IRC Logs for #crux-devel Thursday, 2015-03-19

jaegerprologic: did you ever figure out what causes the BTRFS_BUILD_VERSION error when building docker?00:25
*** Workster has joined #crux-devel00:34
*** Workster has quit IRC00:34
*** Workster has joined #crux-devel00:34
Worksterjaeger, i even pasted a gist in #docker about the issue, no takers.00:36
jaegeryeah, that paste shows up as one of the google hits00:36
*** crash_ has quit IRC00:36
*** crash_ has joined #crux-devel00:38
Worksterlol thats fast00:41
Worksteri can't bump mesa until after work so if anyone wants it sooner go for it.00:42
jaegerI'll take a look at it00:42
Worksteralso TODO32 need to add name change for mesa3d to mesa00:42
jaeger10.5.1 adds the mako dep that's been mentioned00:56
Worksterthat i moved to opt01:01
Worksterso just adding mako and bumping version is all that is needed.01:01
Worksterafk lunch down the street01:01
prologicjaeger, builds fine here?01:18
prologicunless I'm missing something01:18
prologicDocker 1.5.0 here01:18
jaegerprologic: doesn't for me. btrfs-progs is installed, btrfs support is in the kernel01:18
prologicweird01:18
prologiclet me try to build here01:19
prologichow else do I have 1.5.0 installed01:19
prologicif not built :)01:19
jaegerhell if I know, never tried to build it before :)01:19
prologiclol01:19
jaegerunrelated, mesa 10.5.1 works for me on both intel and nvidia systems01:19
prologicgimme 5mins01:19
jaegergoing to build it in a safe env before pushing anything, though01:19
prologic=======> Building '/data/packages/docker#1.5.0-5.pkg.tar.gz' succeeded.01:20
prologicso wtf01:20
jaegerwhere is BTRFS_BUILD_VERSION actually defined?01:20
prologicgood question01:21
prologiclet's see01:21
prologic$ grin BTRFS_BUILD_VERSION01:21
prologic./btrfs/version.h:01:21
prologic   14 : #define BTRFS_BUILD_VERSION "Btrfs v3.18.2"01:21
prologic~/contrib/docker/work/src/docker-1.5.001:22
prologic$ grin BTRFS_BUILD_VERSION01:22
prologic./daemon/graphdriver/btrfs/version.go:01:22
prologic   20 : return string(C.BTRFS_BUILD_VERSION)01:22
jaegerso it's in the docker source?01:22
jaegerhttp://sprunge.us/KSUD01:24
prologicand you have a /use/include/btrfs/version.h ?01:26
jaegeryes, but it doesn't define BTRFS_BUILD_VERSION01:26
prologicoh01:27
prologicupgrade/rebuild btrfs-progs?01:27
*** Feksclaus has quit IRC01:27
jaegerIt's the current one... but your grin output up there seems to show it's in the docker source, unless I read that wrong01:28
*** erdic has quit IRC01:28
prologicit's also in /usr/include//btrfs/version.h:01:29
prologicnot sure why my btrfs-progs is different to yours01:29
jaegeryou're running a different btrfs-progs, I'm guessing01:29
jaegerrather than core01:29
prologiclet's see01:30
*** erdic has joined #crux-devel01:30
prologicjust rebuilt btrfs-progs01:31
prologicthe define is defintiely there01:31
prologicyour btrfs-progs is broken somehow?01:31
jaegerhttp://sprunge.us/ZRSF01:32
prologichttps://gist.github.com/207635478ea8c399965b01:32
jaegerIt's straight out of core01:32
prologic# tar tvf /data/packages/btrfs-progs#3.18.2-1.pkg.tar.gz | grep version01:32
prologic-rw-r--r-- root/root       413 2015-03-19 11:33 usr/include/btrfs/version.h01:32
prologicso *is* mine :)01:32
jaegerIt's also 3.19, you're not running the current one in core01:32
prologicoh!01:32
prologicso they remived the define in later btrfs-progs01:32
prologicgotcha01:32
prologichaha01:32
jaegerlooks that way01:32
prologicthis is why I don't update as eagerly as some :)01:33
prologicokay01:33
prologicso presumably Romster you've filed a bug report then?01:33
jaegerwell, you should test things against ports you publish in contrib, heh01:33
prologicoh I do :)01:33
jaegerwith a current core01:33
prologicbut I didn't think this would become a problem01:33
prologichaha01:33
prologic# prt-get diff01:33
prologicNo differences found01:33
jaegerjaeger's rule for his own ports: always always always test in a clean env01:33
prologicmust have happened quite recently01:33
