2014-01-01 00:32:40 Mp5shooter: wildcard? 2014-01-01 00:32:46 domain.com and *.domain.com? 2014-01-01 00:33:03 can you PM me the url, where you have installed it?:) 2014-01-01 00:33:25 HAPPY NEW YEAR 2014-01-01 00:33:42 33,60 Danish kroner = 4,50911218 Euros ?? 2014-01-01 00:33:46 why 7,50EUR? 2014-01-01 00:52:22 shafire: mp5.im 2014-01-01 00:52:28 and yep shafire 2014-01-01 00:52:31 domain.com and *.domain.com 2014-01-01 00:52:31 :D 2014-01-01 00:52:45 it's great 2014-01-01 00:52:52 cheaper than startssl 2014-01-01 00:52:53 and yep 2014-01-01 00:52:55 it was like 2014-01-01 00:52:58 $7 USD 2014-01-01 00:53:01 with some change 2014-01-01 00:53:32 okay 2014-01-01 00:53:34 cool 2014-01-01 00:53:55 nice timing 2014-01-01 00:54:09 yeah haha 2014-01-01 00:54:10 If I get them today, it's from 01.01.14 - 01.01.19 2014-01-01 00:54:15 :D 2014-01-01 00:54:47 can't believe it's that cheap 2014-01-01 00:55:06 why not 25 years? :D 2014-01-01 00:55:12 haha yeah 2014-01-01 00:55:21 wish they had more than 5 :P 2014-01-01 00:55:44 I will change the html form 2014-01-01 00:55:47 alphasslwild-5 to alphasslwild-25 2014-01-01 00:55:50 maybe, it works, haha 2014-01-01 00:55:52 haha 2014-01-01 00:56:08 it'll outlive their root certificate 2014-01-01 00:56:09 :P 2014-01-01 00:56:55 what have you filled for vat? 2014-01-01 00:57:01 --? 2014-01-01 00:57:02 000000000 2014-01-01 00:57:02 lol 2014-01-01 00:57:07 Ok 2014-01-01 00:58:24 and I used the hostmaster@ email address for the domain 2014-01-01 00:58:27 so it's automated 2014-01-01 00:59:33 argh, ok 2014-01-01 00:59:48 no email on the two wish domains :D 2014-01-01 01:00:04 :P 2014-01-01 01:00:05 but isn't startssl good too? 2014-01-01 01:00:10 yeah 2014-01-01 01:00:12 but you have to pay 2014-01-01 01:00:16 for wildcard and stuff 2014-01-01 01:00:22 yeah 50$ 2014-01-01 01:00:28 but unlimited certs + ucc 2014-01-01 01:00:41 o nice 2014-01-01 01:01:43 I have domain.net and domain.com, want to secure both, only with ucc possible :( 2014-01-01 01:01:51 :( 2014-01-01 01:05:17 have you created a 2048bit rsa csr? 2014-01-01 01:05:35 or should I take 4096 because of 5 years, lol 2014-01-01 01:08:04 I did 4096 2014-01-01 01:08:06 :P 2014-01-01 01:11:56 yeah 2014-01-01 01:12:01 in 5 years, it's the lowest 2014-01-01 01:13:32 mh, I could also take EC instead of RSA/DSA 2014-01-01 07:25:46 we don't have compiled lua-sql-odbc right? 2014-01-01 07:25:51 http://www.keplerproject.org/luasql/2.0/manual.html#drivers 2014-01-01 07:26:01 just a confirm.. 2014-01-01 07:43:39 ncopa, lua sql libraries from keplerproject are not subpackages but separate packages. There's a motivation? 2014-01-01 10:33:09 Where is the best place to start with a new Alpine install? 2014-01-01 11:04:08 http://alpinelinux.org/forum/installation/virtualbox-installation-reboot-kernel-crash 2014-01-01 11:08:16 moinmoin and a happy new year to all 2014-01-01 14:03:18 happy new year to all :) 2014-01-01 14:46:50 Mp5shooter: If I create a CSR for a certificate with 5 years and the key for the CSR is only 2 years valid, will this blow up? 2014-01-01 14:55:47 is there a possibility to reference svn repo as a souce on APKBUILD? 2014-01-01 14:55:56 like 2014-01-01 14:56:11 source="http://foo.bar.org/svn/trunk" 2014-01-01 14:56:13 ? 2014-01-01 14:59:19 OK Mp5shooter bought 2 certs for 6 eur each or so :D 2014-01-01 14:59:22 lol :) 2014-01-01 15:30:43 shafire: a key does not have an expiry iirc 2014-01-01 15:30:51 ok 2014-01-01 15:30:57 Did not know that, thank you 2014-01-01 15:32:20 No email so far :( 2014-01-01 15:32:24 with the cert 2014-01-01 15:32:32 used the hostmaster@ :S 2014-01-01 15:45:20 " Globalsign offer special promotion for Gold Partners since last week. It will be valid till Jan 15th 2014." 2014-01-01 15:45:25 OK 2014-01-01 16:45:07 nice shafire :P 2014-01-01 16:45:23 no cert so far :( 2014-01-01 16:45:31 o 2014-01-01 16:45:49 they didn't get rejected or anything did they 2014-01-01 16:46:06 i forgot to add hostmaster@ to my list and i had to contact globalsign to resend the emails 2014-01-01 16:46:45 first email i got was one where i had to click a link to approve the cert 2014-01-01 16:46:50 then another email came with my cert 2014-01-01 16:47:18 whats the mail from globalsign? 2014-01-01 16:47:38 one was approval@globalsign.com 2014-01-01 16:47:49 another no_reply@globalsign.com i think 2014-01-01 16:47:50 how did you contact them? 2014-01-01 16:47:58 globalsign.com live chat 2014-01-01 16:49:55 they're very nice people 2014-01-01 16:49:55 :p 2014-01-01 17:39:10 Where do the kernel headers installs to? 2014-01-01 17:40:56 I am really having issues getting guest additions for vbox installed properly so I can actually use the system comfortably... 2014-01-01 17:44:08 freedomrun: do you know where the kernel headers get dropped down in Alpine? 2014-01-01 17:44:59 edge226, no sorry 2014-01-01 17:45:39 I am trying to run this in vbox and I want the guest additions to work, It wont build the kernel module due to not being able to find the kernel headers :/ 2014-01-01 18:15:15 edge226: linux-grsec-dev 2014-01-01 18:15:34 edge226: I tried to build guest additions some time ago, but it's not so easy 2014-01-01 18:15:46 barthalion: do I need the linux-grsec other stuff as well? 2014-01-01 18:17:27 barthalion: why the heck is it so difficult? 2014-01-01 18:17:28 well, linux-grsec is the default kerne… 2014-01-01 18:17:57 I've encountered problems with kmod for sure, and then with the code not compatible with POSIX 2014-01-01 18:18:10 but maybe it's easier now 2014-01-01 18:24:19 barthalion: there have been a lot of things that have been slightly off in alpine from my experience. 2014-01-01 18:29:43 barthalion: I am still fighting to get it to see the kernel headers are there... 2014-01-01 18:29:53 barthalion: do you know what I need to set for them to register? 2014-01-01 18:30:53 what makes you think that there is kind of register? 2014-01-01 18:31:09 apparently ncopa fixed the APKBUILD 2014-01-01 18:31:20 take a look at virtualbox-additions-grsec from testing 2014-01-01 18:31:51 barthalion: ah so its not in edge yet? 2014-01-01 18:32:02 it is, but in testing 2014-01-01 18:32:34 edge is "version" (rolling release in this case) 2014-01-01 18:32:39 main and testing are repositories 2014-01-01 18:32:58 you probably enabled only main in /etc/apk/repositories 2014-01-01 18:40:25 barthalion: so what repos do I want enabled? 2014-01-01 18:42:29 okay, I see. Enabled both :) 2014-01-01 18:48:50 drop us a line if it works – we will move it to main then 2014-01-01 18:49:18 doesnt seem to work. 2014-01-01 18:49:27 I dont see that file though. 2014-01-01 18:49:51 which file? 2014-01-01 18:51:05 the virtualbox-additions-grsec when I search virtualbox all I get is xf86-video-virtualbox which I installed. 2014-01-01 18:51:30 did you run apk update? 2014-01-01 18:51:50 barthalion: yes. 2014-01-01 18:52:41 hm, I'll check it in a moment 2014-01-01 18:54:47 and I ran apk upgrade -U --available 2014-01-01 18:58:36 # apk search virtualbox 2014-01-01 18:58:36 xf86-video-virtualbox-4.2.12-r0 2014-01-01 18:58:36 virtualbox-additions-grsec-3.12.6-r0 2014-01-01 18:58:53 can you paste /etc/apk/repositories somewhere? 2014-01-01 19:03:20 ah, those are host additions anyway 2014-01-01 19:08:09 hm 2014-01-01 19:08:20 is there a way to tell apk not to upgrade the kernel 2014-01-01 19:08:22 when you upgrade 2014-01-01 19:09:12 I don't think so 2014-01-01 19:13:57 https://dpaste.de/Pzzj 2014-01-01 19:15:33 barthalion: ^ 2014-01-01 19:16:15 maybe your mirror is out of date 2014-01-01 19:16:44 Ill check later. I need sleep now. 2014-01-01 19:17:08 good night :) 2014-01-02 08:00:21 morning 2014-01-02 08:00:28 the virtualbox drivers might be broken 2014-01-02 08:00:35 i built them and added them to testing 2014-01-02 08:00:39 but got no feedback 2014-01-02 08:01:01 i dont have virtualbox to test on atm 2014-01-02 08:03:13 ncopa, do you need alpine as guest or host? ... and happy new year :) 2014-01-02 08:04:02 hi, i was thinking of edge226's issues above 2014-01-02 08:05:49 ncopa: thanks for getting back to me, I would have no issue being a tester for you if thats what was needed. I can do a fresh vbox install and show what I do to get things up and running so they can be added to the distro. 2014-01-02 08:06:05 that woudl be good 2014-01-02 08:06:42 use 'edge', which is the rolling 'development' version 2014-01-02 08:07:14 ncopa: I have a linux site and a youtube channel so I can do some good feedback ;) 2014-01-02 08:07:37 I have edge enabled vboxed alpine linux on - if there is something that is needed (atm pulling updates) 2014-01-02 08:08:18 ncopa: there was a lot of changes that needed to be done for sudo and su to work properly under my shell of choice. Those things could be useful to have fixes for. 2014-01-02 08:08:34 which is your shell? 2014-01-02 08:08:38 fish 2014-01-02 08:09:01 do you think you could file issues on bugs.a.o for those? 2014-01-02 08:09:14 ncopa: I probably could. 2014-01-02 08:09:41 freedomrun: do you have xf86-video-virtualbox running too? 2014-01-02 08:11:18 i think there is a kernel module too 2014-01-02 08:11:26 ncopa, I have it installed that is for shure as I did pulled all vbox related packages to get proper resolution (didn`t help - no vbox additions) but need to check if it is running .. 2014-01-02 08:11:32 yep 2014-01-02 08:11:38 ncopa: how do you get the kernel headers to link up properly? 2014-01-02 08:11:55 edge226: apk add linux-grsec-dev 2014-01-02 08:12:11 is the development package for kernel modules 2014-01-02 08:12:29 edge226: its for kernel modules? 2014-01-02 08:13:36 linux-grsec-dev :( this is what I was missing ... grr 2014-01-02 08:13:49 ncopa: yeah on most distros you can just run the vbox install script for the virtualbox additions. That is what seems most ideal. 2014-01-02 08:14:31 wouldnt it be better to just: apk add virtualbox-additions? 2014-01-02 08:14:46 that would be expected :) 2014-01-02 08:15:07 and would shurely work with grsec kernel 2014-01-02 08:15:56 alpine repo=alpine security ;) 2014-01-02 08:16:01 the kernel module should be packaged in separate package 2014-01-02 08:16:55 it is, its virtualbox-additions-grsec 2014-01-02 08:16:58 definitely as not needed by every single install ... clever thinking ncopa 2014-01-02 08:17:12 and we have various different kernels 2014-01-02 08:17:24 or we want be able to support different kernels... 2014-01-02 08:18:05 ok update finished (got 1 error hmm) will check for vbox now 2014-01-02 08:18:10 rebooting 2014-01-02 08:18:40 system up (alpine rulez :)) 2014-01-02 08:22:58 ok apk fix upgrade -i seems eliminated 1 error 2014-01-02 08:26:23 ncopa, u said vbox-additions should be loaded as module or? 2014-01-02 08:28:44 normally there are 2 parts: kernel part and userpsace 2014-01-02 08:28:54 i think we currently only have the kernel part 2014-01-02 08:28:58 the kernel module 2014-01-02 08:29:17 i dont know if there are any userspace daemon or something that is needed too 2014-01-02 08:31:13 ncopa, you are reading my mind :) just did some searching about OpenRC(Gentoo) related to vbox-additions and found a bit old thread: https://www.virtualbox.org/ticket/9028 2014-01-02 08:31:30 yes, we have only host additions 2014-01-02 08:31:49 I tried to build guest additions some time ago, but they are real pain 2014-01-02 08:32:08 ah 2014-01-02 08:32:09 for example their build script expects that you use reasonable new kmod to load modules 2014-01-02 08:32:33 there are different for host and guest 2014-01-02 08:32:48 the host additions is kinda useless since we dont support virtualbox host 2014-01-02 08:33:08 and when I modified it to use what we have on Alpine by default, a wild compilation error occured anyway 2014-01-02 08:33:25 atm reading this: http://wiki.gentoo.org/wiki/VirtualBox 2014-01-02 08:35:21 hm 2014-01-02 08:49:20 yep no vbox-additions reported in dmesg, lsmod doesn`t say about anything related to vbox after /etc/conf.d/modules="... vboxdrv vboxnetadp vboxnetflt ..." modification ... maybe this isn`t correct for alpine 2014-01-02 08:50:29 modprobe vboxdrv results lsmod listing vboxdrv = 0 2014-01-02 08:54:17 Ill have to run another vm and record it and get the results that way. 2014-01-02 09:09:39 hmpf 2014-01-02 09:10:21 xf86-video-virtualbox will be tricky on musl 2014-01-02 10:46:29 ncopa: so I am going to work on that video series. I need to do some local space management prior to being able to do so. 2014-01-02 10:52:25 nice 2014-01-02 11:15:30 ncopa: I really think Alpine has some potential :) 2014-01-02 11:15:50 ncopa: I dont think it will ever be my main desktop box but its definitely got a purpose 2014-01-02 11:16:36 ncopa: 20 mins of searching and 120G freed :) 2014-01-02 11:29:59 ncopa: is it better to run 32 or 64 bit? Could it be that the package is only 32 atm? 2014-01-02 11:50:14 found interesting http://www.eclipse.org/mihini/ 2014-01-02 11:50:59 indeed 2014-01-02 11:52:17 that looks something i could have *really* used few years ago... 2014-01-02 12:02:23 edge226: depends on how much mem you have. use 64bit if you have more than 3-4GB ram 2014-01-02 12:04:19 vkrishn: what is M2M applications? 2014-01-02 12:04:30 machine 2 machine 2014-01-02 12:07:50 ncopa: its just in vm only. 2014-01-02 12:09:22 ncopa, it's mainly useful for building automation etc. 2014-01-02 12:09:32 but also other uses are possible 2014-01-02 12:11:13 edge226: if dont plan to assignt more than 3-4G of your vm guest 32bit should be fine, but i dont think 64bit adds any significant overhead either. 2014-01-02 12:11:27 i would expect performance be slightly better with 64bit 2014-01-02 12:11:54 ncopa: But in a matter of packages. are 64 and 32 nearly identical on this distro? 2014-01-02 12:12:02 yeah 2014-01-02 12:12:17 the only diff i am aware of is 2014-01-02 12:12:24 ncopa: kk all good then ill just stick with 64 then. 2014-01-02 12:12:27 xen only exist for 64bit 2014-01-02 12:12:33 wine only exist for 32bit 2014-01-02 12:12:42 other than that i think its identical 2014-01-02 12:12:54 ncopa: not things I am concerned with running it in a VM. 2014-01-02 12:13:07 exactly 2014-01-02 12:15:01 and Ill be honest, I've already found my preferred host system. I think the application of Alpine in a VM is really good though :) 2014-01-02 12:16:09 some rules http://wiki.eclipse.org/Paho/Paho_Websockets still in process of being standard 2014-01-02 12:17:53 and I honestly think better Vbox support could potentially increase alpine's popularity. 2014-01-02 12:20:17 ncopa: it might be a bit until I am @ the point of things being useful but I've gone through this a few times and I am going to Give it a series instead of a single review or something. 2014-01-02 15:54:14 busybox 1.22 is out 2014-01-02 15:56:31 but it's beta 2014-01-02 15:57:19 so we'll only run it in edge ;) 2014-01-02 17:05:17 ncopa, widechars still enabled or reverted back ? 2014-01-03 01:17:39 Hello! 2014-01-03 01:19:10 just installed edge-musl in a chroot 2014-01-03 01:21:03 and wrote an APKBUILD for something I didn't find in main or testing 2014-01-03 04:06:26 is there a tool to build an *.apk package from an APKBUILD? 2014-01-03 04:07:48 abuild 2014-01-03 04:07:53 because I can't find anything about doing that with abuild 2014-01-03 04:08:12 run abuild -r within the directory with the APKBUILD 2014-01-03 04:13:05 thanks 2014-01-03 05:42:29 Any suggestions for what to do when make DESTDIR=... install fails? 2014-01-03 05:44:34 Apparently, it successfully installs everything in $DESTDIR/, then tests if it was installed in /, and returns 2 2014-01-03 07:55:42 Ibidem: sounds like the makefile does not use DESTDIR 2014-01-03 07:55:48 i suppose it depends on the package 2014-01-03 08:24:11 he left anyway 2014-01-03 08:42:15 oh.. 2014-01-03 08:42:28 i got my new laptop 2014-01-03 08:42:58 i'm trying to upgrade windows to 8.1 on it so i can install dualboot 2014-01-03 08:43:11 i wonder how to actually do it with alpine linux 2014-01-03 08:43:25 and efi 2014-01-03 11:27:55 ncopa, I hope mbr change/ dualboot would not void the warranty ? 2014-01-03 13:33:40 kaniini: sha512sum: can't open 'init-xenstore-domain.patch': No such file or directory 2014-01-03 13:33:40 init-xenstore-domain.patch: FAILED 2014-01-03 13:36:02 ncopa, did you re-enable utf8 stuff on busybox that got reverted on upgrade? 