[01:21] <psusi> so I have noticed that boot time is delayed due to friendly-recovery.service... it is conditioned on recovery being in the kernel command line, which it usually isn't so it doesn't run at all, but it is After=systemd-udev-settle.service and Before=systemd-fsck-root.service, so it delays fscking the root fs until after udev-settle, which takes quite some time
[01:22] <psusi> could this be resolved by splitting it into two services, one that is wanted by sysinit, and has the condition, and a second that has the ordering requirements, but is only Wanted by the first service, so if its conditions are evaluated early in boot and found to be false, won't pull in the second service at all, and thus its transitive ordering requirement won't be enforced?
[01:27] <sarnold> psusi: probably best to file a bug and add the tag systemd-boot, that'll get it to pitti's queue
[01:27] <psusi> filed bug, will add tag, thanks
[01:28] <psusi> are there plans to move to systemd for user session management ( instead of upstart ) and in the initramfs in wiley?
[05:04] <micahg> I'm hoping adequate was the last one to fix for perl to migrate from proposed, if not, I'll take another look later
[05:11] <pitti> micahg: oh, you fixed lintian? nice, thanks!
[05:11] <pitti> Good morning
[05:12] <micahg> pitti: good morning, it seems that the fix was in the new version in Debian, just undocumented, I figured to pull the pieces in and get perl migrated and then worry about merging
[05:19] <pitti> wgrant: do we have anything like build recipes for git already? I suppose not directly, but can that be faked with putting some shell commands into the recipe or so?
[05:19] <pitti> ah no, that's just bzr-builder commands
[05:20] <wgrant> pitti: No recipes for git yet (they're currently implemented as a bzr plugin), and you can't execute arbitrary shell commands.
[05:20] <wgrant> For some Git repositories you could register a code import to get them into a Bazaar branch, though.
[05:30] <pitti> wgrant: yeah, that'd be a fallback; I still need to mess around with the patches somehow, but I think that can be done in debian/rules too; I would mostly just need to change debian/source/format
[05:30] <pitti> ... and hope that I can get bzr imports of upstream and Debian packaging of systemd
[05:31] <wgrant> https://code.launchpad.net/~registry/systemd/master
[05:31] <wgrant> imports fine
[05:31] <pitti> nice!
[05:32] <pitti> wgrant: other question, PPA buildds have all non-ancient kernels as they are in VMs, right?
[05:32] <wgrant> pitti: Yes, the virtual builders are all trusty with vivid sbuild
[05:34] <pitti> wgrant: cool, thanks
[06:15] <pitti> micahg: "Jenkins Fixed - wily-adt-adequate" \o/
[06:15] <pitti> micahg: ah, so that missing pkg-config depends affect Debian too, right?
[06:17] <Unit193> Debian #787043
[06:23] <pitti> doko_: FYI, filed bug 1459526 and subscribed you
[07:04] <dholbach> good morning
[07:41] <dholbach> larsu, HAPPY BIRTHDAY! :-)
[07:41]  * dholbach hugs larsu
[07:41]  * dholbach hugs larsu
[07:41]  * dholbach hugs larsu
[07:41] <larsu> thanks dholbach :)
[07:41]  * larsu feels very hugged and hugs back
[07:42] <dholbach> :-)
[07:47]  * seb128 hugs larsu as well :-)
[07:48] <larsu> thanks :)
[08:15] <Laney> @pilot in
[08:16]  * dholbach hugs Laney
[08:17] <Laney> \m/
[08:17] <Laney> still have some procrastination to do first, but the intention is there
[08:17]  * Laney hugs dholbach back
[08:19] <seb128> Laney, let's see if didrocks let you some easy ones after his shift on tuesday ;-)
[08:20] <didrocks> seb128: I took them ALL! :)
[08:20] <seb128> hehe
[08:20] <seb128> Laney, you are out of luck there, sorry ;-)
[08:20] <didrocks> only hard ones remaining for Laney
[08:20] <didrocks> he deserves this! :p
[08:20] <Laney> I can take it!
