent for Alpine Linux? 2025-12-28 02:17:49 LimitlessGrit[m]: I believe commit hooks could be used 2025-12-28 02:17:51 https://github.com/alpinelinux/apk-tools/blob/master/doc/apk.8.scd#configuration-files-relative-to---root 2025-12-28 02:19:28 This is a good solution for ensuring that packages update correctly. 2025-12-28 02:20:11 What if I wanted to keep entire system snapshots for 2-3 updates? 2025-12-28 02:49:49 You can just have an A/B/C upgrade logic in the script? 2025-12-28 02:58:04 I'm more thinking something I can transfer to an external hard drive, or be able to roll back to upon boot. I may just be longing for something that doesn't exist for Alpine though. 2025-12-28 11:23:30 It does exist when you script it yourself but it is unlikely that someone else has done exactly what you want. 2025-12-28 12:00:59 "I'm more thinking something I..." <- If you're still using btrfs, you can make a snapshot and transfer it to a file using `btrfs send`, encrypt the file and send it where ever you want. 2025-12-28 12:00:59 If you want to restore back, just unencrypt it and use `btrfs receive` to set your new system ... 2025-12-28 12:02:36 * If you're still using btrfs, you can make a snapshot and transfer it to a file using `btrfs send`, encrypt the file and send it where ever you want. 2025-12-28 12:02:36 For the roll back upon boot , check snapper 2025-12-28 12:02:36 If you want to restore back, just unencrypt it and use `btrfs receive` to set your new system ... 2025-12-28 22:02:22 adrian: jn: are you okay with me linking the aarch64_be and ppc64 ports in https://codeberg.org/Alpine-Ports/.profile/src/branch/main/ports.md ? 2025-12-28 22:03:48 cool page! for aarch64_be: yes 2025-12-28 22:04:50 okay, also, if you want to be added to the org, feel free to ask in #alpine-ports 2025-12-28 22:12:56 or ask here, but i don't wanna spam the official channels too much :D 2025-12-28 22:50:03 socksinspace: sure :) 2025-12-29 02:30:06 Can alpine linux use alternativs like btrfs or zfs? 2025-12-29 02:30:19 s/alternativs/alternatives/ 2025-12-29 02:38:44 btrfs yes https://wiki.alpinelinux.org/wiki/Btrfs 2025-12-29 02:39:26 zfs yes https://wiki.alpinelinux.org/wiki/ZFS 2025-12-29 02:41:25 Oh fuck yeah! 2025-12-29 02:41:59 Now the next question is do I need to do this on a fresh installation, or can a pre-existing alpine installation be converted to btrfs gracefully? 2025-12-29 03:01:48 read the wiki? 2025-12-29 03:01:51 ;) 2025-12-29 03:18:22 "read the wiki?" <- Ahhh 2025-12-29 03:18:23 lol 2025-12-29 03:18:46 Well It is good for a future reference then. I am not up for going through the gauntlet of a reinstall right now. 2025-12-29 03:19:09 Thank you kindly for the information. 2025-12-29 03:27:16 yw 2025-12-29 06:04:52 longnoserob[m]: If you want to use standard partition layout, just run `export ROOTFS=btrfs` before `setup-alpine`. 2025-12-29 06:06:24 lindsay: thank you, but that should go Limitless Grit 2025-12-29 06:08:13 If you want a more hacky approach, I wrote a blog (in Chinese, hope translator works on it): https://lin.moe/tutorial/2024/07/alpine-efistub-btrfs/ 2025-12-29 14:25:52 WhyNotHugo: is the unbound config for using alpine as a server for resolving .local domains or using alpine as a desktop machine using unbound as a local resolver 2025-12-29 14:29:07 ACTION once set up a router with alpine and the throughput was atrocious. 2025-12-29 14:29:19 Do you think it's a kernel thing? 2025-12-29 14:30:22 Could be deffective hardware 2025-12-29 14:30:28 Broken driver 2025-12-29 14:50:24 quinq: It works perfectly with devuan and even better with openwrt. 2025-12-29 14:50:30 Same hardware. 2025-12-29 15:20:08 Hi All. 2025-12-29 17:05:35 cousin_luigi: openwrt is musl, and the latest openwrt is running kernel 6.6 so your experience is... interesting 2025-12-29 17:06:05 invoked: Indeed it is. I assumed it to be some special kernel optimisation. 2025-12-29 17:59:40 gonna use openwrt as my main desktop os challenge for 69 days 2025-12-29 18:01:18 hell yeah 2025-12-29 19:36:26 even though musl nscd says its only for passwd/group info, would nss mdns still work with it 2025-12-29 19:36:49 i realized the mdns avahi2dns thing doesnt seem to work as it being a server and another alpine machine being a client 2025-12-29 20:04:35 penguinz_rule[m]: for using a desktop setup which resolves .local through unbound. 2025-12-29 20:05:05 it likely works for a local server (if unbound is exposed), not sure about the security implications 2025-12-29 20:07:48 i probably have to futz it more to get it to work as a server 2025-12-29 20:08:33 anyways，in an ideal world all id need is avahi, and i think windows wouldnt need the dns server to access the .local addrs 2025-12-29 23:55:39 Hello everyone! I was looking through gitlab releases, and seems like there is issue with link for `armhf` for apk-tools 2025-12-29 23:55:41 https://gitlab.alpinelinux.org/alpine/apk-tools/-/releases 2025-12-29 23:56:22 apk.static for armv7 works correctly, but for armhf it returns 404 2025-12-29 23:57:54 I scrolled down, and seems like link for v2.14.10 is working or v3.0.0_rc4 . Everything after doesn't work unfortunatelly. 