2014-01-03 13:36:11 ncopa: i already pushed a fix for that 2014-01-03 13:36:48 ncopa: in fact, if you look at: http://git.alpinelinux.org/cgit/aports/commit/?id=4c825a2884ff2910d84d948141f320bd3fad3b6c 2014-01-03 13:36:55 ncopa: you are deleting the file :P 2014-01-03 13:38:07 kaniini: xenstored.initd: FAILED 2014-01-03 13:38:07 sha512sum: WARNING: 1 of 1 computed checksums did NOT match 2014-01-03 13:38:10 still fails 2014-01-03 13:38:13 okay 2014-01-03 13:38:26 but something is wrong with the x86_64 builder too i think 2014-01-03 13:38:38 fabled: the utf8 stuff got reverted? 2014-01-03 13:38:45 it was a mistake if so 2014-01-03 13:39:04 ncopa, yes 2014-01-03 13:39:13 the utf8 config change was reverted in ugprade 2014-01-03 13:39:37 bah.., ok 2014-01-03 13:39:48 hm 2014-01-03 13:40:06 i have generally low expectations for utf8 support in alpine < 3.x 2014-01-03 13:40:16 need figure out why build-edge-x86_64 chokes 2014-01-03 13:40:58 i suspect alpine 3.x will immediately provide superior utf8 support just as a result of it being properly supported in libc 2014-01-03 13:41:06 Your branch and 'origin/master' have diverged, 2014-01-03 13:41:07 and have 1 and 14 different commits each, respectively. 2014-01-03 13:41:07 (use "git pull" to merge the remote branch into yours) 2014-01-03 13:41:19 oh/ 2014-01-03 13:41:24 that is probably because 2014-01-03 13:41:35 i used git push --force to strip out the errant changeset 2014-01-03 13:42:06 and build server pulled before new update 2014-01-03 13:42:14 yep 2014-01-03 13:42:35 if you do git pull 2014-01-03 13:42:37 it should fix it 2014-01-03 13:42:39 i think 2014-01-03 13:42:49 or reset --hard HEAD^ 2014-01-03 13:42:51 yes 2014-01-03 13:43:03 i think i need to do it for x86 too then 2014-01-03 13:43:10 and bugs.a.o... 2014-01-03 13:43:12 ok 2014-01-03 13:43:12 sorry :( 2014-01-03 13:43:29 thats the price we pay for immediate builds.. 2014-01-03 13:43:29 maybe it should just 2014-01-03 13:43:42 reset itself forcefully 2014-01-03 13:43:44 each time 2014-01-03 13:43:56 nah, --force should not really be used 2014-01-03 13:44:06 weird 2014-01-03 13:44:07 unless exception 2014-01-03 13:44:12 i use it on github all the time 2014-01-03 13:44:18 to fix things i just committed 2014-01-03 13:44:24 but forgot a file or whatever 2014-01-03 13:44:29 it messes up for ppl who pulls 2014-01-03 13:44:39 its no problem if you do it before anyone/anything pulls 2014-01-03 13:45:39 in theory you should never rewrite history on public git repos 2014-01-03 13:45:49 (but thats theory) 2014-01-03 13:51:04 ha 2014-01-03 13:52:17 i'm glad i wasn't too hard on kaniini. i did similar mistake myself... 2014-01-03 13:52:20 :) 2014-01-03 13:54:45 bah 2014-01-03 13:54:52 busybox find bug 2014-01-03 14:41:02 kaniini: ^^^ 2014-01-03 14:41:06 console/xencons_ring.c:18:20: error: inlining failed in call to 'notify_daemon': call is unlikely and code size would grow [-Werror=inline] 2014-01-03 14:41:06 static inline void notify_daemon(struct consfront_dev *dev) 2014-01-03 14:41:06 ^ 2014-01-03 14:41:06 notify_daemon(dev); 2014-01-03 14:41:06 console/xencons_ring.c:186:15: error: called from here [-Werror=inline] 2014-01-03 14:42:51 ncopa: i'll look in a bit 2014-01-03 14:43:07 ncopa: however, -Werror should be eradicated across the entire APKBUILD 2014-01-03 14:43:28 ncopa: presently dealing with provisioning 2 x 100gbps links into my infinera dtn-x system 2014-01-03 14:43:58 ah true... -Werror should not be used in releases... 2014-01-03 15:07:20 >>> xen: Eradicating -Werror... 2014-01-03 15:07:22 find: invalid option -- n 2014-01-03 15:07:24 BusyBox v1.22.0 (2014-01-03 14:25:23 UTC) multi-call binary. 2014-01-03 15:07:26 this is why 2014-01-03 15:07:30 new busybox broke it 2014-01-03 15:07:44 bah 2014-01-03 15:08:39 maybe just add 'findutils' to makedepends for now 2014-01-03 15:08:44 okay. 2014-01-03 15:09:19 pushing now 2014-01-03 15:09:50 thanks 2014-01-03 15:10:10 what's -n supposed to do? 2014-01-03 15:10:29 no idea, ncopa wrote that code iirc 2014-01-03 15:11:01 ah 2014-01-03 15:11:17 i thought it was the makefile that had it 2014-01-03 15:12:17 % find -n 2014-01-03 15:12:18 find: unknown predicate `-n' 2014-01-03 15:12:24 find . -name 2014-01-03 15:12:27 ... 2014-01-03 15:12:29 ah 2014-01-03 15:12:34 it says find -name ... 2014-01-03 15:12:40 new busybox choes on that 2014-01-03 15:12:43 chokes 2014-01-03 15:12:52 i think they added -H|-L support 2014-01-03 15:13:10 and while doing so they require PATH argument 2014-01-03 15:13:14 as it is in posix 2014-01-03 15:13:33 yeah, I didn't know that PATH is not required… 2014-01-03 15:13:45 findutils find does not require it i think 2014-01-03 15:14:02 the bsd's has always required it afaik 2014-01-03 15:14:19 no idea why I thought so 2014-01-03 15:14:31 seems like busybox fdisk does GUID 2014-01-03 15:14:35 cool 2014-01-03 15:15:00 'using GPT' 2014-01-03 15:59:41 EFI certaily dont make it easier to boot/dualboot linux 2014-01-03 18:03:43 hi 2014-01-03 18:03:53 any news to an available musl iso image? 2014-01-03 18:41:26 shafire, x86 or x86_64? 2014-01-03 18:41:47 last 2014-01-03 18:41:49 x86_64 2014-01-03 18:47:36 completely untested 2014-01-03 18:47:39 but you can try http://dev.alpinelinux.org/~tteras/alpine-140103-x86_64.iso 2014-01-03 18:47:50 mkinitfs will be broke 2014-01-03 18:48:18 is x86 ready too? 2014-01-03 18:48:49 you can fix mkinitfs with http://sprunge.us/HcLD 2014-01-03 18:49:15 i just built the x86_64 iso for you 2014-01-03 18:50:10 you want x86 too? 2014-01-03 18:50:13 oh 2014-01-03 18:50:20 barthalion, so you want x86 too? 2014-01-03 18:50:30 I'm mainly x86 user, that's why I'm asking 2014-01-03 18:50:46 not big deal if you won't make it, I'll test x86_64 2014-01-03 18:50:56 well 2014-01-03 18:51:36 hold on for a sec 2014-01-03 18:52:57 here you go http://dev.alpinelinux.org/~tteras/alpine-140103-x86.iso 2014-01-03 18:53:04 thank you 2014-01-03 18:53:07 the amount of testing is equal.... that is: zero. 2014-01-03 18:53:33 I'll provide some feedback, I see that you know fix for mkinitfs already 2014-01-03 18:53:52 yes, the mkinitfs is workaroundable with the sprunge change 2014-01-03 18:54:00 I've hit it when I tried to install alpine-musl with apk-static 2014-01-03 18:54:06 yes. 2014-01-03 18:54:39 we plan improving the mkinitfs. but should probably commit that until the new stuff is ready. 2014-01-03 19:01:27 bah. seems that x86 needs at least libgcc_s.so.1 more 2014-01-03 19:09:06 barthalion, i re-uploaded newly generated x86 now. will test if boots in a minute or two 2014-01-03 19:17:25 barthalion, seems the x86 musl iso boots now (on my qemu) 2014-01-03 19:18:26 shafire, the x86_64 one may or may not boot. depends if libgcc_s is needed or not 2014-01-03 19:18:36 for some reason armhf does not need it; but x86 needs it 2014-01-03 19:47:05 fabled: kernel panic, does not boot 2014-01-03 19:48:08 libgcc is missing, yeah 2014-01-03 20:05:55 shafire, re-created the .iso. try again. (did same fix as for x86) 2014-01-03 20:06:40 ok 2014-01-03 21:15:04 now, it works :) 2014-01-03 21:15:09 fa 2014-01-03 21:15:14 fabled: ^ 2014-01-03 21:16:16 will be there a new lte kernel? 2014-01-03 21:16:31 which we could use for 3.0? 2014-01-03 21:17:09 perhaps. 2014-01-03 21:17:14 good that it works 2014-01-03 21:18:15 fabled: autostart doesn't work, had to stop booting and type grsec in 2014-01-03 21:18:25 autostart = kernel panic 2014-01-03 21:18:32 huh. that's weird 2014-01-03 21:19:51 I am testing in Virtualbox 2014-01-03 21:21:30 mmm 2014-01-03 21:21:43 the new busybox (beta) grep is broke 2014-01-05 10:16:11 09:24:38 --> | shafire (b84bd22a@gateway/web/freenode/ip.184.75.210.42) has joined #alpine-devel 2014-01-05 10:16:13 ACTION stab 2014-01-05 19:36:58 umh 2014-01-05 19:37:17 i'm looking for the best way to add a kernel module 2014-01-05 19:37:40 but it's proprietary 2014-01-05 19:37:49 (blackmagic capture card) 2014-01-05 19:38:02 i managed to compile the module 2014-01-05 19:38:22 now i should put it into a usb alpine 2014-01-05 19:38:56 i shoudl add it into initramfs 2014-01-05 20:20:12 fcolista, for which bm device? i've written some userland stuff for atem. 2014-01-05 20:20:59 it's thunderbult 2014-01-05 20:21:02 *bolt 2014-01-05 20:21:09 oh. right. 2014-01-05 20:21:11 Intensity Extreme 2014-01-05 20:21:27 i've compiled the module 2014-01-05 20:21:31 i reversed the atem tv usb h264 stuff earlier. 2014-01-05 20:21:41 wow 2014-01-05 20:21:58 since i've a mac with bm 2014-01-05 20:22:06 i'm looking at the way to use it 2014-01-05 20:22:14 thunderbolt is supported 2014-01-05 20:22:18 but not the otplug 2014-01-05 20:22:20 for the atem stuff you can see https://github.com/fabled/bmd-tools 2014-01-05 20:22:22 *hotplug 2014-01-05 20:22:26 it works. but needs still fixing a bit. 2014-01-05 20:23:01 8-| 2014-01-05 20:23:03 have not tried the thunderbolt stuff 2014-01-05 20:23:56 to use a custom module with alpine, what's the best approach? 2014-01-05 20:24:06 compile it in a build envirnoment 2014-01-05 20:24:12 get the module 2014-01-05 20:24:23 have an alpine with the same kernel 2014-01-05 20:24:30 and put the ko in the initram ? 2014-01-05 20:24:47 alpine boot from usb 2014-01-05 20:25:19 you'd need to create your own initramfs, or just keep the .ko somewhere and load it with separate init.d script 2014-01-05 20:25:35 it's slightly inconvenient to do it atm. 2014-01-05 20:25:44 hopefully it gets easier to do in future 2014-01-06 07:13:46 morning 2014-01-06 08:03:28 morning 2014-01-06 08:04:34 my laptop boots with alpine edge-musl x86_64 now 2014-01-06 08:04:39 in a clumsy way... 2014-01-06 08:04:52 but at least it boots 2014-01-06 08:05:15 i will set up the desktop today i think 2014-01-06 08:06:29 i have macosx 15" retina with alpine x86_64 2014-01-06 08:06:38 yesterday i was able to make it boot 2014-01-06 08:06:56 i've not yes tryied video 2014-01-06 08:07:10 (which is nvidia, so i suppose that nouveau will work). 2014-01-06 08:07:38 what laptop do you have ncopa? 2014-01-06 08:08:39 mac with retina and alpine... you are crazy... LOL 2014-01-06 08:08:52 i mean *COOL* :D 2014-01-06 08:08:52 :-) 2014-01-06 08:09:05 also, it boots from usb :) 2014-01-06 08:09:32 i've a blackmagic thunderbolt capture card and i want to use it 2014-01-06 08:09:32 its a fujitsu siemens lifebook uh552 2014-01-06 08:10:04 its an "older" model (aug 2012?) 2014-01-06 08:10:13 but 1) no touchscreen 2014-01-06 08:10:22 (so no greasy fingers on screen) 2014-01-06 08:10:44 2) it has i5 for the price of amd a4/a6 2014-01-06 08:10:48 which is good 2014-01-06 08:10:55 3) its thin (20mm) 2014-01-06 08:11:07 4) it can expand the RAM up to 16GB... 2014-01-06 08:11:23 5) it has clamied battery time for 7.5 hours (!) 2014-01-06 08:11:35 wow 2014-01-06 08:11:56 6) its really easy to open it up and replace disk/ram 2014-01-06 08:12:30 nice choice 2014-01-06 08:12:38 drawback is that it has no ethernet plug 2014-01-06 08:12:45 so ethernet via usb... 2014-01-06 08:12:48 oh 2014-01-06 08:13:07 or wifi 2014-01-06 08:13:07 maybe because is thin 2014-01-06 08:13:11 yeah 2014-01-06 08:13:22 thunderbolt btw works good 2014-01-06 08:13:26 cool 2014-01-06 08:13:27 on apple 2014-01-06 08:13:31 but no hotplug support 2014-01-06 08:13:44 devices must be connected when alpine start 2014-01-06 08:13:48 oh, 7) it has 2x usb3 + 1 usb2 2014-01-06 08:14:09 fcolista: might be a kernel config option? 2014-01-06 08:14:21 bah 2014-01-06 08:14:28 from what i read no 2014-01-06 08:14:33 ok 2014-01-06 08:14:34 is matter of lacking support 2014-01-06 08:14:37 in the hw? 2014-01-06 08:14:44 basically thunderbolt is a pci-e 2014-01-06 08:14:55 that on a normal pc will work out-of-the-box 2014-01-06 08:15:03 while apple remaps acpi 2014-01-06 08:15:16 in a (strange) way? 2014-01-06 08:15:23 so it does not hotplug on osx either? 2014-01-06 08:15:34 hotplug is ok on macosx 2014-01-06 08:15:43 linux support is not maturefor hotplug 2014-01-06 08:15:49 ok, get it 2014-01-06 08:16:00 its not fixed in 3.12? 2014-01-06 08:16:12 i've 3.10 2014-01-06 08:16:14 http://lkml.indiana.edu/hypermail/linux/kernel/1311.3/03298.html 2014-01-06 08:16:15 ooops 2014-01-06 08:16:16 :) 2014-01-06 08:16:46 btw 2014-01-06 08:16:57 i've 2.7.2 x86_64 2014-01-06 08:17:00 on an usb 2014-01-06 08:17:09 with efi? 2014-01-06 08:17:14 yes 2014-01-06 08:17:17 cool! 2014-01-06 08:17:19 well 2014-01-06 08:17:26 i read up on efi this weekend 2014-01-06 08:17:28 it's a workaround 2014-01-06 08:17:44 how do you do it? 2014-01-06 08:17:59 rEFInd? 2014-01-06 08:18:04 rEFIt? 2014-01-06 08:18:16 there is an hacky tool called ISO-2-USB EFI-Booter for Mac 2014-01-06 08:18:17 no 2014-01-06 08:18:23 no install on mac 2014-01-06 08:18:43 this gives you two files to put on a mbr partition 2014-01-06 08:18:53 just mkdir -p efi/boot 2014-01-06 08:18:59 put those two files there 2014-01-06 08:19:11 then you are able to make alpine boot 2014-01-06 08:19:19 interesting 2014-01-06 08:19:25 yes 2014-01-06 08:19:33 as i said, i read about UEFI this weekend 2014-01-06 08:19:40 i'm wondering to do write an howto on w.a.o 2014-01-06 08:19:44 there are like 5 different ways to do it 2014-01-06 08:19:53 yes, it's a jungle 2014-01-06 08:20:22 well, its 5 different ways to do boot manager and 5 different ways to do boot loader 2014-01-06 08:20:37 and N different combinations ... 2014-01-06 08:21:03 in theory, with EFI you dont need the bootloader 2014-01-06 08:21:24 http://pastebin.com/nRC84r2n 2014-01-06 08:21:33 you can install linux as an efi program 2014-01-06 08:21:34 this is the tree of my usb pen 2014-01-06 08:22:08 bootX64.efi 2014-01-06 08:22:11 yes 2014-01-06 08:22:14 bootIA32.efi 2014-01-06 08:22:19 32 and 64 2014-01-06 08:22:24 its a 64 bit mac? 2014-01-06 08:22:28 yep 2014-01-06 08:22:35 then i think boot IA32.efi will never be used 2014-01-06 08:22:43 exactly 2014-01-06 08:22:50 bootX64.efi 2014-01-06 08:22:58 i think thats the 'fallback' name 2014-01-06 08:23:02 dunno if something we can implement on alpine 2014-01-06 08:23:23 you should be able to set the filepath in the firmware 2014-01-06 08:23:48 so you could have EFI/syslinux/syslinux.efi 2014-01-06 08:23:59 or EFI/alpine/vmlinux-grsec.efi 2014-01-06 08:24:28 it's matter of find the right path :) 2014-01-06 08:24:33 or EFI/somebootmgr/somebootmgr-that-can-chainload-efis.efi 2014-01-06 08:24:45 i dont know how to set it in firware either 2014-01-06 08:24:59 i think we need fix compile options for that in our kernel config 2014-01-06 08:25:07 efi file you mean ? 2014-01-06 08:25:07 and some user space tool 2014-01-06 08:25:13 yes 2014-01-06 08:25:39 i have some idea what we can do 2014-01-06 08:25:46 and some unanswered questions... 2014-01-06 08:25:51 brb 2014-01-06 08:36:30 back 2014-01-06 08:38:30 fabled: btw, i think i will need qemu musl soonsish 2014-01-06 08:38:39 the patches from sabotage works, but breaks uclibc 2014-01-06 08:42:05 we need fix mkinitfs to use scanelf to automatically find ldd deps 2014-01-06 09:35:51 fabled: re issue #2529 2014-01-06 09:36:01 i wonder if we should have a boot option for it 2014-01-06 09:36:21 so you do: alpine_arch=x86_64 2014-01-06 09:36:34 or alpine_config_arch=x86_64 2014-01-06 09:36:52 or alpine_apk_arch=x86_64 2014-01-06 09:37:39 rather than it just convert it no questions asked 2014-01-06 09:37:58 maybe a boolean: alpine_update_arch=yes 2014-01-06 09:52:42 fabled: ok? http://sprunge.us/MhJM 2014-01-06 11:36:10 ncopa, what i'd like 2014-01-06 11:36:20 is to initrd detect which arch the boot media is 2014-01-06 11:36:22 and force it 2014-01-06 11:43:36 mmm 2014-01-06 12:04:01 i had a patch for that but after second thought i ended up with "update_apkovl_arch" boot option 2014-01-06 12:04:11 + an informative error message 2014-01-06 12:04:26 http://sprunge.us/jKEY 2014-01-06 12:04:40 the reason is, if you mix apkovl with boot is arch 2014-01-06 12:04:57 then something is wrong, and might not be what you intended 2014-01-06 12:05:51 if we force user to use boot option, then we are sure user knows what he is doing 2014-01-06 12:06:05 and it is not an accident 2014-01-06 13:03:27 hi 2014-01-06 13:03:57 fabled: are the musl iso links public? http://alpinelinux.org/forum/installation/installing-edge-musl 2014-01-06 13:25:05 shafire, yes 2014-01-06 13:25:20 they are not officially supported. but you can post the link,. 2014-01-06 13:46:43 fabled: in any case: are you ok with apk --show-arch 2014-01-06 13:46:50 or maybe apk --print-arch 2014-01-06 13:47:11 yeah, it's probably good to have it in there 2014-01-06 13:47:22 --show-arch or --print-arch? 2014-01-06 13:48:10 maybe print? 