2017-04-03 06:46:10 oops 2017-04-03 06:48:28 re cloc: actually this is not my fault. The builders didn't fetch the new tarball because they already have a tarball with the same name. Could sameone remove the old tarball from $SRCDEST? 2017-04-04 12:28:01 df -h 2017-04-04 12:28:11 :) 2017-04-04 12:40:22 <^7heo> sh: syntax error: unexpected ")" 2017-04-04 13:05:50 xorg 1.19 coming up here soonish 2017-04-04 13:43:08 <^7heo> Wow, something that built! 2017-04-04 13:43:11 <^7heo> ACTION hides 2017-04-04 22:10:15 humm 2017-04-04 22:10:26 where is build-edge-x86 2017-04-04 22:20:09 kaniini: hm, probably died together with existing x86-only machines :P ¯\_(ツ)_/¯ 2017-04-06 18:48:03 humm 2017-04-06 18:48:06 why did all fail 2017-04-06 19:38:46 <^7heo> same as usual 2017-04-06 23:07:36 humm 2017-04-06 23:07:40 where is s390x builder :D 2017-04-06 23:08:07 it can't do jack man 2017-04-06 23:09:47 fixing go-bootstrap 2017-04-06 23:09:51 on ppc64le 2017-04-06 23:13:29 or perhaps not 2017-04-06 23:13:35 seems go1.4.3 does not have power support 2017-04-06 23:13:38 or aarch64 2017-04-06 23:20:33 ok 2017-04-06 23:20:38 this php7 shit needs to be sorted 2017-04-07 00:18:18 kaniini : re: go. that's a long story 2017-04-07 00:19:39 kaniini : the idea is to tuse go1.4.3 on amd64, cross-compile go-bootstrap targeting s390x/ppc/aarch/etc, use that toolchain to either create *native* bootstrap, or just cross-compile go 2017-04-07 00:19:53 ah 2017-04-07 00:20:04 but fabled proposed better approach 2017-04-07 00:20:11 which is in current APKBUILD 2017-04-07 00:20:25 re: s390x. ncopa is working on it. he's pretty busy 2017-04-07 00:20:51 and I boot alpine s390x on kvm on amd64 just but on native s390x kvm host, it stucks ... 2017-04-07 00:20:57 so that's basically it lol 2017-04-07 00:21:06 getting there tho 2017-04-07 00:23:14 kaniini : can you confirm this ? http://patchwork.alpinelinux.org/patch/3088/ . I have not touch this since then ... 2017-04-07 00:24:36 nvm if it bothers you though... 2017-04-07 00:25:56 i dont think it is correct 2017-04-07 00:35:23 kaniini : shit s390x build is on 2017-04-07 00:47:10 hold on to your hats 2017-04-07 02:37:31 wtf @ aarch64 :D 2017-04-07 02:53:45 kaniini : :D 2017-04-07 02:54:14 php7 xapian binding is located in the wrong place there 2017-04-07 02:54:17 im not sure why 2017-04-07 03:45:32 i wish we had build logs for s390x 2017-04-07 03:58:20 algitbot: retry master 2017-04-07 04:40:27 kaniini : we need to wait for ncopa tomorrow 2017-04-07 04:41:21 my network keeps disconnecting... till now we have these failed on s390x : vala netcat-openbsd py-paramiko liblogging kamailio cvs conky libical i3lock fcgi bwm-ng 2017-04-07 04:43:16 I built most of them alright. Exception : kamailio failed at check(). conky failed unknown error. all other good to me 2017-04-07 04:45:52 conky failed at update_config_sub. hum... 2017-04-07 04:46:24 kamailio hash one test failed for me (setuid() test fail) 2017-04-07 04:46:29 that's it 2017-04-07 04:53:43 yes acpica too. there is Werror bug in it. I thought I fixed it with a patch .. 2017-04-07 05:00:40 nope mlmmj 2017-04-07 05:14:00 strip: Unable to recognise the format of the input file `./usr/lib/python3.6/site-packages/gns3server/modules/docker/resources/bin/busybox' 2017-04-07 05:14:15 that umm raises lots of questions about gns3 2017-04-07 05:21:27 si 2017-04-07 05:47:09 algitbot: retry master 2017-04-07 07:12:36 morning. i will set up the build logs today 2017-04-07 07:12:44 for s390x 2017-04-07 08:01:53 build logs for s390x should be available now 2017-04-07 13:32:42 whats wrong with xapian? 2017-04-07 13:32:47 i cannot reproduce in my dev box 2017-04-07 13:37:05 ncopa, here it fails with same error 2017-04-07 13:40:32 hmm, I have a patch to compile psmisc 2017-04-07 13:41:17 k 2017-04-07 13:42:46 if you sprunge/pastebin it I can apply it right away 2017-04-07 14:02:05 rdutra, let me get the PR for it 2017-04-07 14:03:13 sigh 2017-04-07 14:03:20 rnalrd, https://github.com/alpinelinux/aports/pull/1219 2017-04-07 14:03:25 tnx 2017-04-07 14:03:25 I sent the patches upstream also. 2017-04-07 14:03:32 rnalrd, thank you for helping ppc64le! 2017-04-07 14:03:37 np 2017-04-07 14:51:36 we disable go for ppc64le for now? 2017-04-07 14:52:20 leitao, the psmisc patch does not apply anymore 2017-04-07 14:52:29 rnalrd, hm 2017-04-07 14:52:33 rnalrd, did you bump version? 