=== czchen is now known as czchen_
=== czchen_ is now known as czchen
=== giraffe is now known as Guest52692
=== JanC is now known as Guest69133
=== JanC_ is now known as JanC
[02:07] <tsimonq2> Any chance I can get some help kicking imagemagick through zesty-proposed?
[02:08] <tsimonq2> I'll do some more local testing, but it would help to get it through for Alpha 2.
[02:08] <tsimonq2> http://people.canonical.com/~ubuntu-archive/proposed-migration/update_output.txt - it shows it breaks things
[02:08] <tsimonq2> Like I said, I'll play with it locally, but if someone has a solution, please ping.
[02:09] <tsimonq2> bug 1550210 is in our release notes
[02:09] <ubottu> bug 1550210 in imagemagick (Ubuntu) "Desktop file does not open ImageMagick from the menu" [Medium,Triaged] https://launchpad.net/bugs/1550210
[03:08] <krytarik> tsimonq2: You should better focus your efforts on this one then though: https://launchpad.net/debian/+source/imagemagick/8:6.9.7.4+dfsg-1
[03:10] <tsimonq2> krytarik: I'll file a bug tomorrow
[03:10] <tsimonq2> krytarik: Thanks, nice catch.
[03:11] <juliank> So, fairly close to a workaround for the APT issue.
[03:11] <tsimonq2> juliank: \o/ \o/ \o/
[03:11] <juliank> It turns out that the whole thing is caused by (a) by-hash producing the same URI for empty icons-128x128 and icons-64x64 and (b) apt merging targets with the same URI
[03:12] <juliank> The targets then ended up in some kind of different state, and the whole thing went a bit strange.
[03:12] <tsimonq2> Fun.
[03:12] <juliank> So what I just tried out is only merging files for the same destination.
[03:13] <juliank> That fixes the issue, but breaks a minor part in the test suite testing the merging targets works.
[03:14] <juliank> I'll go to sleep now and hope DonKult has a better idea in the next hours... (it's 4 am now!)
[03:14] <tsimonq2> Thanks juliank :)
=== giraffe is now known as Guest5935
[08:48] <cpaelzer> hi, anybody ever hit "../sysdeps/nptl/fork.c:156: __libc_fork: Assertion `THREAD_GETMEM (self, tid) != ppid' failed." on our platforms that do container based testing?
[08:49] <cpaelzer> I ran into that driving a nova test for a proposed qemu update, but from my search I'm almost convinced it is unrelated to the actual thing that you test
[08:49] <cpaelzer> This is how it looks for me https://objectstorage.prodstack4-5.canonical.com/v1/AUTH_77e2ada1e7a84929a74ba3b87153c0ac/autopkgtest-zesty-ci-train-ppa-service-2379/zesty/s390x/n/nova/20170124_144900_ec0dd@/log.gz
[08:50] <cpaelzer> all efforts on it so far seem to have stalled to a halt https://patchwork.kernel.org/patch/9225661/ https://lkml.org/lkml/2015/2/6/470 https://sourceware.org/bugzilla/show_bug.cgi?id=15392
[08:50] <ubottu> sourceware.org bug 15392 in nptl "Linux, setns, PID namespaces, fork: assert() about pid inequality hit sporadically" [Minor,New]
[08:51] <cpaelzer> Since it seems to be something that might just "randomly happen" I wanted to ask if anybody else has seen that?
[08:52] <cpaelzer> for now I'm rerunning it to check if it is really random for me as well
[10:54] <juliank> Whoa, snapcraft tests take a long time.
[11:03] <juliank> Laney, tsimonq2: OK. I got a complete workaround (the right fix?) for the icon tarball hashsum mismatch issue. I can upload that now. I wonder if that affects anyone else than kubuntu, though, and I don't know if the other workaround (do not fetch 128x128 icons) is already enabled.
[11:04] <Laney> juliank: plasma-discover is the package that enables both - I don't see an upload changing that
=== marcusto_ is now known as marcustomlinson_
=== marcustomlinson_ is now known as marcustomlinson
[11:12] <juliank> Laney: OK, I cherry-picked that from its branch and the fix for partial requests on sourceforge from master and uploaded it :)
[11:12] <juliank> the partial request thing was #1657567
[11:12] <juliank> bug 1657567
[11:12] <ubottu> bug 1657567 in apt (Ubuntu) ""Content-Range: */<file size>" on non-416 responses considered invalid" [Low,Fix committed] https://launchpad.net/bugs/1657567
[11:13] <Laney> juliank: many thanks ;-)
[11:13] <Laney> we can kick a new kubuntu build once apt goes in
=== _salem is now known as salem_
=== salem_ is now known as _salem
=== _salem is now known as salem_
=== hikiko is now known as hikiko|ln
=== hikiko|ln is now known as hikiko
=== salem_ is now known as _salem
[13:40] <fossfreedom_> cjwatson: re this ISO boot issue (screenshot in the bug report) - is this some sort of isolinux theming issue - if so, any ideas how we can resolve this for Ubuntu Budgie? https://bugs.launchpad.net/ubuntu/+source/syslinux/+bug/1659278
[13:40] <ubottu> Launchpad bug 1659278 in syslinux (Ubuntu) "Ubuntu Budgie does not boot directly to ubiquity try/install screen" [Undecided,New]
[13:41] <cjwatson> IIRC there's some horrible thing buried in debian-cd or something
[13:43] <cjwatson> fossfreedom_: yeah, see lp:~ubuntu-cdimage/debian-cd/ubuntu tools/boot/zesty/boot-*, look for HIDDEN_TIMEOUT in particular
[13:44] <cjwatson> the semantics are a bit of a mess but are explained in the gfxboot-theme-ubuntu changelog, items for 0.9.3 and 0.9.9
[13:44] <cjwatson> (sorry ...)