2014-01-06 13:48:18 what do other tools do? 2014-01-06 13:49:26 dunno 2014-01-06 13:50:29 dpkg --print-architecture 2014-01-06 13:50:35 actually 2014-01-06 13:50:40 http://man.he.net/man1/dpkg-architecture 2014-01-06 13:51:06 Instead of: 2014-01-06 13:51:06 ARCH=`dpkg --print-architecture` 2014-01-06 13:51:06 please use the following: 2014-01-06 13:51:06 configure $(ARCH)-linux 2014-01-06 13:51:06 DEB_BUILD_GNU_TYPE := $(shell dpkg-architecture -qDEB_BUILD_GNU_TYPE) 2014-01-06 13:51:07 DEB_HOST_GNU_TYPE := $(shell dpkg-architecture -qDEB_HOST_GNU_TYPE) 2014-01-06 13:51:09 configure --build=$(DEB_BUILD_GNU_TYPE) --host=$(DEB_HOST_GNU_TYPE) 2014-01-06 13:51:34 i think --print-arch is ok 2014-01-06 13:51:40 ok 2014-01-06 13:51:45 let's go with that then 2014-01-06 14:00:16 whoops 2014-01-06 14:00:28 i had another commit in there 2014-01-06 14:00:53 i think that was wrong 2014-01-06 14:01:20 i'm not sure 2014-01-06 14:02:44 fabled: was that commit ok? http://tinyurl.com/mvy6a86 2014-01-06 14:03:16 i think it will not work 100% in all situations 2014-01-06 14:03:22 should probably just turn off the progress flag 2014-01-06 14:03:26 in the main 2014-01-06 14:03:30 not in options parsing 2014-01-06 14:03:44 i've update kernel 3.10.25 on alpine 2.7.2 disk install. On usb i've the same version, but with kernel 3.10.23. How can i upgrade the kernel on usb? Is no matter of copy /boot, since grserc.modloop.squahfs does not exist in disk boot. Someone can put me in the right direction? 2014-01-06 14:04:17 fcolista: you'll have to create a modloop 2014-01-06 14:04:26 i think we in the longrun should have a scriptf or that... 2014-01-06 14:04:30 aaah 2014-01-06 14:04:39 what i normally do is 2014-01-06 14:04:46 i build a new iso with alpine-iso 2014-01-06 14:05:07 and then just copy the isotmp.*/boot 2014-01-06 14:05:26 i think what we want is a couple of more scripts 2014-01-06 14:05:43 one mkmodloop 2014-01-06 14:05:45 or similar 2014-01-06 14:06:03 ncopa, you read my mind. 2014-01-06 14:07:13 and then one like: update-kernel /media/usb/boot 2014-01-06 14:07:35 that will fetch new kernel, generate initramfs, and modloop 2014-01-06 14:07:46 also 2014-01-06 14:07:51 while working on efi 2014-01-06 14:08:05 i came to the conclusion that we should avoid using symlinks in /boot 2014-01-06 14:08:15 due to FAT filesystem 2014-01-06 14:08:29 it might be convenient to be able to mount the EFI system partition as /boot 2014-01-06 14:08:40 which means that the kernel will be stored on fat 2014-01-06 14:09:16 then EFI firmware will be able to boot the kernel directly, without the need for any boot loader like syslinux/grub 2014-01-06 14:09:33 we currently use debian's installkernel 2014-01-06 14:09:38 which creates symlinks 2014-01-06 14:09:43 we might want rethink that 2014-01-06 14:10:16 i'm quite interested on understanding how all these pieces works together 2014-01-06 14:10:34 the main purpose to use symlinks in /boot is so you can have versioned kernels 2014-01-06 14:10:46 eg vmlinuz-3.12.6-0-grsec 2014-01-06 14:11:03 and a symlink vmlinuz-grsec -> default 2014-01-06 14:11:57 we can continue to have the versioned kernels, but then we depend on some hook/script that generates the boot menu 2014-01-06 14:12:10 (eg update-extlinux-conf) 2014-01-06 14:12:41 but that might be tricky if user want boot directly with EFI (and skip the boot loader) 2014-01-06 14:13:17 the versioned kernels in /boot is mostly useful when you are working on kernel sources 2014-01-06 14:13:22 make && make install 2014-01-06 14:13:27 in kernel source 2014-01-06 14:14:08 but i thin in practice when working with kernel, we dont use the versioned kernel feature anyways 2014-01-06 14:14:34 because its not that often that the kernel compile box / dev box is the thing you want reboot 2014-01-06 14:14:50 you normally want boot that development kernel in a virtual machine 2014-01-06 14:15:22 or something similar 2014-01-06 14:15:48 several scripts are needed 2014-01-06 14:15:48 so i think what we really want, is something that generate a /boot dir with modloop 2014-01-06 14:16:29 might be we could have the 'make install' from kernel build call our script (eg our own installkernel) 2014-01-06 14:16:30 Ok. What's the starting point? i clone alpine-iso repo 2014-01-06 14:16:35 yeah 2014-01-06 14:16:49 i think thats the simplest 2014-01-06 14:16:50 is matter of figuring out what part is needed 2014-01-06 14:16:58 of that Makefile 2014-01-06 14:17:06 i think there is a: make modloop 2014-01-06 14:17:07 and create a mkmodloop 2014-01-06 14:17:08 or so 2014-01-06 14:17:17 so you could look in the Makefile 2014-01-06 14:17:32 is there 2014-01-06 14:17:49 fabled: so i revert that apk-tools commit 2014-01-06 14:17:56 fetch: run quiet when --stdout 2014-01-06 14:18:09 where is supposed to go the mkmodloop ? On another project like mkinitfs ? 2014-01-06 14:18:21 dunno yet 2014-01-06 14:18:23 k 2014-01-06 14:18:41 it might be a separate project, like alpine-installkernel 2014-01-06 14:18:44 or similar 2014-01-06 14:18:53 need check what arch linux and fedora does there 2014-01-06 14:18:55 is still not very clear to me the big picture, but i'll try to figuring out by myself. I will ask 2014-01-06 14:19:43 i know where to start: alpine-iso. I need where should i produce: a script that does a little part of alpine-iso: modloop. 2014-01-06 14:20:06 It basically need to take a kernel and all related things, then produce the modloop. 2014-01-06 14:20:14 Or everything is needed for boot. 2014-01-06 14:20:20 Am i right? 2014-01-06 14:20:31 yes 2014-01-06 14:20:37 good. 2014-01-06 14:20:38 basically: mkdir tmpdir 2014-01-06 14:20:41 cd tmpdir 2014-01-06 14:20:58 apk fetch --stdout linux-grsec | tar -zx 2014-01-06 14:21:19 depmod something ... 2014-01-06 14:21:43 mkinitfs -b $PWD 2014-01-06 14:22:23 mksquashfs lib boot/grsec.modloop.something -comp xz 2014-01-06 14:22:30 or something like that 2014-01-06 14:22:41 i think thats basically the steps 2014-01-06 14:22:46 unpack the kernel package 2014-01-06 14:22:54 depmod 2014-01-06 14:22:56 ncopa, we really should do 'setup-bootable-kernel' 2014-01-06 14:22:56 mkinitfs 2014-01-06 14:23:06 mksquashfs 2014-01-06 14:23:32 fabled: yes. those are the things the script would need to do 2014-01-06 14:23:48 fabled: and i woudl like to make it run handy from make install in kernel soruces 2014-01-06 14:24:00 maybe as a hook (installkernel?) 2014-01-06 14:24:42 so you from your kernel build easily can copy/scp to a boot usb 2014-01-06 14:25:18 i want it be able to run from kernel build or apk fetch prebuild kernel 2014-01-06 14:27:05 fcolista: maybe: mkmodloop -a 2014-01-06 14:27:25 mkmodloop -b $BASEDIR 2014-01-06 14:27:40 mkmodloop -o /boot/file.modloop -b $BASEDIR 2014-01-06 14:29:51 ok 2014-01-06 14:30:20 i'll try to do something. 2014-01-06 14:31:55 do we need it in lua or shell script? I think the second since we should not depend from lua for this kind of things... 2014-01-06 14:32:00 but just as confirm 2014-01-06 14:43:20 shell 2014-01-06 14:43:23 or C 2014-01-06 14:43:32 correct, we dont want depend on lua for this 2014-01-06 14:43:46 mkmodloop if fairly easy 2014-01-06 14:43:52 is* 2014-01-06 14:44:09 cd .. 2014-01-06 14:44:46 k 2014-01-06 19:49:21 hi 2014-01-06 19:50:34 on musl x86_64 intel_drv.so says: fbPutImage: symbol not found 2014-01-06 19:55:30 i wonder what lib provides that symbol 2014-01-06 19:56:05 or it might be that lazy linking is not working on musl? 2014-01-06 19:56:47 xvfb? fbdev? 2014-01-06 20:09:13 sounds like that yes 2014-01-06 20:10:04 libfb.so 2014-01-06 20:10:27 in /usr/xorg/modules/libfb.so 2014-01-06 20:12:45 ah 2014-01-06 20:13:03 "musl intentionally does not support lazy binding" 2014-01-06 20:14:48 googling xorg and musl gave that 2014-01-06 21:08:16 hi from hexchat on musl alpine :) 2014-01-07 08:48:43 algitbot: build 2.6-stable 2014-01-07 13:28:14 I am getting this problem: nf_ct_sip: dropping packetIN= OUT=eth1 SRC= DST= LEN=402 TOS=0x00 PREC=0x00 TTL=63 ID=0 DF PROTO=UDP SPT=5060 DPT=5060 LEN=382 2014-01-07 13:28:28 does this apply: http://us.generation-nt.com/answer/bug-715822-linux-image-3-9-1-686-pae-sip-nat-problem-kernel-3-9-help-211829472.html 2014-01-07 13:28:29 ? 2014-01-07 13:29:37 uname -r: 3.6.11-grsec 2014-01-07 14:13:30 rnalrd: i have upgraded the v2.7 kernels to 3.10.25 2014-01-07 14:13:43 i think linux-virt-grsec is 3.10.23 still 2014-01-07 14:14:30 rnalrd: do you want linux-virt-grsec to be 3.10.25 for the v2.7.3 release? 2014-01-07 14:14:35 i'd like to tag release now 2014-01-07 14:15:57 ncopa, give me 5 mins 2014-01-07 14:16:22 actually 10, to compile new ver 2014-01-07 14:51:32 ncopa, done 2014-01-07 14:52:32 thanks 2014-01-07 15:01:48 algitbot: build 2.7-stable 2014-01-08 08:53:02 question: 2014-01-08 08:53:29 any ideas regarding the massive boot problems on older systems using ide-cdrom-drives? 2014-01-08 09:11:31 what "massive boot problems"? 2014-01-08 09:13:05 the system boots from cdrom for some seconds and reboot to start again 2014-01-08 09:13:07 endless loop 2014-01-08 09:13:40 it reboots before something is written to the screen 2014-01-08 09:14:27 sounds like it reboots when loading some of the hardware drivers 2014-01-08 09:14:36 the trick is to find out which :-/ 2014-01-08 09:14:51 at boot prompt you can do: grsec s 2014-01-08 09:15:00 and it should give you a rescue shell early 2014-01-08 09:15:33 http://www.syslinux.org/wiki/index.php/ISOLINUX#What_to_do_when_nothing_works... 2014-01-08 09:15:42 then you could try load the kernel modules one by one and see which driver it is 2014-01-08 09:15:51 hm... 2014-01-08 09:16:05 what hardware is it? 2014-01-08 09:16:25 might be kernel uses some unsupported feature. 2014-01-08 09:16:31 can't say, some of our users gave us this feedback 2014-01-08 09:16:38 so i have to figure out more details 2014-01-08 09:16:53 maybe i have such an old system @home to try and reproduce 2014-01-08 09:17:09 but i think there are many different systems with this problem out in the wild... 2014-01-08 09:17:28 probably yes 2014-01-08 09:17:42 only cdrom boot i do nowdays is in qemu 2014-01-08 09:18:06 yeah, we can't find any problems if using virtualisation 2014-01-08 09:18:18 vmware, virtualbox a.s.o works without problems... 2014-01-08 09:18:38 but with real metal it's different :-/ 2014-01-08 09:31:30 so i think its a hw driver that triggers it 2014-01-08 09:42:23 ok, we built an iso without "quiet" option... 2014-01-08 09:42:39 let's see what happens... 2014-01-08 09:43:52 here are two "problem-boards": 2014-01-08 09:44:00 Asus-CUBX-E mit Pentium 3 http://www.motherboards.org/mobot/motherboards_d/ASUS/CUBX-E/function.mysql-pconnect 2014-01-08 09:44:01 Intel Core 5 mit DH77EB http://www.intel.com/content/www/us/en/motherboards/desktop-motherboards/desktop-board-dh77eb.html 2014-01-08 12:56:24 ncopa: just to let me know: why was grub removed? 2014-01-08 12:59:49 was it ever added? 2014-01-08 12:59:57 grub1 needed N patches 2014-01-08 13:00:06 i dont think we ever added grub2 2014-01-08 13:00:19 due to the grub1 trauma 2014-01-08 13:00:41 i'd rater add gummiboot for UEFI 2014-01-08 13:00:47 or rEFInd 2014-01-08 13:08:53 ncopa: i think the only people who genuinely care about grub2 are those who need multiboot spec? 2014-01-08 13:09:03 or is it supported by something other than grub2 2014-01-08 13:09:48 syslinux supports multiboot 2014-01-08 13:15:45 then indeed no reason to care about grub 2014-01-08 13:17:29 except the support for old systems... 2014-01-08 13:33:15 i'm not so sure syslinux does multiboot with UEFI 2014-01-08 13:33:36 i will have to find out soonish 2014-01-08 13:33:42 my new laptop will dual boot 2014-01-08 13:34:06 i currently boot it from usb due to UEFI 2014-01-08 13:34:50 another question: 2014-01-08 13:35:12 how can i separate between different versions of a source-archive for different alpine-versions? 2014-01-08 13:35:17 lets say, 2014-01-08 13:35:35 the build for alpine 2.7 needs foo-1.1.0.tgz 2014-01-08 13:35:47 and the build for alpine 2.6 needs foo-1.0.0.tgz 2014-01-08 13:40:07 StarWarsFan: you cannot. I create a git branch for each alpine release branch 2014-01-08 13:40:44 hm, that's odd... 2014-01-08 13:41:36 in the APKBUILD you cannot say, for v2.6 use X and for v2.7 use Y 2014-01-08 13:41:51 instead, you have 2 different APKBUILDs for v2.6 and for v2.7 2014-01-08 13:41:55 in differnt git branches 2014-01-08 13:42:31 i c 2014-01-08 13:42:37 or am i misunderstanding the question? 2014-01-08 13:42:44 no 2014-01-08 13:44:00 well, to explain: 2014-01-08 13:44:34 i want to build a package out of one repository for different alpine releases 2014-01-08 13:44:58 this is working without any problems as long as the referenced sources are the same 2014-01-08 13:46:12 i use different git branches and different toolchains 2014-01-08 13:46:20 different build box 2014-01-08 13:46:27 ncopa: why do you need anything for uefi boot? 2014-01-08 13:46:43 because my laptop uses it 2014-01-08 13:46:55 and win 8.1 is preinstlaled on GPT -> UEFI 2014-01-08 13:47:11 and it cannot dualboot unless i install linux as UEFI too 2014-01-08 13:47:12 i boot linux, another linux and my toy os with my laptop's stock uefi 2014-01-08 13:47:26 and it'd boot windows too if i had space for it on ssd 2014-01-08 13:47:43 recent linux kernels are efi binaries if you ask for that in config 2014-01-08 13:47:51 you can boot them directly 2014-01-08 13:47:58 i enabled efivarfs yesterday in kernel 2014-01-08 13:48:02 i read about it yes 2014-01-08 13:48:10 i have a few questions though 2014-01-08 13:48:19 (i spent last weekend reading up on the tpoic) 2014-01-08 13:48:33 as i understand there are multiple ways to do dual booting 2014-01-08 13:48:52 1) your firmware can pick from a list of efi binaries 2014-01-08 13:49:37 2) you can install a boot loader (manager?) like grub2 or gummiboot that can chainload efi binaries 2014-01-08 13:50:01 as i understand option 1 depends on support from the firmware? 2014-01-08 13:50:26 or is there a standard that says that the firmware should provide a boot menu with various efi binaries? 2014-01-08 13:50:51 no, i don't think there is a standard for that. 2014-01-08 13:51:26 so if you want boot vmlinuz.efi directly, then you cannot dualboot unless your firmware has support for it 2014-01-08 13:51:40 (vmlinuz.efi cannot chainload efis...) 2014-01-08 13:51:53 i actually dont think syslinux.efi can do that either 2014-01-08 13:55:52 well, i stand corrected. while the boot menu itself isn't governed by the standard, the bootloader behaviour (including the BootXXXX variables) is included in the uefi spec. 2014-01-08 13:56:49 i suppose i need read the uefi spec 2014-01-08 13:57:11 general feelings about uefi: 2014-01-08 13:57:15 GPT is nice 2014-01-08 13:57:47 requirement of FAT filesystem for efi system partition is not so nice 2014-01-08 13:58:17 well, uefi spec is boring. :-) there are very few realistic uefi implementaions, however, easier to survey them 2014-01-08 13:58:21 (fragile, no online resize support, patents, etc) 2014-01-08 13:58:37 input from me: ami aptio supports multiple boot entries 2014-01-08 13:58:47 remaining to survey: award and phoenix 2014-01-08 13:59:02 might be mine supports it too. i'll have to check/test 2014-01-08 13:59:46 secure boot: scary. its all about gaining control of what you boot on your computer (eg in future they will force hw vendors lock it to microsoft) 2014-01-08 13:59:55 what'd you like instead of fat? 2014-01-08 14:00:02 i'd rather it not be exfat 2014-01-08 14:00:24 dunno, anything open and free? 2014-01-08 14:00:45 ext2/ext3/ext4? 2014-01-08 14:01:03 its all politics 2014-01-08 14:01:18 would be nice with something that supports symlinks too 2014-01-08 14:01:22 well, i wouldn't mind if windows/osx supported ext2 out of the box 2014-01-08 14:01:32 agree 2014-01-08 14:01:43 but its politics... 