2017-04-07 14:52:41 care to submit a new one, or I'll just manually apply it 2017-04-07 14:52:43 yeah 2017-04-07 14:52:53 I will forward patch it. 2017-04-07 14:53:01 When I sent it upstream, it didn t apply also 2017-04-07 14:53:05 they had fixed on problem already 2017-04-07 14:53:37 i'm speaking of the GIT patch 2017-04-07 14:53:51 i did't check on Makefile specific patch or such 2017-04-07 14:54:06 ok 2017-04-07 14:54:10 rnalrd, I will fix it 2017-04-07 14:54:13 error: community/psmisc/APKBUILD: patch does not apply 2017-04-07 14:54:33 looks like complains only on APKBUILD 2017-04-07 14:55:02 tnx 2017-04-07 14:59:03 leitao, nvm i'm basically done with few modifications 2017-04-07 14:59:23 fcolista: perl-term-progressbar is yours 2017-04-07 14:59:49 ncopa: it was installing to wrong place on aarch64. no idea why. 2017-04-07 15:00:01 ncopa, checking 2017-04-07 15:00:08 rnalrd, cool 2017-04-07 15:00:19 i did an initial rebuild overnight, about 95% of it built on x86-64 2017-04-07 15:00:23 fcolista i have a fix 2017-04-07 15:00:26 i push it 2017-04-07 15:00:35 ok 2017-04-07 15:01:38 i'm hitting mostly strange set -e stuff 2017-04-07 15:01:41 so that is the good news 2017-04-07 15:01:47 yeah 2017-04-07 15:01:52 95% is not bad 2017-04-07 15:02:12 and leitao and gromero has already fixed many 2017-04-07 15:03:01 thx ncopa 2017-04-07 15:03:08 np 2017-04-07 15:03:10 that's the same i'm slowly doing for all the packages.. 2017-04-07 15:03:16 yeah 2017-04-07 15:03:32 im very happy im not alone on it :) 2017-04-07 15:03:37 ncopa, by the way, I understand that ppc64le is stuck on go now. 2017-04-07 15:03:40 btw: there's qt5 to be upgraded 2017-04-07 15:03:40 it might not be easy to solve. 2017-04-07 15:03:47 ncopa, can we skip it for now? 2017-04-07 15:03:48 leitao: i just disabled it 2017-04-07 15:03:52 i just did 2017-04-07 15:03:59 ncopa, cool! 2017-04-07 15:04:13 ncopa, gromero has a go version working on ppc64le 2017-04-07 15:04:17 but we will need to bump to 1.8 2017-04-07 15:04:28 and break the circular dependency somehow 2017-04-07 15:04:29 yes, thats how i understood it 2017-04-07 15:04:32 whatta 2017-04-07 15:04:40 whats wrong with aarch64? 2017-04-07 15:04:59 i cannot reporoduce it 2017-04-07 15:06:23 ltb-project-ssp is now hosted on github 2017-04-07 15:07:35 rnalrd: ltp-project-ssp might be one app that depends on php5 2017-04-07 15:07:41 ON IT 2017-04-07 15:07:44 sorry caps 2017-04-07 15:07:51 ok,,ok ok relax... 2017-04-07 15:07:52 :D 2017-04-07 15:08:03 and thanks! 2017-04-07 15:08:12 there's a new release out, check compat with php7 2017-04-07 15:08:19 ah 2017-04-07 15:08:21 cool 2017-04-07 15:08:23 awesome 2017-04-07 15:11:05 <^7heo> ncopa: any chance to see the 3.1.0 of mkinitfs today? 2017-04-07 15:11:18 <^7heo> ncopa: It would really help me for the testing this week end 2017-04-07 15:11:58 whoops sorry 2017-04-07 15:12:02 on worry i have 2017-04-07 15:12:09 what if it break things? 2017-04-07 15:12:12 its friday... 2017-04-07 15:12:20 <^7heo> it can't break things if there's no APKBUILD update. 2017-04-07 15:12:25 ok 2017-04-07 15:12:28 <^7heo> let's wait for monday to do the abump 2017-04-07 15:12:31 <^7heo> and do the release today 2017-04-07 15:12:41 <^7heo> so I can abuild it myself easily for the week end 2017-04-07 15:14:44 <^7heo> ncopa: thanks a zillion. 2017-04-07 15:14:49 <^7heo> ncopa: do you like wine? 2017-04-07 15:15:11 <^7heo> (not gonna send some - it's for future usage ;) 2017-04-07 15:15:23 i do! 2017-04-07 15:15:29 <^7heo> Red or White? 2017-04-07 15:15:36 depends on the food 2017-04-07 15:15:45 <^7heo> yeah without anything ;) 2017-04-07 15:15:50 <^7heo> I drink red by itself 2017-04-07 15:15:58 then it depends on the weather 2017-04-07 15:16:02 lol 2017-04-07 15:16:03 <^7heo> ahah nice one. 2017-04-07 15:16:03 :) 2017-04-07 15:16:06 <^7heo> true dat. 2017-04-07 15:16:31 <^7heo> with very hot summer, you won't drink red wine that much 2017-04-07 15:16:38 <^7heo> s/with/during a/ 2017-04-07 15:16:45 exactly, white wine is better there 2017-04-07 15:17:19 <^7heo> yeah 2017-04-07 15:17:35 <^7heo> Ok I'll try to keep that in mind for next alconf 2017-04-07 15:18:10 alcohonf 2017-04-07 15:18:19 <^7heo> hahahahaha 2017-04-07 15:18:39 open bar is a requirement 2017-04-07 15:18:46 just get some company doing alpine to sponsor it 2017-04-07 15:18:53 i hear docker has billions they dont know what to do with 2017-04-07 15:19:00 <^7heo> ;) 2017-04-07 15:19:14 <^7heo> kaniini: will you even come? 2017-04-07 15:19:23 yes 2017-04-07 15:19:28 <^7heo> even in prague? 