2015-08-02 11:21:00 <vkris> ncopa: ^^ "< ncopa> dir-listing.exclude didnt work as expeceted", was there a patch 'dir-listing.exclude' ?
2015-08-03 06:20:41 <ncopa> vkris: i changed:  dir-listing.exclude = "dirlisting"
2015-08-03 06:20:46 <ncopa> to:  dir-listing.exclude = ("dirlisting")
2015-08-03 06:21:01 <ncopa> dir-listing.exclude requires an array
2015-08-03 07:45:58 <ncopa> algitbot: retry
2015-08-03 07:47:28 <ncopa> algitbot: retry master
2015-08-03 07:48:22 <clandmeter> ok
2015-08-04 17:50:43 <fabled> hmm
2015-08-04 17:50:59 <fabled> libreoffice seems to sometimes fail in the unittests. but it is non-deterministic.
2015-08-06 08:39:36 <fabled> huh
2015-08-06 08:41:24 <fabled> ncopa,  boost may have get broken
2015-08-06 08:41:33 <fabled> boost-dev no longer pulls in all subpkgs
2015-08-06 08:42:00 <fabled> hmm
2015-08-06 08:43:42 <fabled> no
2015-08-06 08:43:44 <fabled> it's glm-dev missing
2015-08-06 08:43:47 <fabled> does boost need that?
2015-08-06 08:43:58 <fabled> no
2015-08-06 08:43:59 <fabled> ok
2015-08-06 08:44:01 <fabled> my mistake then
2015-08-06 13:47:41 <fabled> ncopa, nuttcp failed
2015-08-06 13:48:01 <fabled>  
2015-08-06 13:48:01 <fabled> /home/buildozer/aports/testing/nuttcp/src/nuttcp-7.3.3.c:545:27: fatal error: linux/sockios.h: No such file or directory
2015-08-06 13:48:04 <fabled> linux-headers ?
2015-08-06 13:48:07 <ncopa> bah
2015-08-06 13:48:09 <ncopa> ok
2015-08-07 15:15:52 <ncopa> algitbot: retry 3.1-stable
2015-08-07 15:17:33 <ncopa> algitbot: retry 3.1-stable
2015-08-07 15:18:26 <ncopa> algitbot: retry 3.1-stable
2015-08-07 15:21:11 <ncopa> finally, its weekend
2015-08-11 04:42:06 <pnutzh4x0r> ncopa: http://tpaste.us/2N7v <- adds support for LD_LIBRARY_PATH to findlib
2015-08-11 04:56:09 <pnutzh4x0r> ncopa: http://tpaste.us/GxeW this handles splitting LD_LIBRARY_PATH by :
2015-08-11 06:00:20 <ncopa> mornin
2015-08-11 06:00:30 <ncopa> pnutzh4x0r: cool. thanks.
2015-08-11 06:01:06 <ncopa> i talked with musl devs yesterday and got suggestion to use something like:  /lib/ld-musl-x86_64.so.1 --preload=libssl.so --list
2015-08-11 06:01:27 <ncopa> which should give similar output as ldconfig
2015-08-11 09:12:25 <ncopa> pnutzh4x0r: i ended up using the LD_LIBRARY_PATH anyway
2015-08-11 09:12:36 <ncopa> thanks for that
2015-08-13 11:43:54 <ScrumpyJack> ncopa: thanks for that
2015-08-13 13:39:27 <fabled> ncopa, the umask is needed only with abuild-apk; not with regular apk
2015-08-13 13:40:04 <ncopa> it does happen with sudo apk add too
2015-08-13 13:40:05 <ncopa> i think
2015-08-13 13:42:50 <ncopa> ncdev-3-2-x86_64:~/aports/main/pango$ ls -l /usr/lib/pango/*/modules.cache
2015-08-13 13:42:50 <ncopa> ls: /usr/lib/pango/*/modules.cache: No such file or directory
2015-08-13 13:42:51 <ncopa> ncdev-3-2-x86_64:~/aports/main/pango$ sudo apk add -q pango
2015-08-13 13:42:51 <ncopa> ncdev-3-2-x86_64:~/aports/main/pango$ ls -l /usr/lib/pango/*/modules.cache
2015-08-13 13:42:51 <ncopa> -rw-rw-rw-    1 root     root          1401 Aug 13 15:42 /usr/lib/pango/1.8.0/modules.cache
2015-08-13 13:43:05 <fabled> oh
2015-08-13 13:43:24 <fabled> hm, should we reset umask in apk
2015-08-14 07:24:27 <fcolista> agitbot: retry
2015-08-14 07:24:34 <fcolista> algitbot: retry
2015-08-14 08:23:40 <barthalion> it was something more sophisicated
2015-08-14 08:23:47 <barthalion> algitbot: retry build-edge-x86
2015-08-14 08:23:52 <barthalion> algitbot: build build-edge-x86
2015-08-14 08:23:57 <barthalion> 2
2015-08-14 08:24:03 <barthalion> algitbot: build build-edge-x86_64
2015-08-14 08:35:05 <ncopa> 3
2015-08-14 08:35:17 <ncopa> algitbot: whats up with those numbers
2015-08-14 08:44:39 <fcolista> ncopa, if you can, please delete slock-1.2.tar.gz in bld servers
2015-08-14 08:48:02 <ncopa> done
2015-08-14 08:48:20 <ncopa> algitbot: retry master
2015-08-14 08:49:09 <fcolista> thx ncopa 
2015-08-17 10:02:52 <fcolista> brb
2015-08-17 13:47:46 <fabled> huh
2015-08-17 13:54:20 <fabled> why on earth. it works on my dev-build
2015-08-17 13:55:32 <fabled> ah
2015-08-17 13:55:33 <fabled> bash
2015-08-20 11:51:05 <fcolista> algitbot: what's happening ?