2014-01-08 14:01:47 boring 2014-01-08 14:01:51 without that it's unreasonable to call for efi to standardise on something that 2/3 of the major systems don't support 2014-01-08 14:02:05 i understand that perfectly 2014-01-08 14:02:46 there should be a standard filesystem, which is open and free, and has support for ownership, hard and soft links 2014-01-08 14:03:01 that woudl have been good 2014-01-08 14:03:13 unfortunally, thats not how the world works 2014-01-08 14:03:20 competition instead of cooperation 2014-01-08 14:03:26 yes, this indeed feels like a lost opportunity 2014-01-08 14:03:40 if not for the patent threat, fat is about half the way there 2014-01-08 14:03:49 and as a result we have to live with a technically inferior solution 2014-01-08 14:03:55 no nice features, but you can implement it in half an hour 2014-01-08 14:04:04 still not support for sym/hard links 2014-01-08 14:04:17 i understad perfectly why fat was used 2014-01-08 14:04:27 on the + side, its easy to implement 2014-01-08 14:04:59 reminds me, we use installkernel from debian atm 2014-01-08 14:05:02 i think we want change that 2014-01-08 14:05:08 to something that dont use symlinks 2014-01-08 14:05:20 basically, we can no longer use symlinks in /boot 2014-01-08 14:05:50 ..unless we want depend on a 3rd party linux boot loaader like syslinux, gummiboot or grub 2014-01-08 14:08:01 well, i was actually surprised that i didn't break anything in archlinux when i set it up with a fat /boot 2014-01-08 14:08:38 only need stay away from symlinks and it should jsut work 2014-01-08 14:08:52 i think it will break in alpine linux due to debians installkernel 2014-01-08 14:08:56 we want fix that 2014-01-08 14:13:20 do you support multiple kernels? 2014-01-08 14:13:31 or I'm just missing the reason for installkernel 2014-01-08 14:13:46 yes 2014-01-08 14:13:52 we support multiple kernels 2014-01-08 14:16:43 nice 2014-01-08 14:17:13 we have an update-extlinux script that generates a boot menu 2014-01-08 14:43:14 i have phoenix bios 2014-01-08 15:07:07 ugh 2014-01-08 15:07:30 i cannot access the efivarfs unless i boot in efi mode.. 2014-01-08 15:08:06 i hope this bios can boot usb in uefi mode 2014-01-08 15:46:11 ncopa: makes sense, while you can theoretically just read the variables off nvram, i think linux' efivars use efi runtime services for that 2014-01-08 15:46:58 i wanted mount -t efivarfs /sys/firmware/efi 2014-01-08 15:47:10 trying to mess with windows bcdedit now 2014-01-08 15:47:31 i was able to create a boot entry in the menu that points to \EFI\syslinux\syslinux.efi 2014-01-08 15:47:41 but windows fails to load it 2014-01-08 15:52:13 it thinks it is a windows boot loader naed \EFI\syslinux\syslinux.efi 2014-01-08 15:52:44 ok, i think i will ahve to clone my current boot usb 2014-01-08 15:52:50 create a gpt usb 2014-01-08 15:53:12 and try make phoenix boot the usb as efi 2014-01-08 15:54:09 or build gummiboot or refind 2014-01-08 15:54:16 and let that be the master boot mgr 2014-01-08 15:54:44 i *think* my laptop boots off mbr media just fine in efi mode 2014-01-08 15:55:27 but then again, i can just put the path to the efi executable i want to run in bios setup 2014-01-08 15:56:18 i cant :-/ 2014-01-08 15:56:33 i was kinda hoping that i could to it with efivarfs 2014-01-08 15:56:47 there is a boot menu 2014-01-08 15:56:57 and Windows boot manager is one of the entries 2014-01-08 15:57:07 so it might be i can add it in there 2014-01-08 15:58:04 there is a somewhat braindamaged tool designed to do that, efibootmgr 2014-01-08 15:59:23 currently, i think gummiboot appears to be the lowest hanging fruit 2014-01-08 15:59:33 ok gotta go 2014-01-08 16:09:21 gummiboot is great, I'm using it on my laptop 2014-01-08 18:22:38 Do you (anybody) think that http://bigbluebutton.org/ can be built on Alpine Linux? 2014-01-08 18:24:20 mhavela: last time I/we looked, it was java issue that was blocker 2014-01-08 18:24:32 i think it also wanted to use tomcat or something like that 2014-01-09 07:57:06 morning 2014-01-09 08:42:56 files from v2.4.11-173-g6445167 uploaded 2014-01-09 10:28:46 ncopa, how about efilinux ? suitable from EFI point ? 2014-01-09 10:32:08 what does efilinux provide that syslinux/gummiboot/Refind does not? 2014-01-09 10:41:10 did not do much reading about it 2014-01-09 10:41:39 but code seems neat 2014-01-09 10:42:21 its dependent on gnu-efi 2014-01-09 10:43:06 they all are 2014-01-09 10:44:30 :) 2014-01-09 10:45:03 zram now in mains 2014-01-09 10:55:22 ncopa: any decision on test servers ? 2014-01-09 10:55:37 someone needs to set it up 2014-01-09 10:55:49 and maintain them 2014-01-09 10:56:28 I could probably set up the basic install 2014-01-09 10:56:40 and give someone login to set up the auto part 2014-01-09 10:56:57 also 2014-01-09 10:57:24 someone needs to do the admin work of ssh keys (who should have access to push to it) etc 2014-01-09 10:57:40 I think just a up and running vm/guest with login would suffice 2014-01-09 10:57:59 what do we name it? 2014-01-09 10:58:58 url or server names ? 2014-01-09 10:59:04 should it be x86 or x86_64? 2014-01-09 10:59:08 server name 2014-01-09 10:59:35 i could try get a box up ad run 2014-01-09 10:59:41 and give you ssh login 2014-01-09 10:59:42 _ 2014-01-09 10:59:47 and let you set it up 2014-01-09 11:00:11 a few things 2014-01-09 11:00:41 you should not be forced to push via this staging server to be able to push to official repos 2014-01-09 11:00:52 if that is a requirement it would build 3 times... 2014-01-09 11:00:54 yes, thats the idea 2014-01-09 11:01:00 1) on my dev box 2014-01-09 11:01:03 local box 2014-01-09 11:01:07 2) on staging server 2014-01-09 11:01:12 3) on production 2014-01-09 11:01:35 no direct connect with production 2014-01-09 11:01:52 so it would mean 3x build at minimum? 2014-01-09 11:01:58 for a new pkg 2014-01-09 11:02:00 once things desired go well a document or report can be generated 2014-01-09 11:02:30 what happens if the staging build server is building something heavy 2014-01-09 11:02:33 and i want push? 2014-01-09 11:02:44 its a vm limited by cpu 2014-01-09 11:02:49 lets say its 12:02 pm 2014-01-09 11:02:58 i want push my work before i go to lunch 2014-01-09 11:03:02 so its done when i cam back 2014-01-09 11:03:25 yes, thought about it, some guide can be set as suitable timings 2014-01-09 11:03:39 i'll give you a box and let you set it up ;) 2014-01-09 11:04:06 I notice the servers are generally idle when you leave for the day 2014-01-09 11:04:25 but i am not really happy with the idea of slowing down building of new packages 2014-01-09 11:04:36 ...just because some ppl are to lazy to build locally 2014-01-09 11:04:44 other possibility is requesting time slots 2014-01-09 11:04:57 look 2014-01-09 11:05:03 i do my work on a package 2014-01-09 11:05:04 i push 2014-01-09 11:05:06 done! 2014-01-09 11:05:12 thats all i should need to do 2014-01-09 11:05:40 building locally is not an issue, just give me free/trouble fee internet, new hardware 2014-01-09 11:06:36 I ok with even a 4-5yrs old donated hardware 2014-01-09 11:09:21 if security is an issue, I am ok with a preset vm, user login and IP limits to user and within AL network 2014-01-09 11:10:15 /etc/host can be limited to AL pkg mirror and dev.a.o 2014-01-09 11:13:37 re: 3x, I am not planning to build full aport just few of interest or on need to 2014-01-09 11:42:26 why should it slow down building of new pkgs ? 2014-01-09 11:43:05 devs set it similar to local setup, with git-send-email.. etc and send APKBUILD similarly 2014-01-09 12:57:36 let's call it just test buildbox, if you can't build your changes locally atm 2014-01-09 13:01:19 it wouldn't even keep whole aports tree up to date 2014-01-09 13:05:01 test box ? 2014-01-09 13:35:34 https://commotionwireless.net look interesting 2014-01-09 15:14:12 ncopa: regarding the boot problems we discussed yesterday 2014-01-09 15:14:21 this board runs into a kernel panic too: 2014-01-09 15:14:22 http://www.intel.com/content/www/us/en/motherboards/desktop-motherboards/desktop-board-dh77eb.html 2014-01-09 15:14:50 do you think you could file a bug on bugs.a.o for it? 2014-01-09 15:15:08 and add the link to it 2014-01-09 15:15:24 i tend to lose/forget things that is only mentioned in irc 2014-01-09 15:15:26 sure, no problem 2014-01-09 15:15:29 thanks! 2014-01-09 15:15:43 also, please say what kernel it is 2014-01-09 15:15:50 and arch 2014-01-09 15:16:42 ok 2014-01-09 15:57:09 [x] done: http://bugs.alpinelinux.org/issues/2591 2014-01-09 15:57:16 more info will be added soon... 2014-01-09 20:27:24 hi 2014-01-09 20:28:03 is there any plan to get openjdk7 working on musl? 2014-01-09 20:28:14 timeplan* 2014-01-10 09:50:46 fabled: openssl does not build on musl x86 2014-01-10 09:51:10 it seems like it fails on missing zlib symbols due to it tries to link to the static libcrypto.a 2014-01-10 09:51:21 due to dynamic libcrypto.so is missing? 2014-01-10 09:51:31 ncdev-edge-musl-x86:~/aports/main/openssl/src/openssl-1.0.1f$ find . -name 'libc 2014-01-10 09:51:31 rypto.*' 2014-01-10 09:51:31 ./libcrypto.pc 2014-01-10 09:51:31 ./libcrypto.a 2014-01-10 09:52:28 huh? 2014-01-10 09:52:53 it built for me 2014-01-10 09:53:05 earlier that is 2014-01-10 09:53:09 can you try in your musl x86 again? 2014-01-10 09:53:28 https://dpaste.de/31hP 2014-01-10 09:53:40 i wonder if it is ccache that messes it up for me 2014-01-10 09:53:52 works still 2014-01-10 09:53:55 sounds like ccache issue 2014-01-10 09:54:23 try apk add ccache and set CC="ccache gcc" in /etc/abuild.conf 2014-01-10 09:55:45 i suspect its the c_rehash 2014-01-10 09:56:17 if you have openssl-dev installed, then it will link against that 2014-01-10 09:57:47 why do we build as -j1? 2014-01-10 09:58:57 weird, if i build as: make && make build-shared it works 2014-01-10 10:00:55 mmm 2014-01-10 10:01:04 dunno 2014-01-10 10:01:17 i'll be away for the evening. hope you get it figured out. ttyl. 2014-01-10 10:01:41 just a wild guess: you have openssl-dev installed and it picks up that 2014-01-10 10:01:56 if openssl-dev is not installed, it tries to link against static libcrypto.a 2014-01-10 10:02:10 during make install phase it links the shared libcrypto 2014-01-10 15:14:51 is http://slurm.schedmd.com/ useful from AL point ? 2014-01-10 15:27:24 ncopa: still here? 2014-01-10 15:27:57 probably already having weekend as it supposed to be :) 2014-01-10 15:33:44 hi 2014-01-10 15:33:50 clandmeter: ping 2014-01-10 15:34:05 im here for 20mins more 2014-01-10 15:34:07 before weekend 2014-01-10 15:34:10 hi 2014-01-10 15:34:16 still not going to fosdem? 2014-01-10 15:34:28 unfortunally no :-( 2014-01-10 15:35:13 im thinking of going 2014-01-10 15:35:43 when is it? 2014-01-10 15:35:51 and what does it cost? 2014-01-10 15:35:59 1-2 feb 2014-01-10 15:36:21 where is it? 2014-01-10 15:36:47 Brussels 2014-01-10 15:37:12 isnt it free? 2014-01-10 15:37:14 s/s$// 2014-01-10 15:37:16 it is 2014-01-10 15:39:42 norwegian.no does not appear to go to brussels 2014-01-10 15:40:29 I will be there :) 2014-01-10 15:40:57 I'm even giving a talk... 2014-01-10 15:41:05 royger: yes you told us before 2014-01-10 15:41:30 well, that was last year 2014-01-10 15:41:41 lol really? 2014-01-10 15:41:47 damn time goes fast... 2014-01-10 15:41:50 lol 2014-01-10 15:42:15 i was going over the website and saw your name, so imagined it was this year. 2014-01-10 15:42:35 barthalion: are you going? 2014-01-10 15:42:41 hehehe, no it was last year, I meet barthalion there :) 2014-01-10 15:43:36 in any case, i won't have time 2014-01-10 15:43:44 maybe i should try plan something next year 2014-01-10 15:43:59 would be fun 2014-01-10 15:46:37 if anyone attents, I will be at the Xen booth, come and say hi! 2014-01-10 15:48:58 royger: indeed :p 2014-01-10 15:49:07 unfortunately I have no time this year 2014-01-10 16:05:14 hi 2014-01-10 16:11:52 I have a double room for fosdem 2014-01-10 16:11:55 If you want it 2014-01-10 16:12:01 I paid, but have no chance to go there 2014-01-10 16:12:35 hi 2014-01-10 16:12:38 how come? 2014-01-10 16:13:17 I wanted to go there 2014-01-10 16:14:36 clandmeter: do you need one? 2014-01-10 16:14:43 a room? 2014-01-10 16:14:47 yes 2014-01-10 16:14:59 from 31.01 to 02.01 2014-01-10 16:15:03 from 31.01 to 02.02 2014-01-10 16:15:08 in principle not, but it could be a way to drink more then 1 beer 2014-01-10 16:15:09 :) 2014-01-10 16:15:35 I paid already, so someone must use it 2014-01-10 16:16:04 if you dont mind, i wont mind to use it. 2014-01-10 16:16:40 It's http://www.hoteldefierlant.be/ 2014-01-10 16:17:05 with breakfast and free wlan (I think :D) 2014-01-10 16:17:16 :) 2014-01-10 20:13:43 http://alpinelinux.org/forum/general-discussion/open-letter-concerning-organisation 2014-01-10 20:14:01 does the forum have a search facility? 2014-01-10 20:16:02 sadly, I didn't even remember we had forums :P 2014-01-10 20:16:45 Now, you know it ;) 2014-01-10 20:17:01 ACTION still doesn't have cycles in the day to follow the forums 2014-01-11 13:10:59 hi 2014-01-13 09:04:43 packages that fails to build on musl x86_64: http://sprunge.us/CDNS 2014-01-13 09:04:53 i wonder if we could drop some of them 2014-01-13 09:09:50 ncopa, libunistring is not needed. kamailio was only user of it, and i patched it to not need it anymore 2014-01-13 09:10:32 i can work on few musl issues to build 4-5 more packages. (setkey/encrypt and bsd variant of struct tcphdr) 2014-01-13 09:12:42 sounds good 2014-01-13 12:13:05 ncopa: if you want to move acf-openldap to main/ I have tested it. It is working good. 2014-01-13 12:13:59 alacerda: ok. thanks 2014-01-13 12:14:24 ncopa: Thaks You 2014-01-13 13:24:59 If a Package is: copyright (c) 2003-2013 Bla bla bla.. Does it means it is not a GPL? 2014-01-13 13:26:03 alacerda, no. copyright and license to use are two separate things. 2014-01-13 13:26:25 you find copyright(c) in about every package. and after that the additional license statement 2014-01-13 13:27:07 darkhttpd doesn't seems to have a license statement 2014-01-13 13:27:38 found 2014-01-13 13:27:41 BSD license 2014-01-13 13:27:45 thanks fabled 2014-01-13 13:39:05 hi 2014-01-13 13:39:29 do you prefer nginx or lighttpd? 2014-01-13 13:40:23 I prefer nginx, but I guess that ncopa likes lighttpd more 2014-01-13 13:40:42 I used both, but I can not decide :( 2014-01-13 13:40:48 I have always this problem... 2014-01-13 13:41:06 for me it depends on task ;) 2014-01-13 13:41:20 and my friend uses lighttpd for exactly same thing that I do, so I guess it doesn't matter as long as you don't need something weird 2014-01-13 13:41:40 I need ssl and fcgi 2014-01-13 13:41:52 yeah. i generally like lighttpd; but use nginx for the 'weird' things ;) 2014-01-13 13:42:29 and I usually use nginx and for weird things – lighttpd :p 2014-01-13 13:42:29 Mh, I go with apache 2014-01-13 13:42:32 :P 2014-01-13 13:42:39 test both and decide 2014-01-13 13:42:46 or flip a coin 2014-01-13 13:43:52 One day, I wanted to buy a watch. I had two, I liked, but I could not decide and went home without any watch :( (and saved money) 2014-01-13 13:43:52 :D 2014-01-13 13:44:23 whatever does the job, use it 2014-01-13 13:44:57 I flipped a coin, I will go with lighttpd 2014-01-13 13:45:31 :) 2014-01-13 13:46:00 head: lighttpd, tail: nginx, edge: apache 2014-01-13 13:46:47 Is there already the jvm for musl out? 2014-01-13 13:47:24 http://sprunge.us/CDNS <- openjdk6 is on the list :( 2014-01-13 13:51:24 http://alpinelinux.org/forum/general-discussion/new-alpine-lots-questions 2014-01-13 13:51:35 http://alpinelinux.org/forum/installation/installing-alpine-272-jetway-nf9i-2550-system 2014-01-13 13:53:01 @ guest additions: clicking "mount guest cd" makes it available though /dev/sr0 2014-01-13 13:53:58 metasploit – I'll add it to my todo, but I'm bit busy with earning money now, and unfortunately it doesn't include Alpine hacking 2014-01-13 13:54:41 and the users shell can be set with chsh or editing /etc/passwd by hand 2014-01-13 13:54:56 Do you have a forum account? 2014-01-13 13:55:01 Or should I post it? 2014-01-13 13:55:28 I do, but feel free to post it 2014-01-13 13:56:04 and about the second thread, tell him to report the issue on bugtracker 2014-01-13 13:58:03 Ok 2014-01-13 14:02:08 i have no strong feeling for either nginx or lighttpd 2014-01-13 14:02:36 i think we use lighttpd for serving pages, the mirrors etc 2014-01-13 14:02:51 nginx for reverse proxy 2014-01-13 14:03:00 so yeah, depends on the job 2014-01-13 14:03:10 i kinda like uwsgi too, for web apps 2014-01-13 14:03:16 but havent used it too much 2014-01-13 14:03:54 jvm for musl is not a prio currently, but my guess is that we will get it eventually 2014-01-13 14:04:37 shafire: thanks alot for helping with forums 2014-01-13 14:04:49 no eventually, there will be no alpine 3.0 without openjdk 7 :p 2014-01-13 14:16:10 www.a.o drupal upgrade... press enter to continiu 2014-01-13 14:20:05 http://www.youtube.com/watch?v=aBd-FXmUWyE (do you like such music?) 2014-01-13 14:20:18 not the dancing boy :D 2014-01-13 14:22:53 shafire: does www.a.o work for you? 2014-01-13 14:23:02 no 2014-01-13 14:23:06 a lot errors 2014-01-13 14:23:10 :) 2014-01-13 14:23:15 i get some weird crap 2014-01-13 14:23:19 ��;ySۺ��3s�������l$�@�@�Z���r�NF�G������@�k�>��H�g��3on�ؖtt���ٴt��������>�rש�l�r�gok 2014-01-13 14:23:21 :D 2014-01-13 14:23:39 index.php works ok 2014-01-13 14:23:47 http://alpinelinux.org/index.php 2014-01-13 14:23:47 yeah 2014-01-13 14:24:00 http://alpinelinux.org/forum <- not 2014-01-13 14:24:09 blog too 2014-01-13 14:24:27 looks like url rewrite is broken or something 2014-01-13 14:25:54 ok back to life 2014-01-13 14:26:21 what was it? 2014-01-13 14:26:25 seems there were some updates pending. this drupal update had more to it then just bug fixing. 