[08:20] <didrocks> *challenge accepted*
[08:20]  * Laney starts from the bottom :-)
[08:20] <didrocks> ahah
[08:21] <pitti> I'm piloting tomorrow, don't just leave the sucky ones to me :)
[09:08] <doko> stgraber, known, however I'm still unable to find the issue. not seen in unstable, only in vivid/wily. last known working version is the 5.1 release candidate. for now, I would suggest removing the powerpc binaries
[13:44] <rbasak> infinity: around? I'm merging apache2. We've been carrying some changes you uploaded in http://launchpadlibrarian.net/124947541/apache2_2.2.22-6ubuntu3_2.2.22-6ubuntu4.diff.gz for a while. I have a suspicion these are not needed any more, but I don't know how to verify that.
[13:45] <rbasak> Why did we need cross build for apache2 anyway?
[13:47] <infinity> rbasak: Because it was on someone's list of "packages that should be cross-buildable".
[13:48] <rbasak> infinity: I think it's all bitrotted now anyway.
[13:49] <infinity> rbasak: What proof do you have for that statement? :P
[13:49] <infinity> rbasak: I should have forwarded that to Debian (and I thought I did), but there's no reason any of it would have suddenly stopped working.
[13:49] <rbasak> infinity: "Fix cross-building by passing DEB_{HOST,BUILD}_GNU_TYPE to configure" no longer adds the environment variables you defined to the configure line.
[13:50] <infinity> rbasak: Hrm, indeed, looks like some bad merges along the way.
[13:51]  * infinity looks at the whole source instead of the delta.
[13:52] <rbasak> infinity: I can re-fix the configure line blindly if you'd like me to, and continue to maintain the delta if you wish. I don't recall ever seeing a merge conflict so it's not particularly awkward. I don't see anything relevant from a quick glance at Debian BTS though.
[13:52] <rbasak> infinity: but if nobody wants it or is using it and it has rotted, maybe that's a reason to save us the trouble and drop the delta?
[13:52] <infinity> rbasak: It's not just the configure line.  Someone dropped pretty much my entire delta on a merge at some point.  Probably a blind "well, the hunks don't apply, drop 'em!"
[13:54] <infinity> rbasak: Oh, I guess that mpm/module bit got removed entirely...
[13:54] <infinity> rbasak: So, just maintaining the 'pass HOST/BUILD to configure' bit should be simple enough, but someone should probably test a cross build to see if it still works.
[13:55] <infinity> rbasak: Also a bit confused that debian/patches/086_svn_cross_compiles still applies, given how long ago I backported that from trunk...
[13:56] <rbasak> infinity: 086_svn_cross_compiles still applies> yeah - given it was upstreamed I'd expect it to have arrived by now
[13:57] <infinity> rbasak: Anyhow, bad merges notwithstanding, we shouldn't drop the configure bit.  I'll (re-)forward that to Debian, but not until after I've tested that it actually still crosses.  Which I might not get to today, but we'll see.
[14:02] <rbasak> infinity: OK. So for now, shall I fix the configure line and upload my merge?
[14:04] <rbasak> infinity: anyway, I'll do that. Fancy writing a dep8 test to test your cross to stop future rot?
[14:05] <infinity> rbasak: I'm not sure DEP-8 tests in every single package to make sure it crosses is the right way to track that. :P
[14:05] <infinity> rbasak: A cross-build farm that attempts to cross the entire archive and watch for regressions is the right answer, but needs some tuits of varying roundness.
[14:08] <rbasak> infinity: sure. You can do that instead if you like :)
[14:29] <pitti> stgraber: ça va !
[14:30] <pitti> stgraber: so I naïvely tried this:
[14:30] <pitti> lxc-create -n wily-ppc64el -t download -- -d ubuntu -r wily -a ppc64el
[14:30] <pitti> lxc-start -n wily-ppc64el -F
[14:30] <pitti> stgraber: and it fails; but ISTR that LXC had some clever "run under qemu" integration? how do I make that work?