prologicwhen was the btrfs-progs bump?01:33
prologichaha01:34
prologicyeah I do too :)01:34
jaeger6 days ago01:34
prologicbut I think it's bad luck that I'm01:34
prologic...01:34
prologicthat I'm 6 days behind01:34
prologichaha01:34
jaegerhttps://crux.nu/gitweb/?p=ports/core.git;a=commitdiff;h=7b5c1ad2bb5b14ede081a6a9de3523c8e6a85e1101:34
prologic*sigh*01:34
prologiccan't keep up :)01:34
prologicso sorry about that01:34
prologicwork-around: downgrade btrfs-progs? :)01:34
prologicI'll file a bug report with Docker if Romster hasn't already01:34
jaegerI'll wait for docker to get fixed, I'm not in any rush to mess with it01:35
prologicperhaps see if I can't patch it somehow01:35
prologicsure no worries01:35
prologicsorry about that inconvenience :)01:35
jaegerno worries, at least now we know the cause01:36
prologicindeed01:36
prologicdidn't take long to figure out :)01:37
jaegermany issues don't once you start digging01:37
prologichttps://github.com/docker/docker/issues/1148701:42
Worksternah i never filed a bug but i go ignored in #docker when i pasted a gist of the problem02:17
Workstertest your ports in docker!02:18
Worksterand be sure it's up to date02:18
Worksterromster/crux-build02:18
Worksterjaeger, mesa3d-32 can just have the version bumped since it'll pull in mako from mesa3d dependency.02:19
jaegerI'm testing it now02:19
prologicI do02:33
prologicthat's what I keep telling you :)02:33
prologicI'm just 6 days behind apparently02:33
prologic*sigh* :)02:33
*** Workster has quit IRC03:13
*** Workster has joined #crux-devel03:17
*** mavrick61 has quit IRC03:48
*** mavrick61 has joined #crux-devel03:49
*** Romster has quit IRC04:22
*** nilp has joined #crux-devel06:13
*** Workster has quit IRC06:25
*** Romster has joined #crux-devel06:45
nomiusDid anyone got an invisible mouse pointer with one of the latest upgrades in xorg?07:36
Romsterother than the most recent mesa3d now. nope. my mouse still shows.07:42
nomiusIt's really weird07:49
nomiusI move it and it does things, but it's like invisible07:49
Romsteryou tried regenerating the cursor cache if it even has one?07:52
Romstergtktheme stuff07:52
nomiusNoup, how do I do that?07:54
Romsteri don't feel like researcing what caches there are -_-07:54
nomiusShouldn't rc.fix do that?07:55
nomiuss/Shouldn't/Doesn't/07:55
Romsterthat's only on a upgrade from different crux versions07:56
Romsterhave you ran revdep?07:57
Romsterlooked in Xorg.log08:02
nomiusYeah, nothing weird08:21
frinnst=======> New files found:09:11
frinnstNEW       -rwxr-xr-x      root/root       usr/lib/dri/gallium_drv_video.la09:11
frinnstNEW       -rwxr-xr-x      root/root       usr/lib/dri/gallium_drv_video.so09:11
frinnstanybody else?09:11
*** nilp has quit IRC09:18
juefrinnst: no, works for me10:02
Romsteryeah thats normal and that had a commnet for that and guess who removed that comment10:32
Romstercomment*10:32
Romsterthat was in the Pkgfile.10:32
Romsterhttps://crux.nu/gitweb/?p=ports/xorg.git;a=commitdiff;h=ed4fc9d978a03ffc3d01042144373ae81ec3e38810:34
Romster-_-10:34
frinnstah :)10:37
Romsterhaha :D10:59
Romsterman you fell in that one.11:00
frinnst:)11:13
*** Feksclaus has joined #crux-devel11:30
Romsterhad any time to look at thunderbird i'm still running those changes on mine.11:31
Romsteri'm not sure if it's just me or it takes a little longer to load due to that rpath.11:31
Romsterother than that i haven't noticed any errors or anything.11:32
frinnstIm not sure im a fan of the rpath thingy11:33
frinnstIts just a cosmetic fix for revdep - thunderbird still loads the libraries that revdep errors on11:34
frinnstand removing the devel stuff from firefox may break 3rd party plugins - havent tested that yet11:35
frinnstgranted, doubt many people use that many plugin these days, but still11:35
frinnstIm gonna test if thats the case with firefox some more11:36
Romsterthat is true.11:36
Romsterthe 3rd party stuff removal is on thunderbird11:36
frinnstyep, doubt there are any usecases for that11:37