2014-01-13 14:26:50 you should always run http://alpinelinux.org/update.php 2014-01-13 14:27:19 sure, but this was the first time i couldnt even load the index :) 2014-01-13 14:27:21 delete all txts too http://alpinelinux.org/README.txt 2014-01-13 14:28:26 also this one? http://alpinelinux.org/ROOTPASSWORD.txt ? 2014-01-13 14:28:41 no, you should leave it 2014-01-13 14:31:26 [13/Jan/2014:15:28:29 +0100] "GET /ROOTPASSWORD.txt HTTP/1.0" 404 196 "http://webchat.freenode.net/" "Mozilla/5.0 (Macintosh; Intel Mac OS X 10.9; rv:26.0) Gecko/20100101 Firefox/26.0" 2014-01-13 14:31:30 :P 2014-01-13 14:31:36 heh 2014-01-13 14:31:42 maybe we should addsomething there 2014-01-13 14:31:53 do you see also my ip? :D 2014-01-13 14:32:31 echo "oh no... they found it" > ROOTPASSWORD.txt 2014-01-13 14:32:56 echo "123456" > ROOTPASSWORD.txt 2014-01-13 14:33:34 ncopa: i added it ;-) 2014-01-13 14:34:00 clandmeter: whats my ip? 2014-01-13 14:34:07 shafire: i dont know 2014-01-13 14:34:25 was it ipv4 or ipv6? 2014-01-13 14:34:49 i need to configure lighty to add the correct ip to the log, but im lazy 2014-01-13 14:34:51 oh, alpine is only ipv4, then it was my vpn, ok 2014-01-13 15:26:08 did my darkhttpd patch reached the al-dev mail list? 2014-01-13 15:26:21 yes 2014-01-13 15:27:06 alacerda: is build() supposed to be empty? 2014-01-13 15:29:07 barthalion: I was following another package example.... But now that you pointed this i am not confident 2014-01-13 15:29:25 I am comparing now with openvpn package 2014-01-13 15:29:44 and i see that there are the ./configure and make options 2014-01-13 15:54:33 alacerda: I guess that you will need only mak 2014-01-13 15:54:36 argh 2014-01-13 15:54:37 make 2014-01-13 15:55:38 I am doing: make || return 1 \n install -c $pkgname /usr/bin/ 2014-01-13 15:56:03 make should be in build() 2014-01-13 15:56:07 install in package() 2014-01-13 15:56:09 yeah 2014-01-13 15:56:13 ahh ok 2014-01-13 16:03:45 barthalion: What may I be doing wrong? install: can't create '/usr/bin/darkhttpd': Permission denied 2014-01-13 16:08:12 you forgot $pkgdir 2014-01-13 16:08:59 alacerda: ↑ 2014-01-13 16:11:29 ncopa: watch out, its fragile ;) 2014-01-13 16:11:52 yeah i know.. 2014-01-13 16:11:57 but its broken... 2014-01-13 16:12:08 they depend on non-existsin ruby-gems 2014-01-13 16:12:20 ah ok 2014-01-13 16:12:23 there are a handful left 2014-01-13 16:12:25 ships with ruby now 2014-01-13 16:12:49 grep ruby-gems testing/*/APKBUILD 2014-01-13 16:12:58 i gotta go now 2014-01-13 16:13:00 see u 2014-01-13 16:13:14 bye 2014-01-13 18:43:12 what hapend that some programs doesnt work with openRC scripts? http://pastebin.com/CV2eBNaz 2014-01-13 18:44:11 http://pastebin.com/bKhFbVpq 2014-01-13 18:48:17 it rises 2 instances of the same program and do not create pid file 2014-01-14 07:25:23 fabled: i cleaned up the ruby-gems mess ^^^ 2014-01-14 07:26:07 nice. thanks ncopa 2014-01-14 12:29:12 It seems that unstable directory was replaced by unmantainded: http://wiki.alpinelinux.org/wiki/Aports_tree 2014-01-14 12:29:14 ? 2014-01-14 12:29:15 is it? 2014-01-14 12:29:45 no 2014-01-14 12:29:56 well yes 2014-01-14 12:30:35 i actually got the blackmagic video grabber kernel module compile 2014-01-14 12:30:41 but i dont think i want maintain it 2014-01-14 12:31:00 also 2014-01-14 12:31:03 it is closed source 2014-01-14 12:31:10 so it should likely got to non-free instead 2014-01-14 13:43:06 files from v2.5.4-236-g0cf8b84 uploaded 2014-01-14 13:43:16 files from v2.5.4-236-g0cf8b84 uploaded 2014-01-14 13:49:15 files from v2.4.11-174-ga7ad4c1 uploaded 2014-01-14 14:03:29 files from v2.4.11-175-g566868b uploaded 2014-01-14 14:07:26 files from v2.5.4-237-ge8b94d0 uploaded 2014-01-14 14:07:33 files from v2.5.4-237-ge8b94d0 uploaded 2014-01-14 14:25:00 files from v2.5.4-238-g4dd35aa uploaded 2014-01-14 14:26:57 files from v2.5.4-238-g4dd35aa uploaded 2014-01-14 14:40:46 algitbot: build 2.7-stable 2014-01-14 14:41:09 that zeromq business have been garbage... 2014-01-14 14:56:45 files from v2.5.4-239-g946262a uploaded 2014-01-14 14:56:55 files from v2.5.4-239-g946262a uploaded 2014-01-14 14:59:16 files from v2.4.11-176-g25a0f4d uploaded 2014-01-14 15:00:54 Hi 2014-01-14 15:04:29 hi 2014-01-14 15:10:49 kaniini: was it you, who maintains the openjdk? 2014-01-14 15:17:41 shafire: check the first 2 lines of apkbuild :) 2014-01-14 15:18:16 it will most probably say timo teras aka fabled 2014-01-14 15:18:48 I read timo teras, but I thought, it was kaniini sorry 2014-01-14 15:18:56 but kaniini can not be it, because he was the US guy... 2014-01-14 15:21:49 files from v2.5.4-240-ge421253 uploaded 2014-01-14 15:22:11 files from v2.5.4-240-ge421253 uploaded 2014-01-14 15:29:08 files from v2.5.4-241-g3d38ed2 uploaded 2014-01-14 15:29:39 files from v2.4.11-177-g70463ed uploaded 2014-01-14 15:35:13 i htink we should ship our own install kernel 2014-01-14 18:30:07 did my patch for darkhttpd reached mail list? "adding openRC script" 2014-01-15 07:27:39 i will replace the debian's installkernel with a simplified version 2014-01-15 07:28:07 rather than installing /boot/vmlinuz--grsec it will simply install /boot/vmlinuz-grsec 2014-01-15 07:28:27 reason is that debian's installkernel creates symlinks 2014-01-15 07:28:38 but we want be able to mount /boot on fat for UEFI 2014-01-15 07:28:44 so we cannot use symlinks there 2014-01-15 07:29:03 it will also make the iso image more consistent 2014-01-15 08:26:44 got to love such error messages: configure: error: Could not find a version of the library! 2014-01-15 08:33:57 ncopa: what is drake? 2014-01-15 08:34:25 dunno, an initramfs generator? 2014-01-15 08:34:38 i dont know 2014-01-15 08:34:43 looks like some alternative for make 2014-01-15 08:34:56 http://sources.gentoo.org/cgi-bin/viewvc.cgi/gentoo-x86/media-video/mkvtoolnix/mkvtoolnix-6.6.0.ebuild?view=markup 2014-01-15 08:35:05 nah, i confused with dracut 2014-01-15 08:35:42 ./drake 2014-01-15 08:35:47 so it shipped with the sources 2014-01-15 08:36:18 oh 2014-01-15 08:36:31 lol i was looking for a dep 2014-01-15 08:37:15 omg it needs ruby to build... 2014-01-15 08:42:23 rnalrd: linux-virt-grsec and linux-grsec kernels cannot be installed at same time 2014-01-15 08:42:36 yep 2014-01-15 08:42:39 the both provides /lib/modules/-grsec 2014-01-15 08:42:50 do we want fix that? 2014-01-15 08:42:59 shouldnt be too hard 2014-01-15 08:43:43 i am hacking the installkernel script 2014-01-15 08:43:47 also kernel image name should change 2014-01-15 08:43:53 yes 2014-01-15 08:44:01 i am working on kernel image name 2014-01-15 08:44:24 vmlinuz- 2014-01-15 08:44:40 the idea was to use vmlinuz-$_flavor 2014-01-15 08:44:44 +1 2014-01-15 08:44:52 where $_flavor is what we set in APKBUILD 2014-01-15 08:45:42 however, installkernel only gets the full $kernelversion 2014-01-15 08:45:59 the version that includes it all. 2014-01-15 08:46:07 as in /lib/modules/$kernelversion 2014-01-15 08:46:56 it is constructed to: $pkgver-$pkgrel-$_flavor 2014-01-15 08:47:51 i was thinking of fishing out -[a-z]* 2014-01-15 08:48:11 vmlinuz-3.10.18-0-virt-grsec does not sounds bad 2014-01-15 08:48:19 but should it grab -grsec or -virt-grsec 2014-01-15 08:48:25 i suppose -virt-grsec 2014-01-15 08:48:34 we coudl also rename the flavor to -virtgrsec 2014-01-15 08:48:48 but i think -virt-grsec sounds better 2014-01-15 08:49:04 that's what debian does too iirc 2014-01-15 08:49:21 linux-image-3.11.0-14-generic on my Mint box 2014-01-15 08:49:37 and they make symlinks 2014-01-15 08:49:53 not here 2014-01-15 08:49:56 i'll call it vmlinuz-$_favor only 2014-01-15 08:50:02 lavor 2014-01-15 08:50:06 flavor 2014-01-15 08:50:17 oh maybe they removed the symlinking 2014-01-15 08:50:29 and depend on generated boot menu 2014-01-15 08:57:13 i think we will just remove the "-3.11.0-14" part anyways 2014-01-15 08:58:53 deprecating versioned kernels is step backward IMO 2014-01-15 08:59:16 how so? 2014-01-15 08:59:36 i think the only value they provide is the informative in boot menu 2014-01-15 08:59:55 ah, they are removed anyway on upgrade? 2014-01-15 08:59:59 so it doesn't matter then 2014-01-15 09:00:02 yes 2014-01-15 09:00:25 you cannot install 2 parallel versions of same pkgname 2014-01-15 09:00:47 i thought they would be useful if you built a homemade kernel 2014-01-15 09:00:50 iirc Debian leaves old kernel on upgrade 2014-01-15 09:00:56 yes they do 2014-01-15 09:01:03 we dont have that feature (yet) 2014-01-15 09:01:04 so if new package causes issues, you can easily use previous one 2014-01-15 09:01:29 how can I check, which kernel I use? 2014-01-15 09:01:36 uname -r 2014-01-15 09:01:54 but you want to remove it, or not? 2014-01-15 09:02:09 it will not be removed from uname -r 2014-01-15 09:02:09 ah, or just from /boot 2014-01-15 09:02:24 Ok, I get it 2014-01-15 09:02:25 only from /boot/vmlinuz- 2014-01-15 09:02:57 barthalion: if we want provide keep old in case disaster 2014-01-15 09:03:07 then we could probably do so with pre-upgrade script 2014-01-15 09:03:22 it would be great 2014-01-15 09:03:28 cp /boot/vmlinuz-$flavor /boot/vmlinuz-$flavor.old 2014-01-15 09:03:43 cp -r /lib/modules/something 2014-01-15 09:03:55 the /lib/modules will be a bit tricky though 2014-01-15 09:04:07 since it has to be /lib/modules/$(uname -r) 2014-01-15 09:04:19 we'd need to either add support for it in apk 2014-01-15 09:04:38 or do both pre-upgrade and post-upgrade 2014-01-15 09:04:48 where pre-upgrade renames to tempname 2014-01-15 09:04:54 and post-upgrade rename it back 2014-01-15 09:05:01 to prevent apk to actually remove them 2014-01-15 09:05:06 not that I had any issues, but the way we do it in Arch is idiotic in the best case 2014-01-15 09:07:41 how you do it in arch? 2014-01-15 09:08:13 i wonder if we should have a config option for it "keep-previous-kernel" or so 2014-01-15 09:09:17 we don't 2014-01-15 09:09:20 that's the point 2014-01-15 09:09:29 oh, like alpine then .) 2014-01-15 09:10:04 I don't expect it will change in Arch, but would be really nice in Alpine 2014-01-15 09:10:24 i think it would be nice if we had it as a config option 2014-01-15 09:10:28 keep prevous kernel 2014-01-15 09:10:39 we can also put autoremove in post_upgrade, to keep only N previous kernels 2014-01-15 09:10:56 i think keep 1 previous kernel will be enough for now 2014-01-15 09:11:08 surely will 2014-01-15 09:11:36 fabled: any ideas how to implement that? 2014-01-15 09:11:49 i think we have options: 2014-01-15 09:12:03 mmm 2014-01-15 09:12:10 1) implement suport for multiple versions installed of same pkg in parallel 2014-01-15 09:12:18 (like slot's in gentoo) 2014-01-15 09:12:39 2) we implement keep-old hack in apk 2014-01-15 09:12:58 3) we have pre/post-upgrade hacks 2014-01-15 09:13:08 option 3 is trivial to implement 2014-01-15 09:13:50 pre-upgrade: mv $current $current.old 2014-01-15 09:14:15 post-upgrade: mv /lib/modules/$previous.old /lib/modules/$previous 2014-01-15 09:15:07 i dont think we need keep backup of System.map* or config-* 2014-01-15 09:17:30 hm 2014-01-15 09:17:40 maybe symlink would be useful 2014-01-15 09:17:47 it needs to clean up previous-previous from /lib 2014-01-15 09:18:15 symlink could be used for that yes 2014-01-15 09:18:26 you can't in fat32 2014-01-15 09:18:38 but you can't do it on usb, anyway 2014-01-15 09:18:41 we only need fat for /boot 2014-01-15 09:18:55 the /lib will be on modloop 2014-01-15 09:19:22 so for /lib/modules we could have a /lib/modules/previous symlink 2014-01-15 09:20:04 from post-upgrade, purge what previous points to, and create a new symlink to the current 'old' 2014-01-15 09:22:00 i wonder if we should have a /etc/kernel/linux-grsec.conf or so 2014-01-15 09:22:04 wehre you can set keep-old 2014-01-15 09:22:16 so its optional to keep previous 2014-01-15 09:35:36 ok new installkernel appears to work 2014-01-15 09:35:48 and i think its compatible with update-extlinux 2014-01-15 09:35:53 it shoudl be 2014-01-15 09:36:09 i think I just push it 2014-01-15 10:52:08 does the kernel have all the virtio stuff enabled? 2014-01-15 10:58:11 shafire: i think so 2014-01-15 13:14:31 build64-2-4: rebuild 2014-01-15 13:14:54 build64-2-4: retry 2014-01-15 13:31:28 whoops 2014-01-15 13:31:40 new kernel does not work with current mkinitfs trigger 2014-01-15 13:37:34 files from v2.4.11-177-g70463ed uploaded 2014-01-15 13:37:54 files from v2.4.11-178-gee78239 uploaded 2014-01-15 13:39:11 files from v2.5.4-242-gcca7693 uploaded 2014-01-15 13:39:25 files from v2.5.4-242-gcca7693 uploaded 2014-01-15 13:47:38 files from v2.4.11-178-gee78239 uploaded 2014-01-15 15:20:33 wheee! \o/ 2014-01-15 15:20:44 my laptop now boots using gummiboot 2014-01-15 18:13:26 fabled: do you remember which paxctl flag was neeed for firefox? 2014-01-15 18:13:29 it segfaults for me 2014-01-15 18:13:39 i think it segfaults in mesa libs 2014-01-15 18:19:53 do you guys suggest running stable or edge? 2014-01-15 18:20:07 depends on what you are using it for 2014-01-15 18:20:32 ncopa_: I just made a forum post that describes what my planned usage will be. 2014-01-15 18:20:41 as server/router i'd use stable 2014-01-15 18:20:48 http://sprunge.us/MfRE 2014-01-15 18:20:52 i am fighting firefox... 2014-01-15 18:21:47 Great thanks. 2014-01-16 08:48:57 thanks for cleaning up the mirror:) 2014-01-16 08:53:47 http://alpinelinux.org/forum/general-discussion/suggestions-where-start 2014-01-16 08:53:50 so much text :( 2014-01-16 09:53:48 do we dare upgrade binutils? 2014-01-16 10:07:42 yeah, nothing is impossible ;) 2014-01-16 10:25:20 ncopa: how can i get $JOBS in apkbuild? 2014-01-16 10:26:03 i need to provide it to drake -j 2014-01-16 10:27:25 ${JOBS:-2} 2014-01-16 10:27:35 then you can set JOBS in /etc/abuild.conf 2014-01-16 10:27:39 or it will default to 2 2014-01-16 10:37:35 omg i used regular brackets... thats why it didnt work before. 2014-01-16 10:47:04 ncopa: http://sprunge.us/RPXj 2014-01-16 10:47:10 any idea why this happens? 2014-01-16 10:53:33 depends on what more is in the src/common/bswap.h file 2014-01-16 10:53:39 it is not happy with something there 2014-01-16 10:55:34 ncopa: http://sprunge.us/UNae 2014-01-16 11:03:59 maybe bswap_16 etc already is a defined macro? 2014-01-16 11:04:19 try do #ifndef bswap_16 ... #endif 2014-01-16 11:04:34 or maybe 2014-01-16 11:04:52 it could also be that uint16_t is undefined 2014-01-16 11:05:01 try add: #include 2014-01-16 11:05:11 i think it should be there anyways 2014-01-16 11:06:34 ok thx 2014-01-16 11:09:56 seems already defined, so ifidef fixes it. but needed to add it to all 3 functions. 2014-01-16 11:11:48 ah nice, so APKBUILD's are pretty much the same as PKGBUILD's in AUR :) 2014-01-16 11:12:02 alpine is modelled pretty closely after arch isnt it? 2014-01-16 11:12:43 kind of, but of course alpine is much better. 2014-01-16 11:13:15 I use manjaro as my desktop box, I am really impressed with Alpine. 2014-01-16 11:13:16 i think ncopa used parts of gentoo and arch 2014-01-16 11:13:43 clandmeter: did you see my forum post? 2014-01-16 11:13:51 edge226: not the same, but similiar 2014-01-16 11:13:58 there are some caveats you need to remember 2014-01-16 11:14:30 cool, I am definitely 'in' on this project now and will do what ever I can to help out :) 2014-01-16 11:15:26 edge226: dont say anything negative about arch when barthalion is around. 2014-01-16 11:16:19 clandmeter: I dont distro hate, I dont use arch myself because it has its issues, primarily my own laziness :P 2014-01-16 11:17:14 I use manjaro because I dont want to admin the box all the time during updates so it breaks. I know its my fault lol. 2014-01-16 11:19:04 I saw your long! forum post 2014-01-16 11:19:06 :D 2014-01-16 11:19:23 shafire: yeah I wanted to get it all out at once. 2014-01-16 11:21:30 Has someone fixed the lxc template? 