2017-04-07 15:19:32 well 2017-04-07 15:19:42 assuming trump does not seal the border anyway 2017-04-07 15:19:49 <^7heo> not in the way out. 2017-04-07 15:19:51 <^7heo> ;) 2017-04-07 15:20:14 my main concern is getting stuck because he did something stupid 2017-04-07 15:20:23 <^7heo> stuck out or stuck in? 2017-04-07 15:20:34 i guess there are worse places to get stuck than prague though 2017-04-07 15:20:45 <^7heo> at least for the first two weeks ;) 2017-04-07 15:21:19 however -- he seems to be firing his 'friends' who keep giivng him bad advice 2017-04-07 15:21:24 so i am sure by alconf it will be fine 2017-04-07 15:21:45 <^7heo> :D 2017-04-07 15:24:04 kaniini: at least in prague they'd be able to czech who you are easier 2017-04-07 15:24:17 that is an awful pun 2017-04-07 15:24:48 ^7heo: i mean, home is where my stuff is. so getting stuck would be very unfortunate 2017-04-07 15:30:06 <^7heo> kaniini: yeah, but stuff isn't that important. 2017-04-07 15:30:11 <^7heo> kaniini: home is where my pack is. 2017-04-07 15:32:25 ^7heo: imagine if the unstumpable took your pack 2017-04-07 15:33:19 <^7heo> Xe: FATAL ERROR: Parser faulted at character 26 2017-04-07 15:33:44 come on man, gcc does not have that bad diagnostics anymore 2017-04-07 15:33:47 need to have the ^ 2017-04-07 15:34:28 <^7heo> kaniini: like I'm gonna provide lazy ass IRC lusers with position dependent markers. 2017-04-07 15:35:02 ncopa: aarch64 xapian problem is, libtool: install: /usr/bin/install -c .libs/xapian.so /home/buildozer/aports/community/xapian-bindings/pkg/xapian-bindings/usr/lib/php7/xapian.so 2017-04-07 15:35:14 ncopa: it's installing to wrong place. dont know why. 2017-04-07 15:35:20 <^7heo> kaniini: he's gone 2017-04-07 15:35:23 <^7heo> kaniini: in weekend. 2017-04-07 15:35:30 ah 2017-04-07 15:35:32 <^7heo> yep 2017-04-07 15:35:41 i wonder how to fix it 2017-04-07 15:35:58 <^7heo> rm -rf works fine here. 2017-04-08 22:58:55 duh, ppc64le builder was stuck! 2017-04-08 22:59:59 monitoring-plugins build was on a dead loop 2017-04-08 23:00:05 just restart the daemon 2017-04-10 14:12:39 rnalrd: looks like netcat-openbsd needs an update 2017-04-10 14:12:48 yes, i'm working on it 2017-04-10 14:13:26 looks like that qt5-qtscript is broken, but not due to s390x patch 2017-04-10 14:13:39 yeah 2017-04-10 14:14:02 i suppose we should upgrade qt5 to 5.7 something 2017-04-10 14:22:17 whoops, typo 2017-04-10 14:26:12 leitao: list of package with depends=.*go 2017-04-10 14:26:16 ncdev-edge-x86_64:~/aports/community$ grep depends.*go */APKBUILD | tpaste 2017-04-10 14:26:16 http://tpaste.us/pQyE 2017-04-10 15:30:00 ncopa, ok, I will disable all of them 2017-04-10 15:30:28 thanks 2017-04-10 15:31:26 ncopa, would you prefer just one commit for them all? Or one commit per package? 2017-04-10 16:00:32 leita: no strong opionion 2017-04-10 16:00:38 do whats easiest for you 2017-04-10 16:00:47 leitao i mean 2017-04-10 16:01:39 ncopa, it seems that you list has some packages that does not depend on go 2017-04-10 16:01:42 I filtered it. 2017-04-10 16:04:45 yeah i just did a stupid grep 2017-04-10 16:06:24 ncopa : can you use my built go to bootstrap go ? https://gitlab.com/tmh1999/alpine-s390x-repo/tree/master/alpine/edge 2017-04-10 16:06:37 ncopa : go s390x cannot be bootstrap using go-bootstrap atm 2017-04-10 16:07:09 https://gitlab.com/tmh1999/alpine-s390x-repo/raw/master/alpine/edge/community/s390x/go-1.7.4-r2.apk 2017-04-10 16:11:13 tmh1999: ok 2017-04-10 16:11:41 ncopa : actually we can make go-bootstrap/s390x but it needs manual work. fabled and I discuss about this before. Thanks. 