Romsteractually more so the devel headers idl and stuff that we don't even use11:37
Romsterand i'm using firefox without that ldconfig file and no issues with firefox.11:37
Romsterthat junk on thunderbird juat feels like waste11:38
frinnstyep11:38
Romsterfeel free to do as yo wish but it doesn't hurt to point these things out and test.11:38
frinnstanyways, there will be a few weeks before major firefox and thunderbird releases11:39
Romsteri'm using both them changes on firefox and thunderbird no issues.11:39
frinnstsame here wrt. firefox11:39
frinnstnot done any changes to thunderbird yet11:39
Romsterno point making small changes before the next release anyways, if you do change anything.11:39
Romsterfirefox might of needed that ld path eons ago. seems obsolete now.11:40
Romsternow it just messes up revep and ldd on the libs11:41
Romsterbut really isn't a major thing. but ya know just pointed what i found out.11:41
Romsteri need to have another crack at docbook too.11:41
Romstersee if ic an for once and all fix it.11:42
frinnstyep and it's appreciated :)11:45
Romsterno problem.11:46
juefrinnst: already tried the new openssh?12:07
frinnstnot yet12:09
jueok, seems to work fine, will commit it12:10
jaegerfrinnst: I still get the gallium mismatch on my main boxes but they have a lot more installed; the safe build env doesn't12:30
Romsterjaeger, libva makes it install 2 new files is all12:32
rmullfrinnst: I've had the gallium mismatch from mesa3d for as long as I can remember fwiw12:32
rmullI always just upgrade anyway12:32
rmullI have libva12:32
Romsterbut there is a cyclic dependency between libva and mesa3d each depend on each other.12:32
Romsterits no big deal with ignore new files in footprint12:33
jaegerI never run with ignore on by default12:33
Romsteri'm wondering if that should be on by default on new installs.12:33
RomsterPKGMK_IGNORE_NEW="yes"12:34
jaegernot in my opinion12:35
juejaeger: me neither12:35
Romstermany say about new files and if it's a bug or not or have no idea.12:36
rmullCould extend .footprints to include optional files when in the presence of other packages that generate them :P12:36
rmulljk.12:37
jaegerMy personal preference is that users see the mismatch and learn why it's happening, not ignore it12:37
Romstertoo much complexity for crux.12:37
Romsterah maybe so12:37
Romstergenerates a few questions is all12:37
jaegerObviously that's a personal opinion but I also feel like crux users need to be slightly more advanced than newbie level12:37
rmullCan just link directly to: https://crux.nu/Wiki/FaqPkgUtils#ntoc212:39
Romstertrue12:39
rmullMaybe it could be developed a little more12:39
rmullTo differentiate between new and missing12:39
Romsterwhen the wiki works again12:39
rmulloh, haha :\12:39
jaegerI fixed the wiki yesterday12:39
Romsteroh12:39
frinnstseems the openssl update is out14:36
frinnstdebian just pushed it14:36
frinnst19-Mar-2015:    Security Advisory: twelve security fixes14:37
jaegerheh14:37
frinnstI cant push it for 2 more hours14:38
jaegerI'm testing it here, I can push it if you like14:38
frinnstlovely14:39
frinnstI have to battle a vmware appliance thats running out of diskpace and doesnt supply a proper shell...14:41
jaegeryuck14:41
frinnstvshield manager crap14:42
jaegerso far openssl 1.0.2a doesn't give me any trouble. openssh works, s_client works14:46
frinnstyeahm should be wsafe14:46
jaegerminor update, yeah14:46
jaegerRomster: are there updated sources for the two openssl-32 patches or do I need to remake those manually?14:55
*** frinnst has quit IRC14:57
*** frinnst has joined #crux-devel15:01
*** frinnst has quit IRC15:01
*** frinnst has joined #crux-devel15:01
frinnststupid slow centos mirrors15:14
frinnstor wait, redhat hasnt pushed it at all yet15:14
*** Feksclaus has quit IRC15:19
*** jue has quit IRC18:02
*** jue has joined #crux-devel18:26
*** Feksclaus has joined #crux-devel18:59
*** Workster has joined #crux-devel22:15

Generated by irclog2html.py 2.11.0 by Marius Gedminas - find it at mg.pov.lt!