2014-01-16 11:21:35 templates* 2014-01-16 11:23:46 clandmeter: c'mon, I'm far from being a fanboy :p 2014-01-16 11:24:03 :) 2014-01-16 11:24:16 i couldnt resist ;-) 2014-01-16 11:24:39 essentially we are brothers in the same stream even if manjaro and arch devs do not agree. We use the same software essentially :) 2014-01-16 11:24:42 he is a true alpine fanboy 2014-01-16 11:25:44 shafire: and about templates… no fwik, packages are still hanging on my todo 2014-01-16 11:25:58 Ok 2014-01-16 11:36:57 damn, i need to learn how to code C 2014-01-16 11:39:27 just a background on my experience beyond the post I made. I am rusty but I did take Computer Information Technology back in 2003-2005 where I learned C, C++, Java and Assembly ;) 2014-01-16 11:40:34 I have been looking for a distro to help out and Alpine may just be the one that gives me a reason to brush up on my languages :) 2014-01-16 11:41:23 can i open a ticket on your website? 2014-01-16 11:53:28 ncopa: looks like it has been defined in uclibc-dev. 2014-01-16 11:53:42 so i chanced it to: #if !defined(MTX_COMMON_BSWAP_H) || !defined(_BYTESWAP_H) 2014-01-16 11:54:06 not sure if thats correct, or i should specify uclibc directly? 2014-01-16 11:57:12 clandmeter: on mine? 2014-01-16 11:57:26 edge226: yes 2014-01-16 11:57:58 what do you mean open a ticket? its just a blog. 2014-01-16 11:59:29 If you would like it modified to reflect the existance of Alpine and for me to make some posts about alpine I can certainly do that. 2014-01-16 12:04:30 updating the Linux distro's menu to add alpine now. 2014-01-16 12:06:43 i think i need glasses 2014-01-16 12:07:18 all of a sudden i have problem to focus on the screen 2014-01-16 12:07:37 I need new one, so maybe we'll get discount :p 2014-01-16 12:11:48 clandmeter: no, do: #ifndef bswap_16 2014-01-16 12:12:00 around the bswap_16 func 2014-01-16 12:12:22 all 3 are already defined 2014-01-16 12:12:35 then do it on all 3 2014-01-16 12:12:54 i cannot do it in one time? 2014-01-16 12:13:02 #ifndef bswap_32 2014-01-16 12:13:04 not really 2014-01-16 12:13:10 menu on site updated to reflect existance of alpine with links to the official site, wiki, downloads and forums :) 2014-01-16 12:13:20 !defined(_BYTESWAP_H) is not really portable 2014-01-16 12:13:30 so it might break lets say glibc or musl 2014-01-16 12:13:46 ok 2014-01-16 12:14:03 or i can replace it with __uclibc__? 2014-01-16 12:14:47 ncopa: btw, new llvm/clang is out 2014-01-16 12:15:09 barthalion: i saw 2014-01-16 12:15:26 clandmeter: __any_sys__ is hackish 2014-01-16 12:15:39 ok 2014-01-16 12:15:55 if (feature) is better than if (platform) 2014-01-16 12:16:13 for that reason musl dont have any __musl__ 2014-01-16 12:22:27 I am not using glasses. I love my ortho-k contact lenses :-) 2014-01-16 12:23:20 I've got prisms, so I'm sentenced to glasses unfortunately 2014-01-16 12:24:06 what is prisms? 2014-01-16 12:24:44 https://en.wikipedia.org/wiki/Prism_correction 2014-01-16 12:27:48 barthalion: glasses make you look smarter anyways ;) 2014-01-16 12:29:30 barthalion: Ok I see 2014-01-16 12:29:38 ncopa: can i send this upstream? http://sprunge.us/OUAF 2014-01-16 12:32:35 looks good to me 2014-01-16 12:32:51 ok thx! 2014-01-16 12:33:33 [A 2014-01-16 12:33:50 oops, where did that come from.. 2014-01-16 13:24:37 barthalion: i was holding xf86-{video,input}-* til after mesa-10 and xorg-server-1.15 2014-01-16 13:24:45 they will need rebuild then 2014-01-16 13:24:54 but np 2014-01-16 13:26:36 whee \o/ 2014-01-16 13:26:47 i just got a contributor patch for abuild 2014-01-16 13:27:02 basically, add -dbg to subpackages 2014-01-16 13:27:11 and it will create a subpackage with debugging symbols 2014-01-16 13:27:13 ncopa: I hopefully will push new mesa today 2014-01-16 13:27:17 ncopa: and then new xorg 2014-01-16 13:27:51 I've also new libgcrypt in queue, and minor poppler release (without breakage this time) 2014-01-16 13:27:52 barthalion: iirc you'll need rebuild xf86-{video,input}-* after new xorg 2014-01-16 13:28:07 poppler without break.. incredible... 2014-01-16 13:28:10 :) 2014-01-16 13:28:31 don't worry, they released version with soname bump too :p 2014-01-16 13:28:53 bah 2014-01-16 13:29:19 firefox segfaults with musl libc in libgl from mesa 2014-01-16 13:30:06 this new abuild -dbg improvement will make those things *much* nicer 2014-01-16 13:31:39 btw, why musl requires config.sub update in some packages? 2014-01-16 13:36:31 ncopa: what? "abuild -dbg" is working now? 2014-01-16 13:37:59 StarWarsFan: you just add a subpackage="$pkgname-dbg" 2014-01-16 13:38:01 and thats it 2014-01-16 13:38:24 way cool! :-) 2014-01-16 13:38:27 if you have a crashing app, which has a -dbg subpackage 2014-01-16 13:38:35 you can just apk add thepackage-dbg 2014-01-16 13:38:46 ACTION still needs to dig deeper why quasselcore crashes... 2014-01-16 13:38:48 and you'll get a nice backtrace 2014-01-16 13:39:08 quasselcore is a candidate for this feature 2014-01-16 13:39:13 musl 2014-01-16 13:39:14 uclibc 2014-01-16 13:39:18 might be nice to have those too 2014-01-16 13:39:22 maybe glib 2014-01-16 13:40:06 its issue #626 2014-01-16 13:40:09 very nice. 2014-01-16 13:40:27 ack 2014-01-16 13:40:34 StarWarsFan: do you suggest quassel over weechat? 2014-01-16 13:41:46 ACTION is using quassel 2014-01-16 13:41:57 didn't know weechat 2014-01-16 13:43:29 I am thinking of scrapping weechat for quassel, both have nice network functionality and relay capabilities but I cant seem to get the weechat stuff to work over localhost. 2014-01-16 13:44:30 so is there something like a weechat-core? 2014-01-16 13:44:56 because that's the quassel killer feature: 2014-01-16 13:45:08 i'm running quassel-core at my server @home 2014-01-16 13:45:20 so i'm always online and did not miss anything 2014-01-16 13:45:26 tmux + irssi works too... 2014-01-16 13:45:38 on every device i have the quassel client 2014-01-16 13:45:44 to connect to my core instance 2014-01-16 13:45:50 StarWarsFan: in weechat you create a relay and then can access it via other programs that are only clients. 2014-01-16 13:46:07 you can see where you leave last time? 2014-01-16 13:46:25 StarWarsFan: you're supposed to. I've not gotten it working properly. 2014-01-16 13:46:39 and nobody in the weechat channel has any answers for me as to why its not working. 2014-01-16 13:47:00 so you need to use quassel ;-) 2014-01-16 13:47:09 but the problem is: 2014-01-16 13:47:23 the alpine version crashes right after a client has connected 2014-01-16 13:47:34 that's why i need the dbg feature 2014-01-16 13:47:43 because the problem is on some other dependency 2014-01-16 13:47:52 already tried to build a debug version of quassel 2014-01-16 13:48:06 but that leads not to the problem... 2014-01-16 13:48:37 StarWarsFan: thats what I am thinking but I wanted something with a base that is CLI. 2014-01-16 13:49:04 the overhead comparison is pretty crazy but I guess it doesnt really matter too much. 2014-01-16 13:49:04 it is 2014-01-16 13:49:18 there is the core and the client 2014-01-16 13:49:31 interesting. When I looked it up on manjaro the core was still GUI 2014-01-16 13:49:51 sure, possible 2014-01-16 13:49:57 i've setup three packages on alpine 2014-01-16 13:50:03 or better: ncopa did ;-) 2014-01-16 13:50:48 nice, that means I can ditch a seperate box for IRC and have alpine run everything :) 2014-01-16 13:52:36 exactly 2014-01-16 13:52:43 see edge/testing 2014-01-16 13:53:42 alpine really is amazing, I went from potentially needing 3-4 VM's to just Alpine for everything :) 2014-01-16 13:55:49 now I just need to take the time to set it all up. I think Ill set up all the server stuff and then get the firewalling going. 2014-01-16 13:57:29 I think I can get chromebook connecting to openvpn as well :) 2014-01-16 13:58:26 Through my experience with my chromebook it seems a bunch of people in the Manjaro community want them now lol. 2014-01-16 14:00:42 because I dont know of any other small notebook you can get for $200 and upgrade the ram to 6GB and the SSD to 120GB at a total cost of $350 with even similar specs. 2014-01-16 14:22:09 we are climbing on distrowatch 2014-01-16 14:22:38 i think if we had some easy desktop install it would go even higher 2014-01-16 14:22:45 like what manjaro did with arch 2014-01-16 14:31:20 lets finish musl first :) 2014-01-16 14:43:04 edge226: what was your site? 2014-01-16 14:59:44 alpinelinux has a twitter account? 2014-01-16 15:12:16 ACTION hates twitter personally 2014-01-16 15:13:11 alacerda: what would we use it for? 2014-01-16 15:13:40 ACTION loves twitter 2014-01-16 15:18:27 ACTION agrees with rnalrd, how can anybody hate twitter? 2014-01-16 15:40:33 clandmeter: To promote al. Send news like new releases, bug fix, new packages and things like that 2014-01-16 15:40:47 new articles... 2014-01-16 15:41:42 I follow securitytube, clavis security and others... I think it is useful 2014-01-16 15:41:50 alacerda: nice, we are still looking for a marketing manager. are you interested in the job? 2014-01-16 15:41:56 ;-) 2014-01-16 15:42:31 It would be a pleasure. :) 2014-01-16 15:43:39 when do I start ;) 2014-01-16 15:47:10 another question: which tool/bot is used here to create links if one enters an issue number? 2014-01-16 15:47:23 seriously, i think alpine can use somebody that promotes it better then we do now. 2014-01-16 15:47:50 like the response to this: 2014-01-16 15:47:56 foo #123 bar 2014-01-16 15:47:56 #1 2014-01-16 15:48:18 hey, it doesnt listen to me anymore. 2014-01-16 15:48:23 he got tired. 2014-01-16 15:48:48 StarWarsFan: ncopa configed it. 2014-01-16 15:49:35 clandmeter: I can add my hand to the promote work. That is what I am doing in my portuguese blog. 2014-01-16 15:50:15 I dont think i can do it alone. But I can help 2014-01-16 15:50:34 alacerda: ok, if you want to sign-up for those services, i think you should request an alpine email address 2014-01-16 15:50:39 or use the one from infra 2014-01-16 15:50:40 ncopa: we can just spamclick distrowatch with curl 2014-01-16 15:50:56 ncopa: distrowatch rank is not indicator at all 2014-01-16 15:51:26 jbilyk can help you setup an alpine email address. 2014-01-16 15:51:57 jbilyk _0/ 2014-01-16 15:52:37 if you email jbilyk@alpinelinux.org, I'll get to them when I find time :) 2014-01-16 15:54:00 jbilyk: maybe create social@a.o ? 2014-01-16 15:54:25 clandmeter: right now, they're forwarders to an email address of your choice 2014-01-16 15:54:31 where would it forward to? 2014-01-16 15:54:37 to alacerda 2014-01-16 15:54:38 :) 2014-01-16 15:54:50 if he wants it, send in the request 2014-01-16 15:54:59 ACTION doesn't care, but ncopa usually has final say :) 2014-01-16 15:55:03 I'm writing the e-mail to you jbilyk 2014-01-16 15:55:11 ncopa doesnt respond so he doesnt count. 2014-01-16 15:55:23 ncopa's at end of work day, so not surprising :) 2014-01-16 15:55:39 jbilyk's also at work, trying to not blow up stuff while chatting :D 2014-01-16 15:55:49 *g* 2014-01-16 15:56:01 for such services itrs best to use a.o address. if anything ever happends, we can always get access to the account. 2014-01-16 15:58:01 StarWarsFan: i think we use http://git.alpinelinux.org/cgit/sircbot with a hook 2014-01-16 15:58:43 probably some way to get json/xml responce from redmine by providing the bug id. 2014-01-16 15:58:50 I sent you an e-mail jbilyk 2014-01-16 15:59:32 clandmeter: thx, will take a look... 2014-01-16 15:59:40 http://bugs.alpinelinux.org/issues/296.xml 2014-01-16 15:59:54 http://bugs.alpinelinux.org/issues/296.json 2014-01-16 15:59:55 alacerda: ok, will get to it soonish 2014-01-16 16:00:05 thanks jbilyk 2014-01-16 16:00:10 yeah we use sircbot 2014-01-16 16:00:14 its a bit buggy 2014-01-16 16:01:57 barthalion: re distrowatch, i know its not a good indicator, but getting high up there has some "marketing" value 2014-01-16 16:04:38 ACTION will be right back 2014-01-16 16:09:49 ACTION is back 2014-01-16 16:17:25 I want a alpine email too :D 2014-01-16 16:17:29 an* 2014-01-16 16:17:49 But I am not that active :S 2014-01-16 16:18:56 I hadn't al.org address for 1,5 years, it's possible to live without it 2014-01-16 16:19:06 and I rarely use it anyway 2014-01-16 16:19:25 its just prestige 2014-01-16 16:20:00 filling out a student form application with @a.o :P 2014-01-16 16:21:12 and getting more spam :p 2014-01-16 16:24:59 ncopa: makepkg is guaranteed to start in $srcdir since some time, can we get something similiar in abuild? 2014-01-16 17:00:15 have a nice evening bye 2014-01-17 08:14:12 barthalion: how do you mean? 2014-01-17 08:14:34 that when build() is called the cwd is "$srcdir"? 2014-01-17 08:31:55 ncopa: if i have a unresolved symbol when mixing repos, how can i find out which pkg is causing it? 2014-01-17 08:34:28 you mean what package that provides the symbol? 2014-01-17 08:34:42 or which file that needs a missing symbol? 2014-01-17 08:34:48 i guess some pkg is updated 2014-01-17 08:34:57 in edge which has the new symbol 2014-01-17 08:35:16 _ZN11libmatroska13KaxCodecDelay10ClassInfosE 2014-01-17 08:35:35 sounds like matroska c++ 2014-01-17 08:36:18 i guess i can add that one from edge, and it should (hopefully) work 2014-01-17 08:37:56 works :) 2014-01-17 08:38:33 otherwise, nm -D /path/to/file 2014-01-17 08:38:37 you cannot make apk smarter to do that automatically? :) 2014-01-17 08:38:51 we dont store symbol information in index 2014-01-17 08:39:02 so apk cannot tell what package provides/needs what symbol 2014-01-17 08:39:15 i guess thats a huge db 2014-01-17 08:39:29 nm -D will show what symbol is provided 2014-01-17 08:39:34 and which is needed 2014-01-17 08:40:02 T means its proviced 2014-01-17 08:40:19 U means its needed 2014-01-17 08:41:00 T 2014-01-17 08:41:00 Global text symbol. 2014-01-17 08:41:06 U 2014-01-17 08:41:07 Undefined symbol. 2014-01-17 12:30:05 hi 2014-01-17 12:54:39 hi shafire 0/ 2014-01-17 12:54:52 alacerda: :) 2014-01-17 12:57:28 ok i have some fun going on with lennart... 2014-01-17 12:57:34 https://bugs.freedesktop.org/show_bug.cgi?id=73729 2014-01-17 12:59:08 ouch 2014-01-17 13:04:04 that says actually alot 2014-01-17 13:13:51 Is someone good at webdesign? 2014-01-17 13:14:16 I would like to have a (responsive modern) new alpine design for version 3 2014-01-17 13:14:37 if you don't mind 2014-01-17 13:15:02 version 3 of what? 2014-01-17 13:15:15 alpine 2014-01-17 13:16:22 the alpinelinux.a.o has ver static content 2014-01-17 13:16:25 very* 2014-01-17 13:16:47 i have been thinking of making a site from asciidoc 2014-01-17 13:17:07 and keep the asciidoc sources in git 2014-01-17 13:17:21 so people who wants contribute can git send-email 2014-01-17 13:17:34 great idea ncopa 2014-01-17 13:17:53 nevertheless, a new cool design is required :) 2014-01-17 13:18:21 ncopa: also the wiki? 2014-01-17 13:18:36 shafire: you mean design of www.a.o? 2014-01-17 13:18:43 clandmeter: yes 2014-01-17 13:19:16 you dont like my design? 2014-01-17 13:19:20 :( 2014-01-17 13:19:37 i think this concept is interesting: http://suckless.org/wiki 2014-01-17 13:19:41 it's good, but outdated 2014-01-17 13:19:47 jk, i hope you can make it better. 2014-01-17 13:19:57 i have been thinking of new logo... 2014-01-17 13:20:26 so you want to stop with drupal or just create a new theme? 2014-01-17 13:21:32 looks like so at moment, or? asciidoc + new theme? 2014-01-17 13:22:07 replace forum by mailing list consequently 2014-01-17 13:23:03 and we have a package browser 2014-01-17 13:23:52 I dont see that advantage of using git to manage our pages. people who dont know git cannot submit changes. 2014-01-17 13:24:14 or we asume every infra memeber to know git. 2014-01-17 13:25:11 I think it's easy to leap the git hurdle 2014-01-17 13:25:59 twitter @alpinelinux 2014-01-17 13:26:50 who is responsible for twitter ? 2014-01-17 13:27:08 alacerda handles it 2014-01-17 13:27:33 what's alpine target group? 2014-01-17 13:27:56 I thought embedded and server, but people are using it as desktop recently :D 2014-01-17 13:28:14 if you add alpine-desktop... 2014-01-17 13:46:32 about changing the logo... it'd be the third logo. I dont think it is a good idea changing the logo again 2014-01-17 13:48:05 not chaning, more improving 2014-01-17 13:48:27 http://www.coca-cola.co.uk/125/history-of-coca-cola-logo.html 2014-01-17 13:49:39 http://thumbnails.visually.netdna-cdn.com/the-history-of-logo-design_51fa58c8f3881_w1500.jpg 2014-01-17 13:49:49 look at mercedes-benz (daimler) 2014-01-17 13:55:56 we won't replace drupal unless someone actually *does* it 2014-01-17 13:56:57 the idea was to reduce workload of constant drupal upgrades 2014-01-17 13:57:16 generated static pages requires less maintenance 2014-01-17 13:57:53 it would mean we need a package browser rewrite and a new forum so its not gonna happen anytime soon 2014-01-17 14:21:29 why not replacing the forum by the mailing list? 2014-01-17 14:46:59 What you think of we add a "Follow us on Twitter" on main al.o page? 2014-01-17 14:48:02 shafire: what kind of logic is that? 2014-01-17 14:49:41 what do you mean? 2014-01-17 14:50:18 i mean, what do you mean :) 2014-01-17 14:50:50 forum and ml are 2 diffent things. 