2017-04-10 17:12:26 rnalrd, I have a workaround for openblas on ppc64le, disabling the level3 tests on ppc64le 2017-04-10 17:12:29 do you mind if I upload it? 2017-04-10 18:33:34 some package's log say it's successful but algitbot report failure. hum. 2017-04-11 16:49:00 the comminuty/monitoring-plugins is broken on s390x 2017-04-11 16:49:08 it floods the log 2017-04-11 16:49:17 looks like it is running yes | something 2017-04-11 16:49:30 or end up running: yes something 2017-04-11 16:49:43 so it goes into endless filling up of the log 2017-04-11 16:54:24 ha 2017-04-11 16:54:36 monitoring-plugins configure script is stupid 2017-04-11 16:56:11 its not just stupid 2017-04-11 16:57:30 AC_ARG_WITH(perl, 2017-04-11 16:57:30 ACX_HELP_STRING([--with-perl=PATH], 2017-04-11 16:57:30 [sets path to perl executable]), 2017-04-11 16:57:30 with_perl=$withval,with_perl=$PERL) 2017-04-11 16:57:30 AC_SUBST(PERL, $with_perl) 2017-04-11 16:57:57 if you run ./configure --with-perl ... 2017-04-11 16:58:00 to enable perl 2017-04-11 16:58:12 then autoconf will set with_perl=yes 2017-04-11 16:58:16 and PERL=yes 2017-04-11 16:58:28 so when you later try to execute $PERL ... 2017-04-11 17:11:37 ncopa, monitoring-plugins froze the ppc64le builder 2017-04-11 17:11:47 and generate a 100G build log 2017-04-11 17:18:10 https://github.com/monitoring-plugins/monitoring-plugins/issues/1480 2017-04-11 17:18:19 it was not the only builder if froze 2017-04-11 17:26:44 stupid autoconf 2017-04-11 17:29:34 whatta 2017-04-11 17:29:37 libtool: error: cannot determine absolute directory name of 'yes/lib' 2017-04-11 17:32:34 looks like clandmeter has a happy monday :) ... on a tuesday 2017-04-11 17:34:25 huh 2017-04-11 17:34:56 well at least it does not fill the build logs forever... 2017-04-11 17:35:39 why it doesnt first build its dep 2017-04-11 17:38:41 oh 2017-04-11 17:38:43 crap 2017-04-11 17:39:34 ncopa, make it pretty print please! 2017-04-11 18:13:42 ncopa, hi! Crackmapexec is failing because there is a dependency that was not built 2017-04-11 18:13:49 how do we force the dependency to be built? 2017-04-11 18:14:06 on my system, crackmapexec builds fine if the depedency chain is built 2017-04-11 18:18:49 leitao: do you have the build log for that dependency? 2017-04-11 18:19:34 http://build.alpinelinux.org/buildlogs/build-edge-ppc64le/community/py-greenlet/ 2017-04-11 18:19:37 hum 2017-04-11 18:19:38 ok 2017-04-11 18:19:40 right 2017-04-11 18:20:26 ncopa, I fixed it yesterday FWIW 2017-04-11 18:22:07 py2-gevent-1.1.2-r1[py2-greenlet] 2017-04-11 18:22:14 it fails to figure out the build order 2017-04-11 18:22:24 i need to go 2017-04-11 18:22:27 will check that tm 2017-04-11 18:23:07 ah i know 2017-04-11 18:23:28 we need to add a hack to py-gevent 2017-04-11 18:23:57 we need to add the depends in _py2 and _py3 to makedepends 2017-04-11 18:24:17 to make it possible to calculate the build order 2017-04-11 18:24:42 basically add py2-greenlet and py3-greenlet to makedepends 2017-04-11 18:24:52 in py-gevent 2017-04-11 18:24:52 for what package? 2017-04-11 18:24:57 ok 2017-04-11 18:24:59 let me try that 2017-04-11 18:25:00 py-gevent 2017-04-11 18:26:08 ncopa: I have a more elegant solution 2017-04-11 18:27:02 jirutka, what is your idea? 2017-04-11 18:27:11 leitao: https://wiki.alpinelinux.org/wiki/APKBUILD_examples:Python 2017-04-11 18:29:16 dont hink that solves it 2017-04-11 18:29:20 problem is: 2017-04-11 18:29:31 supakcage() { depends=foo; } 2017-04-11 18:29:39 while foo is not in makedepends 2017-04-11 18:29:51 there is no way to know that foo needs to be built first 2017-04-11 18:30:30 otherwise you might end up with a subpackage that cannot be installed due to the missing (not ye built) dependency foo 2017-04-11 18:31:08 ncopa: see https://git.alpinelinux.org/cgit/aports/commit/?id=2905100a 2017-04-11 18:31:20 ncopa: it’s on wiki, just some ppl don’t bother to read… :/ 2017-04-11 18:32:05 ok i get it 2017-04-11 18:32:06 nice 2017-04-11 18:32:06 oh crap, i forgot to bump pkgrel 2017-04-11 18:32:11 im on my way out... 2017-04-11 18:32:32 ncopa, jirutka: thanks! 2017-04-11 18:32:33 have a nice evening 2017-04-11 18:32:49 but I guess that it failed to build on ppc64le dues to missing build dep, so it’s not needed to bump pkgrel? 2017-04-11 18:33:00 ncopa: you too, nice evening! 2017-04-11 18:33:43 ls 2017-04-11 18:36:00 ls: command not found 2017-04-11 18:37:10 XD 2017-04-12 00:29:39 algitbot: retry master 2017-04-12 02:07:02 ncopa : here is the list of packages I have been working on and found solution for them. I hope the description should be enough for you to fix it in the s390x builder : https://gitlab.com/tmh1999/alpine-s390x-repo/blob/master/status 2017-04-12 02:07:26 ncopa : in progress packages in here, fyi : https://gitlab.com/tmh1999/alpine-s390x-repo/blob/master/in-progress 2017-04-12 09:03:15 algitbot: retry master 2017-04-12 15:05:57 jirutka, I am working on your question about || true on rm. Give me more minutes. 