2015-08-20 11:51:19 <fcolista> ping algitbot 
2015-08-20 13:28:52 <ncopa> algitbot: how are you doin?
2015-08-20 14:40:52 <alacerda> ping algitbot 
2015-08-20 15:28:16 <fcolista> ping algitbot 
2015-08-20 19:11:30 <alacerda> algitbot, wake up
2015-08-21 09:58:51 <ncopa> algitbot: ping
2015-08-21 11:12:05 <alacerda> algitbot, you don't like me? When I ping you you do not answer me, but when ncopa does you promptly answer him
2015-08-21 11:12:07 <alacerda> :P
2015-08-21 11:12:33 <ncopa> :)
2015-08-23 08:29:05 <barthalion> ncopa: I see no reason why would this fail ↑
2015-08-23 08:29:11 <barthalion> ncopa: builds fine on my buildbox
2015-08-23 08:29:31 <barthalion> also I can't connect to bld1 and bld2 to see the logs
2015-08-23 08:29:52 <barthalion> http server is apparently down
2015-08-23 08:38:16 <barthalion> armhf always happy
2015-08-23 15:06:00 <barthalion> algitbot: retry build-edge-x86_64
2015-08-23 15:06:08 <barthalion> or something
2015-08-23 15:06:10 <barthalion> plz
2015-08-24 06:12:44 <ncopa> morning
2015-08-24 06:12:46 <ncopa> http://bld2.alpinelinux.org/buildlogs/build-edge-x86_64/main/nginx-lua/nginx-lua-1.8.0-r3.log
2015-08-24 06:12:59 <ncopa> i started up http service on bld1 and bld2
2015-08-24 06:25:38 <barthalion> ncopa: adding -doc is enough to make man pages compressed?
2015-08-24 06:27:03 <ncopa> barthalion: yes
2015-08-24 06:27:39 <ncopa> since abuild-2.23 or 2.24
2015-08-24 07:57:46 <ncopa> algitbot: retry master
2015-08-24 09:39:07 <clandmeter> ncopa: thx
2015-08-24 09:40:34 <ncopa> clandmeter: np
2015-08-24 13:51:09 <alacerda__> oops
2015-08-25 06:59:55 <ncopa> algitbot: retry master
2015-08-25 07:00:21 <ncopa> abuild is buggy
2015-08-25 07:00:38 <ncopa> the source download is racy
2015-08-25 07:00:50 <fabled> ncopa, recent issue?
2015-08-25 07:01:25 <ncopa> the issue has alwasy been there but was exposed now when build-edge-x86 and x86_64 share the distfiles dir
2015-08-25 07:01:34 <ncopa> i mounted the distfiles over nfs
2015-08-25 07:01:56 <fabled> oh
2015-08-25 07:02:27 <ncopa> 	flock -n -x 9 || msg "Waiting for ${lockfile##*/}..."
2015-08-25 07:02:27 <ncopa> 	flock -x 9
2015-08-25 07:02:36 <ncopa> i suppose that fails over nfs
2015-08-25 07:04:00 <ncopa> hm
2015-08-25 07:04:03 <ncopa> it should work
2015-08-25 07:04:20 <ncopa> this is the func: http://git.alpinelinux.org/cgit/abuild/tree/abuild.in#n318
2015-08-25 07:17:00 <ncopa> i don tknow why it failed
2015-08-25 07:18:08 <ncopa> https://dpaste.de/doZP
2015-08-25 09:00:38 <ncopa> definitively bug in abuild: http://build.alpinelinux.org/buildlogs/build-edge-x86_64/main/quagga-nhrp/quagga-nhrp-0.99.24.1_git20150825-r0.log
2015-08-25 09:01:11 <fabled> algitbot: retry master
2015-08-25 09:02:51 <alacerda> algitbot, are you sick ?