2025-12-30 09:34:39 anyone know of a kanban web app in apk? 2025-12-30 17:52:58 why i can't use lbu commit without usb or floppy? 2025-12-30 17:55:22 I am not sure if i understand lbu correctly, Is it like if the my system ware in a repo, And i could save reversions of my system 2025-12-30 18:18:10 Hi, community/croc: upgrade to 10.3.1 <- Problem with my MR (1 month) ? : https://gitlab.alpinelinux.org/alpine/aports/-/merge_requests/93895 2025-12-30 19:06:50 Hi, what is the way of getting a newer version of a package like mise (apk info mise)? Do we need a pull request? 2025-12-30 19:12:58 l1x: easiest way is to make a merge request to aports or contact the maintainer. are you familiar with abuild tooling? 2025-12-30 19:27:26 elagost: i am LLM familiar with abuild. Do i need to fork aports or I push directly into the central repo (with my branch)? 2025-12-30 19:28:58 i see people are using forks 2025-12-30 19:29:05 thanks, i am submitting the PR 2025-12-30 21:07:52 https://gitlab.alpinelinux.org/l1x/aports/-/jobs/2155099 do you usually solve this by export MAKEFLAGS="-j4"? 2025-12-30 22:16:12 ERROR: mise: build failed 2025-12-31 01:15:09 l1x: It's a relatively common workaround for such issues 2025-12-31 14:43:16 Anyone else where KeepassXC autotype no longer works on X11? 2025-12-31 14:54:22 ikke: i don't use KeepassXC, but i found this: https://github.com/keepassxreboot/keepassxc/issues/12723 2025-12-31 14:54:49 I just saw that as well, but for me it's consistently not working 2025-12-31 14:56:16 hmmm, i wasn't very thorough reading through the comments on that issue, but it seemed that it was consistently broken for some, inconsistent for others 2025-12-31 15:01:22 Right, so waiting for that to be fixed 2025-12-31 15:04:13 there is a fix pending, so hopefully gets merged soon 2025-12-31 17:09:48 hi, is there a way for apk to list packages that are installed from a community repository? 2025-12-31 17:39:25 Hi, I allways Use "apk list -I". But list have a lot of options, try apk list --help 2025-12-31 17:41:42 apk itself does not track what url a package was installed from, though apk policy can show what index provides a package 2025-12-31 20:45:42 Hello! How can I disable ipv6 entirely in alpine? 2025-12-31 21:00:29 "Hello! How can I disable ipv6..." <- Add `ipv6.disable=1` to the kernel cmd line 2025-12-31 21:06:31 "Add `ipv6.disable=1` to the..." <- Thanks for the reply! But I still get an ip using `ip addr show | grep inet6` 2025-12-31 21:07:03 do all of those addresses start with fe80? 2025-12-31 21:07:22 yes! 2025-12-31 21:07:34 those are link-local methinks 2025-12-31 21:07:54 those are link-local, so nothing would attempt to connect to the general internet with them 2025-12-31 21:08:16 I just saw that I have this as well inet6 ::1/128 scope host 2025-12-31 21:08:39 ::1 is localhost 2025-12-31 21:09:36 Ah okay, thanks for the headsup. 2025-12-31 21:19:12 chatterino[m]: check `cat /proc/cmdline` to confirm that the option is actually set 2025-12-31 21:20:42 ikke: Hmm nothing in there. 2025-12-31 21:22:04 How did you attempt to set the option? 2025-12-31 21:22:11 I have it in there. 2025-12-31 21:22:11 cat /etc/sysctl.d/local.conf 2025-12-31 21:22:11 ``` 2025-12-31 21:22:11 ipv6.disable=1 2025-12-31 21:22:11 ``` 2025-12-31 21:24:44 For them to apply in sysctl you'd have to use: 2025-12-31 21:24:44 ``` 2025-12-31 21:24:44 ``` 2025-12-31 21:24:44 net.ipv6.conf.default.disable_ipv6 = 1 2025-12-31 21:24:44 net.ipv6.conf.all.disable_ipv6 = 1 2025-12-31 21:24:46 Then `sysctl -p /etc/sysctl.d/local.conf`. Ensure the `sysctl` runs on boot to apply this option at boot (`rc-update add sysctl`, though this should already be setup) 2025-12-31 21:27:58 Cool! Using ip addr show | grep inet6 shows nothing now. However cat /proc/cmdline still prints nothing. 2025-12-31 21:28:17 It would only show something if you actually provide it as a kernel argument 2025-12-31 21:28:20 The kernel commandline and sysctl are two different ways that you can use to disable ipv6. 2025-12-31 21:28:21 not a sysctl setting 2025-12-31 21:29:49 Ok got it. How would I be able to disable ipv6 per NIC? 2025-12-31 21:30:22 I had multiple ipv6 due some docker containers and I actually might need it only for one. 2025-12-31 21:30:30 net.ipv6.conf..disable_ipv6 = 1 2025-12-31 21:30:43 Something like this for sysctl: 2025-12-31 21:30:43 ``` 2025-12-31 21:30:43 ``` 2025-12-31 21:30:43 net.ipv6.conf.eth0.disable_ipv6 = 1 2025-12-31 21:30:44 Ha okay, thought about it at first. 2025-12-31 21:45:18 Thanks for your help.