[14:31] <stgraber> pitti: oh, you tried that on x86?
[14:31] <pitti> stgraber: yes, amd64
[14:31] <stgraber> yeah, not gonna work :)
[14:32] <stgraber> the only one that vaguely works is armhf and that's when using the good old ubuntu template
[14:32] <pitti> stgraber: ok; maybe I mixed that up with mk-sbuild or so? I know that we had some way of running a schroot or container or so under qemu
[14:32] <pitti> stgraber: ah, ok
[14:32] <pitti> stgraber: ok, so back to looking for a porter box I figure :)
[14:32] <pitti> (if only we had sensible schroots on them, but let's see..)
[14:32] <stgraber> qemu-user-static sort of works for armhf, so long as you don't use ptrace or netlink (which is already pretty restrictive)
[14:32] <stgraber> other arches tend to segfault earlier on
[14:32] <pitti> stgraber: no, just want to reproduce a build failure
[14:35] <pitti> $ sudo apt-get build-dep umockdev
[14:35] <pitti> AttributeError: 'apt_pkg.SourceRecords' object has no attribute 'Lookup'
[14:35] <pitti> meh, porter box hates me today
[14:36] <pitti> (and it builds fine under Debian)
[14:37] <LocutusOfBorg1> what did happen to the ppa builders? all "cleaning" state
[14:37] <LocutusOfBorg1> are they cleaning the room? :p
[14:39] <pitti> stgraber: ok, nevermind; abusing a temp container on the ppc64el CI machines; thanks for the explanation!
[15:22] <seb128> hum
[15:23] <seb128> what should we do when packages are blocked proposed migration on "Boottest result: Regression"
[15:23] <seb128> where the error is "EnvironmentError: Unsupported device, autodetect fails device"
[15:24] <seb128> e.g https://jenkins.qa.ubuntu.com/job/wily-boottest-address-book-app/lastBuild/console
[15:24] <pitti> seb128: I think step 1 is to retry, step 1/2 yell at CI to fix it, step 3 to override
[15:24] <seb128> pitti, who in CI do you know?
[15:24] <pitti> seb128: ah, that would be "cihelp:" matter
[15:24] <seb128> thanks
[15:24] <pitti> seb128: just talk to the vanguard (they have rotational shifts)
[15:24] <seb128> pitti, thanks
[15:33] <Laney> @pilot out
[15:33] <Laney> oops!
[15:53]  * dholbach hugs Laney
[16:03] <LocutusOfBorg1> dholbach, do you think I have chances to get sponsored curl and cmake?
[16:03] <LocutusOfBorg1> I had some spare time to look at them
[16:03] <LocutusOfBorg1> :)
[16:04] <LocutusOfBorg1> (if you have free time I can do some merges in the next few days)
[16:06] <dholbach> LocutusOfBorg1, not right now
[16:06] <dholbach> I'm in meeting until 20:00 my time
[16:08] <LocutusOfBorg1> no problem at all :)
[18:27] <smoser> infinity, what needs to be done to make a vivid-netboot appear at http://archive.ubuntu.com/ubuntu/dists/trusty-proposed/main/installer-amd64/current/images/
[18:27] <smoser> and then ultimately at all arches and then -updates ?
[18:33] <zyga> ev: hi
[18:34] <zyga> ev: canonical irc is down for me
[18:34] <zyga> ev: have a look at the email I sent
[18:34] <infinity> smoser: I need to upload it.
[18:34] <zyga> ev: and if you want we can have a quick hangout if you have some questions but don't want to talk on irc
[18:35] <smoser> infinity, can you do that ? and to -updates even ?
[18:35] <infinity> smoser: I can't, since LP is broken. :P
[18:35] <smoser> oh yeah. that.
[18:35] <infinity> smoser: But yes, I had planned to finish it off this week.