2015-08-25 09:09:25 <fabled> ncopa, care to priorize the wget issue?
2015-08-25 09:09:32 <fabled> if every push makes things fail, it's bad
2015-08-25 09:09:40 <fabled> or revert to local distcache
2015-08-25 09:14:49 <ncopa> algitbot: retry master
2015-08-25 09:14:57 <ncopa> algitbot: retry 3.2-stable
2015-08-25 09:14:58 <ncopa> yeah
2015-08-25 09:15:05 <ncopa> i need figure out what and why things go wrong
2015-08-25 09:15:12 <ncopa> code looks ok
2015-08-25 09:15:30 <ncopa> but it might be that i need to do something special for getting lockfiles working over nfs
2015-08-25 09:57:51 <ncopa> it seems the flock(1) is totally broke over nfs mount
2015-08-25 12:54:33 <ncopa> ok. i dont understand why flock(2) does not work over nfs
2015-08-25 12:54:36 <ncopa> i think it should
2015-08-25 13:59:55 <ncopa> file locking on linux is a mess
2015-08-25 14:10:11 <fabled> ncopa, locks do not work on nfs without special mount options
2015-08-25 14:10:18 <fabled> even then they are racy
2015-08-25 14:10:44 <fabled> ncopa, please disable nfs sharing for now
2015-08-25 14:10:48 <fabled> it's causing more problems than solving
2015-08-25 14:11:23 <ncopa> locks shoudl work by default
2015-08-25 14:11:31 <ncopa> but not with flock
2015-08-25 14:11:45 <ncopa> apparently if you use flock() on nfs client
2015-08-25 14:12:05 <ncopa> it will be converted to fcntl(F_SETLK) on server side
2015-08-25 14:12:16 <ncopa> if fcntl is used both sides it should work
2015-08-25 14:12:35 <ncopa> busybox flock uses flock(2)
2015-08-25 14:13:35 <ncopa> i dont know if there are any other simple tricks we can do to indicate that file is downloading form other place
2015-08-25 19:34:47 <barthalion> yes
2015-08-25 19:34:54 <barthalion> ncopa: literal lockfile
2015-08-25 19:35:25 <barthalion> with hostname and pid of blocking process, for example
2015-08-26 13:10:32 <ncopa> ok, i think I have a permanent solution for the nfs locking issue
2015-08-28 06:21:41 <fabled> ncopa, seems there's again cyclic dependency. this time between gtk+2.0 and librsvg
2015-08-28 06:23:08 <ncopa> i thought i NACK'ed that patch
2015-08-28 06:23:45 <fabled> hmmh
2015-08-28 06:48:14 <ncopa> i unintentionally added librsvg dep
2015-08-28 06:48:18 <ncopa> fixed now
2015-08-28 09:47:03 <fabled> ncopa, you need to disable go-bootstrap too
2015-08-28 09:47:39 <ncopa> i thought i did
2015-08-28 10:09:07 <fabled> >>> swig*: Running postcheck for swig
2015-08-28 10:09:07 <fabled> >>> WARNING: swig*: Found /usr/share/man but package name doesn't end with -doc
2015-08-28 10:09:08 <fabled> >>> ERROR: swig*: Found uncompressed man pages:
2015-08-28 10:09:08 <fabled>         /usr/share/man/man1/ccache-swig.1
2015-08-28 10:09:08 <fabled> >>> ERROR: swig*: prepare_subpackages failed
2015-08-28 10:09:08 <fabled> >>> ERROR: swig: all failed
2015-08-28 10:09:10 <fabled> >>> swig: Uninstalling dependencies...
2015-08-28 10:09:12 <fabled> ncopa, ^^
2015-08-28 13:50:31 <ncopa> hum whats up with 3.1 arm builder?
2015-08-28 13:50:35 <ncopa> is it tracking edge?
2015-08-28 14:01:49 <fabled> i booted it
2015-08-28 14:01:59 <fabled> maybe it caught up with something
2015-08-28 17:04:10 <fabled> >>> gcc: Cleaning temporary build dirs...
2015-08-28 17:04:11 <fabled> Waiting for /var/cache/distfiles/gcc-5.2.0.tar.bz2.lock ...
2015-08-28 17:04:11 <fabled> abuild-fetch: fcntl(F_SETLKW): Stale file handle
2015-08-28 17:04:11 <fabled> >>> ERROR: gcc: all failed
2015-08-28 17:04:13 <fabled> ncopa, ^^^
2015-08-31 08:51:54 <ncopa> i wonder what that "abuild-fetch: fcntl(F_SETLKW): Stale file handle" was