2017-04-12 16:23:18 ncopa : could help me review some patches in main/ I sent and fire up again the s390x builder ? There was some network/checksum error on the builder that I cannot fix : https://gitlab.com/tmh1999/alpine-s390x-repo/blob/master/status 2017-04-12 16:55:55 tmh1999, nice summary 2017-04-12 16:59:46 leitao : Thanks. I hope to have s390x build good asap 2017-04-12 17:00:07 tmh1999, is it missing a lot of packages? 2017-04-12 17:00:48 leitao : no, not much. missing main/ are in that url. about community, about 15 packages 2017-04-12 17:01:46 cool 2017-04-12 17:01:52 where is the builder currently? 2017-04-12 17:02:19 afaik it's in one of a sles vm provided by IBM to ncopa 2017-04-12 17:02:29 he built a lxc in that sles 2017-04-12 17:02:45 there's no 'real' s390x alpine yet as we are working on the bootloader 2017-04-12 17:03:08 I have an s390x alpine in an kvm (on x86 host), still struggle to run it in native s390x machine 2017-04-12 17:12:09 its lxc in ubuntu nowdays 2017-04-12 17:12:18 the sles got replced with an aubuntu box 2017-04-12 17:12:46 cool, good to know 2017-04-12 17:51:20 leitao: fwiw, i dont think its worth spending time on the old webkitgtk^^^ 2017-04-12 17:51:36 ncopa, cool. 2017-04-12 17:51:41 we want get rid of it 2017-04-12 17:51:43 I was trying to build without jt 2017-04-12 17:51:45 jit* 2017-04-12 17:51:53 Debian has the exact same version built. 2017-04-12 17:51:54 apps should use webkit2gtk 2017-04-12 17:51:58 ncopa, ok 2017-04-12 17:52:10 i think its like 2 or 3 that still needs the old 2017-04-12 17:52:10 ncopa, should I disable it for ppc64le? 2017-04-12 17:52:21 i suppose we should check those and disable those too 2017-04-12 17:52:44 geary 2017-04-12 17:53:21 testing/handbreak 2017-04-12 17:53:28 py-webkitgtk 2017-04-12 17:53:42 i think we can just disable all those for ppc64le 2017-04-12 17:54:01 ok 2017-04-12 17:54:13 i looked at geary today 2017-04-12 17:54:27 and they have in their git master already migrated to webkit2 2017-04-12 17:55:41 right, should not be a problem 2017-04-12 17:55:53 I think we will need to skip nodejs also. 2017-04-12 17:58:24 leitao: what's the matter with nodejs? 2017-04-12 18:00:18 gromero, is it ported to power at all? 2017-04-12 18:02:49 alpine-ppc64le:~$ sudo apk add nodejs 2017-04-12 18:02:49 (1/4) Installing libcrypto1.0 (1.0.2k-r0) 2017-04-12 18:02:49 (2/4) Installing http-parser (2.7.1-r1) 2017-04-12 18:02:49 (4/4) Installing nodejs (6.10.1-r0) 2017-04-12 18:02:49 (3/4) Installing libssl1.0 (1.0.2k-r0) 2017-04-12 18:02:49 Executing busybox-1.26.1-r3.trigger 2017-04-12 18:02:51 OK: 1386 MiB in 329 packages 2017-04-12 18:02:53 alpine-ppc64le:~$ vim hello.js 2017-04-12 18:02:55 alpine-ppc64le:~$ node hello.js 2017-04-12 18:02:57 Hello World 2017-04-12 18:03:42 leitao: ^ 2017-04-12 18:03:44 gromero, bummer! 2017-04-12 18:03:51 help us and send the patches 2017-04-12 18:06:58 leitao: https://github.com/alpinelinux/aports/commit/184cbe9b717f5ae699187e7264b7bc7fd97b3181 2017-04-12 18:07:59 gromero, cool. Did you have a chance to look at nodejs-current? 2017-04-12 18:08:46 leitao: https://github.com/alpinelinux/aports/commit/ecff4dad4aeac6cec74666dd39adc3001f6e82dd 2017-04-12 18:09:03 leitao: but it's not working as in main. I'll have a look at it 2017-04-12 18:09:35 gromero, cool 2017-04-12 19:54:03 how was libbvirt built earlier on ppc64le? 2017-04-12 19:56:09 oh its netcat-openbsd that introduced a dep of libbsd? 2017-04-12 19:56:47 i suppose we should build busybox-telnet 2017-04-12 20:01:43 ncopa, checking libvirt 2017-04-12 20:04:50 isn't libbsd ported to ppc64le? 2017-04-12 20:05:24 leitao, ncopa : new libbsd broke netcat-openbsd on s390x too. I'm wondering how Debian guys have it on s390x. 2017-04-12 20:05:46 leitao : looks like only x86/x86_64 has support for new version of libbsd 2017-04-12 20:05:47 the errors seems weird 2017-04-12 20:05:52 yeah a.out.h thing 2017-04-12 20:05:53 nlist.c:43:25: fatal error: linux/a.out.h: No such file or directory 2017-04-12 20:06:04 never saw this header ever 2017-04-12 20:06:17 reading some ml yesterday, a.out.h is only generated on x86* 2017-04-12 20:18:00 leitao : debian folks build libbsd 0.8.3 on ppc64le too. 2017-04-12 20:18:14 tmh1999, those bastards! 