2014-01-17 14:51:26 you are right 2014-01-17 14:51:45 both have different users, ppl on ml prefer not be on the forum and vice versa. 2014-01-17 15:06:28 shafire, what kind of msgs are planned for twitter ? 2014-01-17 15:06:57 alacerda: ^^^ 2014-01-17 15:07:10 thanks 2014-01-17 15:12:15 ircbot, seems nice, http://www.energymech.net/ 2014-01-17 15:21:33 gotta go 2014-01-17 15:33:17 ncopa: didnt we agree an alpine email address needs some didication, or we start to charge 10$ per month? 2014-01-17 15:33:57 I thought, it was 10 UAH?! 2014-01-17 15:33:57 dedication even 2014-01-17 15:36:42 for while I plan twitte new releases, new packages added ( maybe not *all* new ones ), package upgrades.... But if someone has other ideas I am interested ... 2014-01-17 15:37:06 some articles added to HowTo wiki page 2014-01-17 15:37:46 ideas can be sent to social@al.o 2014-01-17 15:49:40 new packages maybe not that good 2014-01-17 15:49:49 what if they do a massive pkg bump 2014-01-17 15:50:56 who are "they"? 2014-01-17 15:51:07 i mean we 2014-01-17 15:51:15 doing to many windows at the same time. 2014-01-17 15:52:58 good call 2014-01-17 16:01:49 ACTION suggests just new releases at first, and possibly major new features in main 2014-01-17 16:01:59 otherwise people will just hear noise 2014-01-17 16:05:56 AL has msg.alpinelinux.org, I would suggest adding #wiki/ #pkg/ #forum/ #ml/ 2014-01-17 16:06:19 a script can be used to bot between twitter and msg.a.o 2014-01-17 16:07:11 new topic/articles title from suggested topic above can be pushed to twitter 2014-01-17 16:07:36 i think we could create a list of major packages and verify the version has been updated (not just the pkgrel) and only publish that. 2014-01-17 16:08:26 hmm.. I think the script msg-->twitter could do that 2014-01-17 16:09:34 it just tracks 'build/build-edge-x86 building testing/mkvtoolnix-6.7.0-r0', removes rX and keeps a state 2014-01-17 16:09:58 than compares and pushes to Twitter 2014-01-17 16:11:09 thus making the mosquitto msg.a.o robust and things centered around it 2014-01-17 16:16:56 and yes missed, title from bugs.a.o #bugs 2014-01-17 16:19:59 vkrishn: sound ok 2014-01-17 16:20:22 this is something like http://www.alert-project.eu/content/alert-conceptual-architecture-0 2014-01-17 16:20:28 pg. 14 2014-01-17 16:20:46 of ALERT Conceptual Architecture.pdf 2014-01-17 16:21:19 except that mqtt method is much easier an lighter, though not as feature full 2014-01-17 16:22:14 to be honest, i have looked in the recent msging changes. 2014-01-17 16:22:30 vkrishn: seems you are an expert compared to me. 2014-01-17 16:44:42 ncopa, re: 'that zeromq business have been garbage... ' ... I could not follow, maybe I missed something ? 2014-01-17 16:44:56 something to do with current msg.a.o implementation? 2014-01-17 16:57:30 vkrishn: is there any doc on how to use msg.a.o? 2014-01-17 17:04:52 clandmeter, I currently monitor via clog 2014-01-17 17:04:59 https://dpaste.de/K1tr 2014-01-17 17:05:21 i made a typo, i didn't look into new msg framework :) 2014-01-17 17:05:21 its colored lines and text making it easy know what is going 2014-01-17 17:05:41 ok 2014-01-17 17:05:42 install current mosquitto, it has mosquitto_sub 2014-01-17 17:05:49 client 2014-01-17 17:06:27 i will but not now 2014-01-17 17:06:30 its weekend 2014-01-17 17:06:44 there are 2 new nice log analyzer in testing, not tested yet 2014-01-17 17:06:56 going to go afk. 2014-01-17 17:07:02 testing/lnav and testing/goaccess: 2014-01-17 17:07:20 gotta go too 2014-01-17 17:07:30 have a nice weekend 2014-01-17 23:35:29 rnalrd: have you checked out tweetdeck? 2014-01-18 01:48:40 ACTION wonders why the USB creation article does not cover the most basic and probably least failure rate option. dd. 2014-01-18 02:43:11 just a FYI, the confirmation email is not sending properly for the wiki. 2014-01-18 17:35:37 algitbot: build master 2014-01-18 23:02:36 anyone have any idea why it wont send me a confirmation email on the wiki sign up? 2014-01-19 02:23:57 edge226: someone else just reported that to me yesterday - i haven't had time to check out the email relay on that box yet 2014-01-19 02:24:03 sorry about that 2014-01-19 02:26:47 jbilyk__: ah kk. 2014-01-19 16:15:53 ping clandmeter 2014-01-19 16:42:04 I guess I screwed up buildbots with new libgcrypt 2014-01-20 08:23:45 we don't have libatomic_ops? 2014-01-20 08:24:32 ncopa, gc build fails ^^^ 2014-01-20 09:33:37 ncopa: what does msg.a.o do? 2014-01-20 12:39:30 rnalrd: god, why they were dead when I pushed it 2014-01-20 12:39:40 rnalrd: I'll fix it in the couple of minutes 2014-01-20 12:39:53 rnalrd: I forget to update gc deps 2014-01-20 12:43:35 the fun begins 2014-01-20 12:45:04 rnalrd: I guess it will take a while, I pushed gigantic rebuild 2014-01-20 12:48:46 barthalion, tnx 2014-01-20 12:51:39 ncopa: we really need a way to peek what's happening behind algitbot 2014-01-20 12:52:04 ncopa: less noise is cool, but two days ago buildbots hung and I had no idea what was wrong 2014-01-20 12:52:20 ncopa: and now suddenly everything is ok, though I don't know what they're doing right now 2014-01-20 13:13:39 barthalion: isnt that what msg.a.o is for? 2014-01-20 13:17:22 alacerda, do have a look at https://pypi.python.org/pypi/python-twitter and https://github.com/jpmens/twitter2mqtt might be useful 2014-01-20 13:17:56 though we may want something like mqtt2twitter 2014-01-20 13:18:10 I got a goot result using supertweet 2014-01-20 13:18:20 but i will check these options too 2014-01-20 13:18:21 url ? 2014-01-20 13:18:23 thanks 2014-01-20 13:18:51 supertweet.net 2014-01-20 13:26:01 clandmeter: why I haven't heard about it? 2014-01-20 13:26:58 clandmeter: and how can I use it? while "welcome to nginx" is quite welcoming, it doesn't make me happier 2014-01-20 13:27:26 barthalion: i think its not yet ready 2014-01-20 13:27:53 although msg.a.o is 2014-01-20 13:28:03 linux 3.13 out with nftables 2014-01-20 13:28:29 you need to push info to it, so others can subscribe to it and see it. 2014-01-20 13:29:26 hm, right, I guess ncopa wrote about it here some time ago 2014-01-20 13:29:34 still, I like IRC-based development 2014-01-20 13:30:03 currently you can subscribe to it to see commits which have been pushed to aports. 2014-01-20 13:30:15 so having something like "algitbot: sudo gimme the logs now" would be better from my point of view 2014-01-20 13:30:23 barthalion, http://www.irclogger.com/.alpine-devel/2013-11-29#1385732668 2014-01-20 13:30:39 mosquitto_sub -h msg.alpinelinux.org -t 'git/aports/master' 2014-01-20 13:30:44 well, still 2014-01-20 13:30:52 buildbots are dead again 2014-01-20 13:30:55 algitbot: build master 2014-01-20 13:31:45 one benefit with mqtt is it can host lots of other msgs that may a little too much or irc 2014-01-20 13:32:12 may be a little too much for irc 2014-01-20 13:32:28 i suggested to use a complete solution to do our building infra, but like always ncopa wants to do as much by hand as possible :) 2014-01-20 13:32:35 I don't need to print logs directly here 2014-01-20 13:32:54 I just want to view the current one 2014-01-20 13:33:11 while buildbots are running, not post mortem 2014-01-20 13:33:19 you could 2014-01-20 13:33:51 but I'm too lazy to reverse engineer the url 2014-01-20 13:34:28 if we have the build log in msg.a.o we could connect to it after we push to aports 2014-01-20 13:34:36 that will give you the log of the current build 2014-01-20 13:37:40 it would be appreciated if someone could say why our buildbots are drinking tee again 2014-01-20 13:37:58 it's not five o'clock 2014-01-20 13:38:26 i dont have access i think 2014-01-20 13:38:57 fabled and ncopa are away 2014-01-20 13:39:14 ncopa never adds my key to new servers. 2014-01-20 13:39:21 he doesnt trust me ;-) 2014-01-20 13:39:33 but we have an alternative 2014-01-20 13:39:38 jbilyk jbilyk__ 2014-01-20 13:40:01 nobody home :P 2014-01-20 13:40:11 home is not needed 2014-01-20 13:40:26 please leave a msg after the beep....BEEP 2014-01-20 13:40:30 jbilyk: haha 2014-01-20 13:40:41 jbilyk: got access to bld2? 2014-01-20 13:40:45 jbilyk: can you take a look what's wrong with build bots? 2014-01-20 13:41:07 ncopa and fabled have done the work on them - I'm not sure how they're currently setup to be honest 2014-01-20 13:41:19 clandmeter, I think its still building 2014-01-20 13:41:28 last time I played around trying to see how they worked, I busted edge/x86 building :) 2014-01-20 13:41:31 log will be fine 2014-01-20 13:41:43 last msg was ~2min back 2014-01-20 13:41:59 barthalion, won't this http://dev.alpinelinux.org/buildlogs/ help 2014-01-20 13:42:01 ? 2014-01-20 13:42:21 kinda old 2014-01-20 13:43:02 wee 2014-01-20 13:43:13 this is what I wanted 2014-01-20 13:44:02 a build failure? 2014-01-20 13:44:20 i added tail some time ago 2014-01-20 13:44:37 but ncopa keeps changing settings so the logs break 2014-01-20 13:48:10 puf 2014-01-20 13:48:12 alacerda, api.supertweet.net may be back compat with twitter api ? 2014-01-20 14:03:27 http://www.supertweet.net/about/api looks neat 2014-01-20 14:26:41 barthalion, have you subscribed to msg.a.o ? 2014-01-20 14:27:08 I have IRC client for a reason :p 2014-01-20 14:27:34 on msg.a.o after every git pull and build start it emits "building v131211-553-g972aeb1" 2014-01-20 14:28:03 http://bld1.alpinelinux.org/buildlogs/build-edge-x86.v131211-553-g972aeb1.log 2014-01-20 14:28:13 this translates to ^^^ 2014-01-20 14:28:38 so even if is does not fail it can be tracked 2014-01-20 14:29:10 the logs name keep changing based on git commit and tag 2014-01-20 14:30:44 I can use git describe as well 2014-01-20 14:34:23 so it is not a feature for me 2014-01-20 14:42:20 :) 2014-01-20 14:45:54 so you fetch the logs or ssh into server to view it ? 2014-01-20 14:48:35 http://bld[1|2].al.o/buildlogs/$builder.$(git describe).log 2014-01-20 14:50:59 neat, add watch -n 5 wget -c to it and you can then tail it 2014-01-20 15:06:56 gotta go 2014-01-20 16:12:44 jbilyk: there are still issues with the authentication on the wiki. It successfully sent the confirmation now but I am still not authenticated. I would also prefer being able to uncheck having others email me... 2014-01-20 16:39:52 edge226: mind opening a ticket on bugs.a.o about it pls? 2014-01-20 16:40:05 nangel just fixed one issue, but might've introduced another one :/ 2014-01-20 17:03:59 jbilyk: I do not mind, however the question is where? 2014-01-20 18:06:34 edge226: http://bugs.alpinelinux.org, in the Alpine Linux project 2014-01-20 18:39:40 why cant we set it up so you sign up for 1 thing instead of having to sign up for a bunch of things... 2014-01-20 18:44:09 jbilyk: and how do I actually submit a bug with this tracker... that doesnt seem to exist. Anywhere. 2014-01-20 18:46:16 edge226: http://bugs.alpinelinux.org/projects/alpine/issues 2014-01-20 19:07:06 jbilyk: why are there no links to this stuff? 2014-01-20 19:07:46 edge226: on www.alpinelinux.org, at the bottom, under "Services" on the main page, "Bug Tracker" is the link you're looking for ;) 2014-01-20 19:10:13 well thats a really odd place to put it if you want people to actually see it :/ 2014-01-20 19:11:11 that's why I didn't design it...ncopa and clandmeter did the www.a.o frontpage 2014-01-20 19:12:01 I am just saying, I literally would have never looked there for the bug tracker. 2014-01-21 07:57:00 moin 2014-01-21 07:57:05 :-( 2014-01-21 07:57:20 ncopa: we still got some problem with current alpine release 2014-01-21 07:57:31 now kernel panic on virtual box 2014-01-21 07:57:39 see http://download.eisfair.org/installer-kernel-error.webm 2014-01-21 07:57:48 the iso tries to boot 2014-01-21 07:57:50 which fails 2014-01-21 07:57:53 but: 2014-01-21 07:58:18 after around 20 turns, it boot but fails with a kernel panic 2014-01-21 07:58:28 very strange... 2014-01-21 08:01:11 it fails to find the kernel 2014-01-21 08:01:19 morning StarWarsFan 2014-01-21 08:01:27 hi 2014-01-21 08:01:35 problem is that syslinunx.cfg does not find the kernel 2014-01-21 08:02:27 mh 2014-01-21 08:02:29 hm 2014-01-21 08:02:30 maybe not 2014-01-21 08:02:37 it does find the /boot/grsec 2014-01-21 08:02:43 and grsec.gz 2014-01-21 08:02:53 but says invalid argument 2014-01-21 08:03:09 very strange... 2014-01-21 08:03:10 is your syslinux.cfg modified? 2014-01-21 08:06:15 uh, just a second, need to ask this because i did not build the iso... 2014-01-21 08:10:39 until i got the answer: if you want to try out "eisfair-ng", which is based on alpine: 2014-01-21 08:11:26 see https://ssl.nettworks.org/wiki/display/eng/Download 2014-01-21 08:14:59 we modified /usr/share/mkinitfs/initramfs-init to add some more packages 2014-01-21 08:57:55 hm, brb... 2014-01-21 09:11:57 ncopa: syslinux is not modified 2014-01-21 09:12:06 Original alpine 2014-01-21 09:13:31 ok 2014-01-21 09:13:42 i have a win8.1 box i can test on 2014-01-21 09:13:48 will check it up soonish 2014-01-21 09:24:26 Cool, thx 2014-01-21 15:00:44 i am seriously disapointed in win 8.1 2014-01-21 15:00:56 it broke web browsing in firefox and IE 2014-01-21 15:00:59 o 2014-01-21 15:01:06 dns resolution works from cmd.exe 2014-01-21 15:01:13 you use the app or the desktop application? 2014-01-21 15:01:14 ping works 2014-01-21 15:01:41 i used windows 8.1, the operating system from microsoft 2014-01-21 15:01:51 ah of IE 2014-01-21 15:01:55 the desktop application 2014-01-21 15:02:12 cause I don't know why they have the windows store app that makes it full screen 2014-01-21 15:02:15 so annoying 2014-01-21 15:02:41 yeah... kinda confusing. 2 different systems 2014-01-21 15:02:57 its not a tablet and not a desktop 2014-01-21 15:03:25 how can i open IE as an app? 2014-01-21 15:03:58 it gives me the desktop app when using IE from the "start menu" 2014-01-21 15:05:43 from the windows store, originally they had it, dunno if that changed 2014-01-21 15:05:55 maybe when you goto the metro screen and open it from there 2014-01-21 15:05:55 i dont think i can get windows store either 2014-01-21 15:06:10 it just hangs 2014-01-21 15:06:18 i think its due to browsing is broke 2014-01-21 15:06:37 windows store does take a while to load 2014-01-21 15:07:35 we arent able to connect to the Store.... 2014-01-21 15:07:39 http://community.spiceworks.com/how_to/show/16215-how-to-re-enable-the-internet-explorer-10-app-in-windows-8 2014-01-21 15:08:06 any reason your messing with win8.1? 2014-01-21 15:09:41 i wanted help StarWarsFan with his virtualbox issue 2014-01-21 15:09:45 my new laptop is dualbooting 2014-01-21 15:10:07 thanks for the tip, it needs to be the default 2014-01-21 15:10:09 thats why 2014-01-21 15:10:14 but it didnt help 2014-01-21 15:10:18 still no workey 2014-01-21 15:12:32 only on windows 8.1 virtualbox? 2014-01-21 15:13:49 win 8.1 runs natively on the hw 2014-01-21 15:13:57 so i can run virtualbox there 2014-01-21 15:13:59 i had 8.0 2014-01-21 15:14:03 and upgraded to 8.1 2014-01-21 15:14:08 and removed norton 2014-01-21 15:14:08 i meant I can test on windows 7 2014-01-21 15:14:14 ok 2014-01-21 15:14:25 well, i was thinking of: test and resolve... 2014-01-21 15:14:36 or is it windows 8 issue 2014-01-21 15:15:22 i have a general windows issue 2014-01-21 15:15:41 so i'm generally disaponted of windows 2014-01-21 15:15:49 which version of vbox and alpine? 2014-01-21 15:15:59 I can test it with win7 2014-01-21 15:16:09 i havent got that far... 2014-01-21 15:16:21 i think its any virtualbox and alpine 2.7.3 2014-01-21 15:25:48 it boots up with alpine x64 and vbox 4.3.6 2014-01-21 15:25:53 at the setup screen 2014-01-21 15:29:09 finish install and rebooted and it works 2014-01-21 16:00:37 StarWarsFan: ^^^ 2014-01-21 16:00:47 thanks Frosh 2014-01-21 16:07:10 ooo its possible he selected a different linux selection at the beginning when selecting the hardware selection 2014-01-21 16:07:29 he has to select the right linux kernel and either 32 or 64 2014-01-21 16:14:21 Frosh: http://download.eisfair.org/installer-kernel-error.webm 2014-01-21 16:14:36 See the capture 2014-01-21 16:22:01 try linux 2.5 instead of other linux 2014-01-22 10:10:24 whee! 2014-01-22 10:10:31 cgit-0.10 has lua support 2014-01-22 10:39:48 fabled: what is the current status of rpi? possible to get a browser running? 