[18:35] <zyga> oh, lp is down too?
[18:36] <sarnold> much of LP is down, among other services
[18:36]  * zyga takes note not to trip over those cables the next time
[18:36] <sarnold> irc.canonical.com was listed amont the affected services, but it has continued working for me and many others
[18:36] <sarnold> perhaps just new connections are affected, ldap lookups and so forth
[18:36] <zyga> sarnold: it doesn't work for me, so perhaps there's some round robin in the loop?
[18:37] <infinity> smoser: The proposed upload will probably happen this afternoon/evening, then like any SRU, it should get some validation before moving it to updates.
[18:37] <smoser> k
[18:39] <infinity> smoser: I'll subscribe you to the SRU bug, so you know when it's accepted, and can perhaps test an arch or two for me.
[18:48] <smoser> infinity, yeah. i can.
[21:03] <Unit193> So what would it take to get another ISO built from a seed?
[21:52] <zyga> how can I talk to is while the IRC is down?
[21:53] <Unit193> robert_ancell: Re: https://bugs.launchpad.net/ubuntu/+source/pidgin/+bug/1295207 I've been running with those gst1.0 patches since posting my comment there, no crashes.  All else that would need be done is a patch for pidgin-sipe.  This would be a great effort in getting gst0.10 gone! ;)
[21:54] <wgrant> zyga: #canonical-sysadmin
[21:55] <wgrant> zyga: What issues are you seeing connecting to IRC?
[21:55] <robert_ancell> Unit193, is there a patch / branch for that?
[21:57] <zyga> wgrant: I want to add a record to the incident report
[21:57] <Unit193> robert_ancell: There is, I linked to the Fedora one, and it's been merged into the unreleased upstream 2.x branch (who knows when they'll release), and I have a mostly-merged .11 with it from March(?): https://launchpad.net/~unit193/+archive/ubuntu/test/+sourcepub/4862307/+listing-archive-extra
[21:58] <wgrant> zyga: You should still be able to connect to irc.c.c
[21:58] <Unit193> robert_ancell: Need to go now though. :3
[21:58] <robert_ancell> Unit193, ok
[21:59] <zyga> wgrant: I cannot, since wednesday
[21:59] <wgrant> zyga: Uh
[22:00] <wgrant> Have you reported this?
[22:00] <zyga> wgrant: I was super busy (demo day today) so I ignored the problem
[22:00] <zyga> wgrant: then this happened and I assumed it was the cause
[22:00] <zyga> wgrant: (no, I have not)
[22:41] <robert_ancell> Unit193, I can't seem to download your pidgin changes https://launchpadlibrarian.net/201273988/pidgin_2.10.11-1~vanir1~15.04.debian.tar.xz
[23:05] <Unit193> robert_ancell: Launchpad is/was having issues.
[23:10] <robert_ancell> Unit193, do you have the patch files? Can you attach to the bug report or send them to me / pastebin them?
[23:15] <Unit193> It's been a while, it's all now only on LP. :/
[23:15] <Unit193> I can refresh the Fedora one, but it applies with only offsets.
[23:16] <Unit193> Dang this is annoying.
[23:26] <robert_ancell> Unit193, I just pushed pidgin 2.10.11 so I just need a patch file that appies against that. The one from Fedora has some conflicts
[23:40] <wgrant> robert_ancell, Unit193: It should be working now.
[23:41] <Unit193> wgrant: Fantastic, thanks.  Yes it does now. \o/
[23:41] <robert_ancell> wgrant, works here, thanks!
[23:51] <robert_ancell> Unit193, your package doesn't seem to have any patch for farsight 0.2 - does that seem correct?
[23:53] <robert_ancell> ah, it is in the gstreamer1.patch. Sorry, missed that.
[23:58] <Unit193> robert_ancell: Right, not the best merge ever, but at the time I was doing a quick one to try it in a PPA.
[23:58] <wgrant> Launchpad should be completely OK again. Let me know if anything looks awry.