2017-04-12 20:18:26 lol n 2017-04-12 20:18:33 lol not sure how they did it 2017-04-12 20:55:26 leitao : https://git.yoctoproject.org/cgit.cgi/poky/tree/meta/recipes-support/libbsd/libbsd/0003-Fix-build-breaks-due-to-missing-a.out.h.patch 2017-04-12 20:56:27 tmh1999, humm. does he get the copyright for himself? 2017-04-12 20:58:03 leitao : I guess it's under yocto project license ? 2017-04-12 21:00:18 leitao: https://github.com/alpinelinux/aports/pull/1246 2017-04-12 21:01:03 nice 2017-04-12 21:21:31 leitao : that yocto patch work on s390x. is the copyright thing a constraint for us to use the patch ? 2017-04-12 21:21:50 tmh1999, I don't know. 2017-04-12 21:21:58 was the patch accepted upstream? 2017-04-12 21:22:13 Upstream-Status: Pending 2017-04-12 21:36:36 no 2017-04-12 21:36:40 it is fine 2017-04-12 21:36:53 he does not change the license, only the copyright 2017-04-12 21:39:05 kaniini : Thanks. 2017-04-13 16:23:55 ncopa: could you please fix qt5-qtscript to build on x86? I’ve looked into it a bit, found some references, but no simple patch and cause I now nothing about qt infrastructure, can’t figure out what’s going on 2017-04-13 16:35:37 screw it, I’m gonna just disable qt5-qtscript on x86 for now >_< 2017-04-13 16:37:29 algitbot: retry master 2017-04-13 16:38:37 jirutka, how do I see the commands that the bot understands? 2017-04-13 16:38:55 leitao: I don’t know :( I just learned that this command works 2017-04-13 16:39:30 jirutka, what does this command does? 2017-04-13 16:40:12 maybe we should write some script that would massively spam maintainers, who push changes that fails to build… 2017-04-13 16:40:33 jirutka, yes! 2017-04-13 16:40:37 obviously setting builder to not skip failures does not work to force people fix their shit 2017-04-13 16:40:49 armhf was blocked for days… 2017-04-13 16:41:58 leitao: hmm, to be honest, I’m not sure what this command really do; just know that when builder is stuck, this sometimes help 2017-04-13 16:42:17 leitao: I know that it’s lame answer, but really don’t know what exactly it do :/ maybe clandmeter ? 2017-04-13 16:42:34 jirutka, no worries. 2017-04-13 16:42:43 I am wondering how to grab more information on the builder. 2017-04-13 16:43:00 leitao: our current build system is quite bad :/ 2017-04-13 16:43:12 <^7heo> we have all the infinite future to improve it ;) 2017-04-13 16:43:20 jirutka, I never saw a good build server in fact. 2017-04-13 16:43:31 you would not trust debian and ubuntu if you know buildd 2017-04-13 16:43:35 ^7heo: I know :) I’m partially working on it :) 2017-04-13 16:43:40 heh 2017-04-13 16:43:52 I don’t trust Debian and Ubuntu even when I don’t know their build system XD 2017-04-13 16:44:03 buildd is still not able to build in paralell 2017-04-13 16:44:05 <^7heo> jirutka: yeah we're all gonna improve all the things. One word at a time. 2017-04-13 16:44:08 all the builds should be -j1 2017-04-13 16:44:15 <^7heo> leitao: why? 2017-04-13 16:44:35 ^7heo, because build scripts are makefiles 2017-04-13 16:44:46 and, it is hard to create dependency on Makefile 2017-04-13 16:44:49 oh how debianish XD 2017-04-13 16:44:53 no one cared about it in the past 2017-04-13 16:44:56 now, it is hard to change 2017-04-13 16:45:15 whole debian pkg system is based on shitty make files 2017-04-13 16:45:54 <^7heo> leitao: I don't understand either what you mean or how are makefiles, dependencies and parallelization even related 2017-04-13 16:46:07 <^7heo> (I get parallelization and mutexes but that's it) 2017-04-13 16:46:53 as far as I know debian, I can imagine even impossibly bad solutions and reasons why something can’t be fixed :) 2017-04-13 16:47:05 Debian/rules, which is Debian's APKBUILD, is a makefile. 2017-04-13 16:47:40 when you run a make with -J2, it runs what is independent in parallel. 2017-04-13 16:47:59 so, if we set buildds to run with -j2, it tries to pack stuff that are still being compiled. 2017-04-13 16:48:01 things like that 2017-04-13 16:48:11 XD 2017-04-13 17:08:41 ncopa: clandmeter: okay, we have really a problem; it’s not just qt5-qtscript, it seems that maybe all pkgs depending on qt5-qtbase are affected… 2017-04-13 18:52:10 jirutka: i plan to upgrade qt to 5.8 2017-04-13 18:52:17 will try look at it tomorrow 2017-04-13 18:52:23 ncopa: great! 