2014-01-22 11:12:26 clandmeter, i have not tried. it might be not easily possible - at least HW accel is not supported yet. 2014-01-22 11:12:39 hi 2014-01-22 11:12:50 i did get vlc and omxplayer working 2014-01-22 11:12:56 so it's possible to watch movies 2014-01-22 11:12:58 what about https://github.com/ssvb/xf86-video-fbturbo 2014-01-22 11:12:59 without X 2014-01-22 11:13:50 clandmeter, that won't help much, as rpi has non-open gpu 2014-01-22 11:14:02 it'd need to be using the rpi specific libraries 2014-01-22 11:14:53 isnt rasbian using fbdev driver for x? 2014-01-22 11:14:58 likely, yes 2014-01-22 11:15:13 but all the apps doing anything useful are likely built using rpi omxil 2014-01-22 11:15:22 including Qt and media players 2014-01-22 11:16:10 im just intestested in a browser 2014-01-22 11:16:19 something like a kiosk system 2014-01-22 11:38:42 clandmeter, depends what you browse. simple stuff might work with xf86-video-fbdev, though i have not tested it. 2014-01-22 11:39:32 firefox did not build iirc, but midori should be there 2014-01-22 11:39:59 midori is fine for webkiosk 2014-01-22 11:52:14 hi 2014-01-22 11:52:25 musl update please:) 2014-01-22 11:56:11 stats are: 2014-01-22 11:56:19 musl-x86: 1664 of 1731 2014-01-22 11:56:19 musl-x86_64: 1655 of 1731 2014-01-22 11:56:20 musl-armhf: 1604 of 1734 2014-01-22 12:02:17 Frosh: thx for your test 2014-01-22 12:02:32 i've forwarded your suggestion to the corresponding person with the problem 2014-01-22 12:02:43 let's hope the best... 2014-01-22 12:14:18 fabled: looks good :) 2014-01-22 12:15:40 shafire, yeah, we got older gnupg, claws-mail and other stuff to compile recently 2014-01-22 12:16:14 so it's probably in usable state soonish. though getting x86 to run is still slightly tricky due to musl ld 2014-01-22 12:16:41 there are a few runtime bugs too 2014-01-22 12:17:07 udev was broke due to some glibc specific assumptions 2014-01-22 12:17:22 firefox subprocess? is segfaulting 2014-01-22 12:17:44 I do use musl x86_64 on my laptop now though 2014-01-22 12:17:51 so its kinda useful already 2014-01-22 12:18:03 ncopa, in lxc or as main environment? 2014-01-22 12:18:20 as main env 2014-01-22 12:18:24 wow. cool! 2014-01-22 12:18:49 there are a few things not working that i suspect is related musl 2014-01-22 12:19:02 firefox + grsec kernel. 2014-01-22 12:19:21 something segfaults and triggers the bruteforce protection in grsec 2014-01-22 12:19:37 with vanilla kernel something segfaults but firefox window shows up 2014-01-22 12:19:50 i think dhcpcd does not work 2014-01-22 12:19:55 udhcpc works 2014-01-22 12:20:01 wpa_supplicant works 2014-01-22 12:21:44 i think polkit or something does not work 2014-01-22 12:22:53 so i cannot get use bluez DBUS api unless root 2014-01-22 12:24:15 ERROR: vlc-2.1.2-r3: trying to overwrite usr/lib/vlc/plugins/video_output/libxcb_xv_plugin.so owned by vlc-xorg-2.1.2-r1. 2014-01-22 12:24:26 + bunch of other modules 2014-01-22 12:24:33 sounds like something is wrong in the split function 2014-01-22 12:24:54 sounds like that yes 2014-01-22 12:25:16 iirc it scanelf and find anything that links to libX11 or similar 2014-01-22 12:26:26 fabled: btw, i got dual booting with UEFI working with gummiboot too 2014-01-22 12:36:20 http://alpinelinux.org/forum/installation/php-http-wrapper 2014-01-22 12:39:25 i have some php fpm issues as well on my home box. 2014-01-22 12:39:35 but probably unrelated. 2014-01-22 13:40:30 anyone had this: (mod_fastcgi.c.2673) FastCGI-stderr: Primary script unknown 2014-01-22 13:40:55 php 5.5 fpm lighthttp 2014-01-22 14:10:51 ncopa: i think php-fpm is broken 2014-01-22 14:10:59 but my conclusion makes no sence :| 2014-01-22 14:11:33 broken as in misconfigured. 2014-01-22 14:15:45 dunno 2014-01-22 14:16:52 even with listen mode, i need to set it to lighttpd as user/group. 2014-01-22 16:35:41 okay guys 2014-01-22 16:35:52 happy days at fosdem :p I have holidays now 2014-01-22 16:35:52 bye 2014-01-22 18:50:21 is 2.7 builder stuck? 2014-01-22 18:51:42 build-2-7-x86: retry 2014-01-22 18:52:23 build-2-7-x86: rebuild 2014-01-22 18:52:53 algitbot: build 2.7-stable 2014-01-22 18:53:55 barthalion: thanks 2014-01-22 18:54:20 build servers keep changing and I can't keep up :( 2014-01-22 18:54:32 you're welcome 2014-01-22 18:54:37 nothing happened though 2014-01-22 18:55:12 algitbot: build 2.7 2014-01-22 18:55:29 well, 2.7-stable should have done the job 2014-01-22 18:55:42 but buildbots don't like me these days 2014-01-22 18:58:16 algitbot: build 2014-01-22 18:58:24 algitbot: build 2.7-x86 2014-01-22 18:59:55 thanks jbilyk 2014-01-22 19:00:02 np - hope that works 2014-01-22 19:12:04 barthalion / jbilyk: nothing yet :( 2014-01-22 19:13:55 tdtrask: http://bld1.alpinelinux.org/buildlogs/build-edge-x86.log 2014-01-22 19:14:52 seems like it built properly? 2014-01-22 19:15:06 oops - that's edg:) 2014-01-22 19:15:53 tdtrask: about all I know how to do is restart build-2-7-x86 lxc container 2014-01-22 19:17:56 algitbot: build 2.7-stable 2014-01-22 19:18:36 tdtrask: checked the buildmg lua scripts, "build 2-7.stable" seems to be the right command 2014-01-23 00:52:21 ping clandmeter 2014-01-23 00:52:39 ncopa are you there? 2014-01-23 00:52:52 they're both in europe, so unlikely...they're probably sleeping 2014-01-23 00:53:33 here it is 22:57 hs.. time to sleep too 2014-01-23 00:55:09 jbilyk_, bye 2014-01-23 06:26:13 ncopa, got this: 2014-01-23 06:26:15 ERROR: perl-error-0.17021-r0: trying to overwrite usr/share/perl5/vendor_perl/Error.pm owned by perl-git-1.8.5.3-r0. 2014-01-23 07:08:30 morning 2014-01-23 07:08:36 fabled: i'll look at it 2014-01-23 07:09:39 http://search.cpan.org/~shlomif/Error-0.17021/lib/Error.pm 2014-01-23 07:09:49 "Using the "Error" module is no longer recommended" 2014-01-23 07:10:06 does rt4 still need it? 2014-01-23 07:12:50 hum 2014-01-23 07:12:56 dunno 2014-01-23 07:13:47 perl-cache-chache depends on it 2014-01-23 07:21:17 ok i figured it out 2014-01-23 07:21:36 git has a bundled Error.pm which gets installed if the system Error.pm is not found 2014-01-23 07:23:30 argh 2014-01-23 07:23:34 checkapk is still broke 2014-01-23 14:36:12 files from v2.4.11-182-gf0071a9 uploaded 2014-01-23 16:01:29 2.7-stable buildbots are lazy 2014-01-23 16:01:34 or simply down 2014-01-23 16:05:07 probably just down 2014-01-23 16:06:37 algitbot: build 2.7-stable 2014-01-23 16:06:52 there we go 2014-01-23 16:07:03 going home. 2014-01-23 16:07:06 see you tm 2014-01-23 19:24:18 barthalion, you around ? 2014-01-23 19:45:00 vkrishn: well, a bit busy 2014-01-23 19:45:03 doing some chores 2014-01-23 19:45:21 ok, then later 2014-01-23 19:46:31 write what you want, I'll read it in hour or so 2014-01-23 19:51:06 vkrishn: ↑ 2014-01-23 19:52:46 just wanted to know how you update the main git aports from your local edit ? 2014-01-23 19:53:47 git push directly , or log into git.a.o and pull ...? 2014-01-23 19:58:06 I pull it from ssh://git.a.o or sth like that 2014-01-23 19:58:19 before I push something I do git pull --rebase 2014-01-23 19:58:22 and then push 2014-01-23 20:15:22 how do you check for `git describe` ? local or git.a.o ? 2014-01-23 20:26:42 I mean when you check for what log file to view ? 2014-01-23 20:49:24 I found another reason why you may like msg.a.o, ;) 2014-01-23 20:49:37 while checking for `git describe origin/master`, the return may not be same if a cherrypick is being done 2014-01-23 20:57:21 but then you can always 'git pull' to update, then 'git describe' 2014-01-23 21:17:04 gotta go, but would be checking logs for any replies 2014-01-24 10:24:29 yes, that really is FreeBSD ipfw. 2014-01-24 10:25:12 http://inai.de/images/nf-packet-flow.png 2014-01-24 10:25:18 this made me less than amused 2014-01-24 10:25:21 so, it occured to me... 2014-01-24 10:25:38 why don't we just hijack netfilter to emulate BSD netisr 2014-01-24 10:25:52 and then some helpful google chap did half the work for me already 2014-01-24 10:28:28 after spending literally 9 hours of my time figuring out why iptables was not seeing bridge packets... 2014-01-24 10:28:30 it occured to me 2014-01-24 10:28:35 "hey i can just replace this shit" 2014-01-24 10:29:04 anyway 2014-01-24 10:54:27 linux-grsec package at edge 2014-01-24 10:54:32 is really screwed up 2014-01-24 10:56:53 buffer 7 2014-01-24 10:56:57 buffer 8 2014-01-24 10:56:59 buffer 18 2014-01-24 10:57:01 oops 2014-01-24 11:02:18 but yeah 2014-01-24 11:02:20 ipfw 2014-01-24 11:02:22 is pretty much working 2014-01-24 11:02:24 ;) 2014-01-24 12:23:41 performance-wise, ipfw is 2.5x as fast as iptables in my testing 2014-01-24 12:23:53 even though it uses netfilter for netisr emulation 2014-01-24 12:23:56 ;) 2014-01-24 12:24:14 sent to mailing list instructions on how to use it 2014-01-24 12:25:46 then again, it's missing nat and stuff which probably adds a lot of the netfilter overhead 2014-01-24 12:36:14 kaniini: nice :) 2014-01-24 14:27:41 Is alpine known to work on the Raspberry Pi? 2014-01-24 14:28:06 No.... 2014-01-24 14:28:08 jdhore, we have experimental musl based armhf build that works on raspberry pi 2014-01-24 14:28:24 it's not complete, and some things are broke. but many things work too. 2014-01-24 14:41:18 fabled: openjdk is still broken? 2014-01-24 14:41:47 i see a git msg about openjdk broken and xbmc disabled because of it. 2014-01-24 14:41:55 clandmeter, ? 2014-01-24 14:41:57 on musl? 2014-01-24 14:41:58 yes 2014-01-24 14:42:03 openjdk is not yet built 2014-01-24 14:42:05 x86_64 2014-01-24 14:42:11 mmm 2014-01-24 14:42:16 i thought it worked 2014-01-24 14:42:21 it does 2014-01-24 14:42:25 i build it correctly 2014-01-24 14:42:31 i was just wondering 2014-01-24 14:42:31 huh 2014-01-24 14:43:09 sounds like someone just forgot to enable it 2014-01-24 14:43:16 ah ok 2014-01-24 14:43:55 i will push it 2014-01-24 17:12:26 fabled: nat is actually easily enabled... the ipfw does support divert sockets it just requires making a dev packages 2014-01-24 17:12:36 fabled: and then packaging the natd 2014-01-24 17:13:28 but, nat isn't our motives obviously... wirespeed filtering at 10gbps is our motive ;) 2014-01-24 17:13:53 with bridge devices 2014-01-24 17:14:00 which on iptables requires brouting 2014-01-24 17:14:04 ipfw does not 2014-01-25 16:10:37 re 2014-01-25 16:10:38 hi 2014-01-25 16:11:00 question regarding the new debug-build-functionality of abuild 2014-01-25 16:11:17 did someone test this, if the package still has subpackages? 2014-01-25 16:12:09 in example see APKBUILD from quassel 2014-01-25 16:12:13 http://git.alpinelinux.org/cgit/aports/tree/testing/quassel/APKBUILD 2014-01-25 16:12:36 i've added "$pkgname-dbg" to the list of subpackages 2014-01-25 16:12:44 build the package 2014-01-25 16:13:08 and installed common-, core- and dbg-package 2014-01-25 16:13:31 but if i try to debug a core dump, gdb says "no debugging symbols found" 2014-01-25 16:14:18 the binary which i want to debug is /usr/bin/quasselcore 2014-01-25 16:14:50 but the debug-package installs a file /usr/lib/debug/usr/bin/quassel.debug 2014-01-25 16:16:39 ls -l /usr/lib/debug/usr/bin/quassel.debug /usr/bin/quasselcore 2014-01-25 16:16:39 ... 3629665 Jan 25 12:50 /usr/bin/quasselcore 2014-01-25 16:16:39 ... 687852 Jan 25 12:56 /usr/lib/debug/usr/bin/quassel.debug 2014-01-25 16:17:37 the real binary is much bigger than the debug-file 2014-01-25 16:20:13 so i have no idea, if the debug build is working proberly to create a debug-package to already build subpackage 2014-01-25 16:20:43 there is a package "quassel-core" build and i expected another one "quassel-core-dbg" 2014-01-25 16:20:47 hm... 2014-01-25 16:21:28 ok, removing den dbg-subpackage and use -DCMAKE_BUILD_TYPE=Debug for now... 2014-01-25 16:21:45 but would be great if someone could explain what i did wrong... 2014-01-25 23:38:24 n8@all 2014-01-26 06:42:11 hmm 2014-01-26 06:42:15 new prosody release available 2014-01-27 13:26:41 ncopa: :-) 2014-01-27 13:26:53 by the way: when does a package come from testing to main? 2014-01-27 13:30:58 whem someone report that it works and is ok 2014-01-27 14:00:27 ncopa: ok, then I will check this today and let you know 2014-01-27 14:02:27 built it very often over the weekend and finaly it works as expected 2014-01-27 14:02:58 but a final check based on the official build system should be done anyway 2014-01-27 14:03:14 so i will let you know 2014-01-27 18:51:47 StarWarsFan: seems like quassel core and client works... 2014-01-27 19:02:37 arrived @home some minutes ago 2014-01-27 19:02:42 going to test it now... 2014-01-27 19:02:50 ncopa_: cool! 2014-01-27 20:01:54 ncopa_: yes, quasselcore works :-) 2014-01-27 20:02:24 would be nice to move it to main, so it could be used on 2.7 too... 2014-01-27 20:03:55 it won't 2014-01-27 20:04:07 it will be moved to main on edge 2014-01-27 20:04:20 and then? how to go ahead? 2014-01-27 20:04:22 we usually don't move packages from testing to main in stable releases 2014-01-27 20:04:42 there are exceptions of course, but you need to annoy ncopa 2014-01-27 20:05:01 i don't want to annoy somebody ;-) 2014-01-27 20:05:37 so when will it be available for 2.7 "the normal way" or which will be the next release where it will be available? 2014-01-27 20:05:46 2.8? 2014-01-27 20:06:27 2.8 2014-01-27 20:08:44 Any timeline somewhere? 2014-01-27 20:12:41 there surely is one, but I don't know it :p 2014-01-27 20:13:14 I'm hardly a good developer to ask about plans, sorry 2014-01-27 21:23:58 StarWarsFan: release every 6 months, so May timeframe 2014-01-27 21:39:49 i c 2014-01-27 21:51:56 another question: 2014-01-27 21:52:04 if a package comes with an init script 2014-01-27 21:52:19 and I add it using rc-update add ... 2014-01-27 21:52:35 and then uninstall the package 2014-01-27 21:53:05 is it task of the package or the user to remove the link in /etc/runlevels/default/ ? 2014-01-27 22:16:37 n8@all 2014-01-28 01:31:18 StarWarsFan|afk: user i'd think since they're the ones that added it 2014-01-28 03:13:42 is there a way to force a package to be deleted even if it's a dependency? 2014-01-28 03:14:08 i need to remove lua-sec so I can install lua-sec-prosody in its place but I can't because lua-sec is listed as a dependency of prosody 2014-01-28 03:14:16 --force, -f isn't working 2014-01-28 03:18:16 oh 2014-01-28 03:18:25 when I installed lua-sec-prosody it overwrote lua-sec's files 2014-01-28 03:18:30 but now apk complains of 1 error 2014-01-28 03:18:31 :P 2014-01-28 07:45:36 moin 2014-01-28 09:20:55 hi 2014-01-28 09:21:04 Mp5shooter: apk fix 2014-01-28 20:46:14 ncopa: interesting that I just stumbled across this: http://bugs.alpinelinux.org/issues/103 2014-01-28 20:46:23 ACTION started on such a bot this past weekend 2014-01-28 20:46:48 I have the simple redmine issue linking done, but am working on a few other features before I make available the repo for the app 2014-01-28 20:48:34 ACTION wanted more features than the ones that were listed on that third part page 2014-01-29 07:04:42 interesting 2014-01-29 07:04:49 i think we use sircbot for that 2014-01-30 04:56:50 why does postfix have krb5-conf as a dependency o.O 2014-01-30 07:01:29 Mp5shooter: is it a direct dependency? 2014-01-30 07:02:06 try apk info -r krb5-conf 2014-01-30 07:02:56 its heimdal-libs 2014-01-30 07:03:14 which is pulled in by libsasl 2014-01-30 07:04:41 it should been a plugin 2014-01-30 07:48:52 Mp5shooter: I have a fix for heimdal dep for postfix (sasl) 2014-01-30 07:49:11 i will move the gssapi and gs2 api cyrus-sasl plugins to subpackages 2014-01-30 07:49:19 do you think we should push that to v2.7? 2014-01-30 07:53:48 Mp5shooter: if you want the postfix/heimdal fix for v2.7 could you please file an issue on bugs.a.o 2014-01-30 14:01:46 i need make a decision what to do with lua 5.1/5.2 2014-01-30 14:01:58 the modules 2014-01-30 14:02:09 continue to have lua5.1-module and lua5.2-module 2014-01-30 14:02:21 or ship both 5.1 and 5.2 in lua-module 2014-01-30 14:29:42 <@ncopa> do you think we should push that to v2.7? 2014-01-30 14:29:52 hm 2014-01-30 14:30:07 idk 2014-01-30 15:35:14 any chance someone can push firefox v26 to stable repo please please please 2014-01-30 15:35:41 it is working really well ... the latest nss module too 2014-01-31 04:21:46 hello 2014-01-31 04:21:56 anybody uses freeradius? 2014-01-31 04:22:05 it is not working now in the latest alpine 2014-01-31 04:22:12 localhost:~# radiusd -X libssl version mismatch. Built with: 1000105f Linked: 1000106f 2014-01-31 04:22:23 let me look at the apk file 2014-01-31 04:29:52 abuild ing. If it solved the problem, the package is out of date 2014-01-31 16:59:44 files from v2.4.11-186-gabd9421 uploaded 2014-01-31 17:05:51 files from v2.4.11-186-gabd9421 uploaded 2014-01-31 19:12:26 nothing like http://p.pw/DLV jaja (wait to pass the advertising) xD