2017-04-16 00:50:13 algitbot: retry master 2017-04-16 00:51:15 algitbot: retry master 2017-04-16 11:20:54 ^ rnalrd could you please take a look at build failure of lua-rapidjson? it fails because LuaJIT segfaults when trying to run tests 2017-04-17 12:25:25 leitao: could you please look at lua-dromozoa-utf8? it’s failing on ppc64le in check, LuaJIT segfaults here 2017-04-17 12:26:24 jirutka, hello. Sure, I will take a look 2017-04-17 12:26:29 is the builder stuck at this package now? 2017-04-17 12:26:40 leitao: thanks! yes, it’s currently stuck on it 2017-04-17 12:26:55 ok 2017-04-17 12:27:20 leitao: I can disable it for ppc64le, but I think that it’s an indicator of that there’s something wrong in LuaJIT for ppc64le 2017-04-17 12:29:51 jirutka, let me take a look first. 2017-04-17 12:29:57 okay :) 2017-04-17 18:14:01 jirutka, we reproduced the luajit issue and gromero found an workaround. 2017-04-17 18:14:15 pie is causing the toc pointer to be corrupted. 2017-04-17 18:14:34 do you mind if we compile with -no-pie temporarily until we have a final patch? 2017-04-18 10:39:09 algitbot: retry 3.3-stable 2017-04-18 10:39:20 algitbot: build 3.3-stable 2017-04-18 13:38:19 it seems that lua-dromozoa-utf8 is still using the old luajit build 2017-04-18 13:38:24 should we force rebuild of luajit? 2017-04-18 18:05:53 ncopa : could you help me to fire up the s390x builder ? I just want to know which packages currently failed to build then I can continue to work with them. 2017-04-18 19:48:48 tmh1999_: i was not aware that it had stopped 2017-04-18 19:48:52 i will check it 2017-04-18 19:49:15 ncopa : thanks. if it is too late for you at night, tomorrow would be okay :) 2017-04-18 20:07:22 tmh1999: community/rancid failed in configure script 2017-04-18 20:07:29 ended up in endless ping 2017-04-18 20:07:53 33516 32000 buildoze S 952 0% 0 0% /bin/ping 127.0.0.1 56 1 2017-04-18 20:09:33 ncopa : I got that one too. working on it 2017-04-18 20:10:17 the current list of build failures: http://tpaste.us/ELyD 2017-04-19 02:44:26 kaniini : are you online ? Could you help me to check if there was any message from algitbot on s390x builder between go-bindata and thermald ? I was dc during that window ... 2017-04-19 02:45:03 around 20:40 - 21:08 EDT time 2017-04-19 02:45:49 1 hour ahead of your tz 2017-04-19 06:57:12 tmh1999: http://build.alpinelinux.org/buildlogs/build-edge-s390x/testing/thermald/thermald-1.5.4-r0.log 2017-04-19 06:57:55 tmh1999: oh no 2017-04-19 06:58:08 tmh1999: ebook-tools, corkscrew, radare2, lxd, otrs, mycli, ossec-hids, dovecot-antispam-plugin, py-bottle-request, envonsul, wok, py-sqlalchemy-utils 2017-04-19 06:59:24 kaniini : yeah? 2017-04-19 06:59:39 kaniini : thank you :) 2017-04-19 08:30:59 ncopa, can you add the missing patch for tlsdate? 2017-04-19 09:29:03 uh 2017-04-19 09:29:25 tlsdate builds just fine here 2017-04-19 09:37:55 rnalrd, did you update your env? 2017-04-19 09:38:05 is against new libressl iirc 2017-04-19 09:40:44 rnalrd, ncopa added that patch yesterday, so i guess its needed for something. he just forget to include it in git. 2017-04-19 09:45:11 yes, I did 2017-04-19 09:46:19 unless it's has been pushed <1h ago and mirror did not update yet 2017-04-19 09:48:03 https://git.alpinelinux.org/cgit/aports/commit/main/tlsdate/APKBUILD?id=e321fe70b8f9f36940a0070e7f734f729287ce64 2017-04-19 09:48:29 ncopa, and add the patch to apkbuild :) 2017-04-19 09:48:32 rnalrd removed it 2017-04-19 09:49:50 uhm my mirrors don't have libressl-2.5 yet 2017-04-19 09:49:57 that's why it built here 2017-04-19 09:50:02 sorry about that 2017-04-19 09:50:30 i'm sorry :-) 2017-04-19 09:50:45 might be an idea to git pull and build libressl locally 2017-04-19 09:50:51 yeah 2017-04-19 10:08:00 algitbot: build master 2017-04-19 10:17:57 is that a new feature? 2017-04-19 10:22:45 ncopa, maybe "OK $nickname" would have been nicer ;-) 2017-04-19 10:23:18 or Yes Master :) 2017-04-19 10:23:31 i thin its old feature 2017-04-19 10:23:44 the responce? 2017-04-19 10:25:57 probably debug code leftovers 2017-04-19 10:25:58 :) 2017-04-19 15:53:05 <^7heo> damn 2017-04-19 15:53:12 <^7heo> I really need to ignore 'failed to build' too. 2017-04-19 15:53:20 <^7heo> that happens WAY too often. 2017-04-21 16:07:02 ncopa, apk? 2017-04-21 16:07:23 hi 2017-04-21 16:07:46 hum 2017-04-21 16:07:48 not me 2017-04-21 16:07:54 weird 2017-04-21 16:08:10 probably someone that just use our buildrepo script 2017-04-21 16:08:13 with default config 2017-04-21 16:08:17 next week ill setup secure mqtt so we can test. 2017-04-21 16:08:22 yeah 2017-04-21 16:08:25 sounds good 2017-04-21 16:08:41 what do you think if we move mqtt to proxy? 2017-04-21 16:08:57 proxy? 2017-04-21 16:09:01 other lxc? 2017-04-21 16:09:09 other host? 2017-04-21 16:09:19 or else we need to move the certs on creation 2017-04-21 16:09:34 so other container 2017-04-21 16:09:39 yes 2017-04-21 16:10:00 msg runs on its own container 2017-04-21 16:10:06 we could keep it that way 2017-04-22 00:54:07 some builders took it in wrong order, trying to build clang 4.0.0 before llvm 4.0.0 :( it’s probably b/c clang depends on llvm, that is provided by llvm4… 2017-04-22 04:06:07 funny thing llvm4 build good on my s390x machine but failed on the builder, hum 2017-04-22 10:17:51 algitbot: retry master 2017-04-22 10:20:55 algitbot: retry master 2017-04-22 11:47:06 algitbot: retry master 2017-04-22 12:31:21 algitbot: retry master 2017-04-22 12:40:21 algitbot: retry master 2017-04-22 13:59:56 clandmeter: could you please take a look at this strange error on aarch64? it seems that there’s something wrong on aarch64 builder, http://build.alpinelinux.org/buildlogs/build-edge-aarch64/main/postfix/postfix-3.1.4-r1.log 2017-04-23 10:36:16 algitbot: retry master 2017-04-23 16:20:07 The sound of silence :) 2017-04-23 19:30:18 <^7heo> Go to France if you want some noise. 2017-04-24 11:24:12 master mirror is down while we add more diskspace to it 2017-04-24 17:29:14 ncopa, it seems that the builder is not resolving it. 2017-04-24 17:29:15 ip* 2017-04-24 17:29:16 curl: (6) Couldn't resolve host 'search.cpan.org' 2017-04-24 17:36:52 leitao_: we are working on it. i think its a dns issue 2017-04-24 17:37:17 oh 2017-04-24 17:37:24 didnt read it correctly 2017-04-24 17:37:29 thats another issue 2017-04-24 17:37:32 :| 2017-04-24 17:37:46 ACTION should learn to read first then talk... 2017-04-24 17:38:04 yeah clandmeter how dare you 2017-04-24 17:38:59 we are changing rsync.a.o so that could resolve incorrectly for a while. 2017-04-24 17:50:37 no more scaleway?!!!!? 2017-04-24 17:55:42 kaniini: yeah we need get off scaleway 2017-04-24 18:06:23 ncopa: where do you plan to move? or actually moving atm? 2017-04-24 18:06:42 we will talk about that tm 2017-04-25 12:22:58 nice to see green messages for a change :) 2017-04-25 12:44:36 <^7heo> ncopa: wooow, GREEN! 2017-04-25 12:44:48 <^7heo> ncopa: are we now using efficient energy or? 2017-04-25 16:15:15 whee \o/ 2017-04-25 16:15:18 more green 2017-04-25 20:17:23 nginx is broken on 3.6 2017-04-25 23:29:38 this checksum error on eventlog for the 3.6 build seems to not be reproducible anymore 2017-04-25 23:29:44 should we force the rebuild of the package? 2017-04-25 23:47:09 i dont have access to the s390x builder :( 2017-04-27 13:58:26 ncopa, any idea why eventlog is having checksum issues 2017-04-27 13:58:31 I am not able to reproduce here. 2017-04-27 19:46:01 there is something wrong with the ppc64le builder. A lot of build fails because of checksum 2017-04-27 20:09:39 algitbot: retry master 2017-04-28 01:04:57 algitbot: retry master 2017-04-28 07:19:19 algitbot: retry 3.6-stable 2017-04-28 07:20:30 algitbot: retry 3.6 2017-04-28 07:21:11 algitbot: retry master 2017-04-28 07:45:40 who maintains build-edge-s390x? i think its using ccache which makes spl build fail. 2017-04-28 13:34:34 algitbot: build master 2017-04-28 16:55:45 algitbot: retry master 2017-04-28 17:13:32 algitbot: retry master 2017-04-28 19:41:02 jirutka, kaniini: Would you mind reviewing https://github.com/alpinelinux/aports/pull/1330 ? 2017-04-28 19:41:08 please. :-) 2017-04-28 20:08:50 jirutka, thanks for the last fix! it conflicted with the cache files, right? 2017-04-28 20:09:02 leitao: yes 2017-04-28 20:09:14 jirutka, ok, thanks! 2017-04-28 20:15:34 hum 2017-04-28 20:15:38 seems gcj is broke 2017-04-28 20:39:34 hmmh, seems gcj does not work due to PIE; probably boehm-gc damage 2017-04-28 22:48:33 algitbot: retry master 2017-04-29 11:22:05 algitbot: retry master 2017-04-29 11:23:27 algitbot: start building these god damn aarch64 pkgs on edge! >_< 2017-04-30 22:52:12 o no