[13:50] <fossfreedom_> cjwatson: oh... that is making my head spin trying to understand that.
[13:51] <rbasak> cyphermox: are you aware of the network-manager and network-manager-applet SRUs in Xenial?
[13:51] <rbasak> Just wondered if you were still involved and/or had an opinion.
[13:53] <cjwatson> fossfreedom_: it is unpleasant, sorry - but I think the above is probably enough information to fix the bug
[14:12] <fossfreedom_> cjwatson: k - thanks. think I understand this.  How can I test a proposed merge request?
[14:12] <cjwatson> fossfreedom_: la la la
[14:12] <cjwatson> fossfreedom_: best is probably to remaster an image with hidden-timeout=2 or whatever hacked into the relevant config file, but you'll need to figure out where ...
[14:13] <cpaelzer> jamespage: coreycb: I've hit THREAD_GETMEM in the nova dep8 Test triggered by the install of ca-certificates
[14:13] <cpaelzer> jamespage: coreycb: I've asked about that about 5.5h before now in the chan, but nobody else responded to have seen it
[14:14] <cpaelzer> jamespage: coreycb: since it shows up "again" on s390 on my retest I wonder how common that might be - have you seen those issues?
[14:14] <jamespage> cpaelzer, I'll defer to coreycb and zul for that answer
[14:14] <coreycb> cpaelzer, i've never seen it unfortunately
[14:16] <cpaelzer> coreycb: this was my former question with a bit more details https://irclogs.ubuntu.com/2017/01/25/%23ubuntu-devel.html#t08:48
[14:16] <cpaelzer> coreycb: really never - weird
[14:16] <zul> cpaelzer:i dont remember seeing that either
[14:17] <cpaelzer> thanks zul and coreycb
[14:18] <cpaelzer> I'm just puzzled on how random it might be as I hit it on ppc&s390 at first and s390 only on the retry
[14:18] <coreycb> xnox, have you come across this at all on s390? ^
[14:18] <coreycb> cpaelzer, it sounds like it's container related then
[14:18] <cpaelzer> seems to be able to happen on any of the autppkgtest environments that use containers
[14:18] <cpaelzer> yes that is what I found from my debugging this morning
[14:19] <cpaelzer> or even "pid namespace" related to be even more specific
[14:19] <mitya57> Mirv, hi! Do you have any plans to update qtcreator? It currently blocks qbs migration because it was built against old version, and with new qbs it FTBFS (https://launchpad.net/~ci-train-ppa-service/+archive/ubuntu/2401/+packages).
[14:19] <coreycb> cpaelzer, yeah i'm surprised i haven't seen it before
[14:21] <cpaelzer> for now I'll rerun it again to see how transient it is, but the 3/4 I've seen so far do not appear as "almost never" to me
=== _salem is now known as salem_
[14:32] <coreycb> cpaelzer, possibly related: https://groups.google.com/forum/#!msg/linux.kernel/G7aBRlEGHqw/tXT9cJV_AwAJ
[14:33] <coreycb> cpaelzer, https://sourceware.org/bugzilla/show_bug.cgi?id=15392
[14:33] <ubottu> sourceware.org bug 15392 in nptl "Linux, setns, PID namespaces, fork: assert() about pid inequality hit sporadically" [Minor,New]
[14:33] <coreycb> cpaelzer, https://lkml.org/lkml/2015/2/6/470
[14:37] <cyphermox> rbasak: sorry, I can look but I don't know off the top of my head
[14:42] <cpaelzer> coreycb: yeah that is the bug I linked this morning
[14:49] <coreycb> cpaelzer, oh, sorry :)
[14:50] <cpaelzer> coreycb: no reason to excuse
[14:50] <cpaelzer> coreycb: I just hope you didn't spend too much time to re-find that
[15:23] <Mirv> mitya57: hi. not special plans, feel free to update though as the delta is smallish these days and easy to keep.
[15:31] <mitya57> Mirv, OK, will try to update it.
[15:34] <rbasak> xnox: please can you explain _why_ you make changes in your changelog entries? Take bug 1657909 for example, which is in limbo because we don't know why you made a change.
[15:34] <ubottu> bug 1657909 in gnome-keyring (Ubuntu Yakkety) "SSH service should be hidden from Startup Applications too" [Low,In progress] https://launchpad.net/bugs/1657909
[15:34] <rbasak> xnox: also if you could explain in the bug, that would be useful.
[15:46] <fossfreedom_> andyrock: Hi - seb128 signed you up on this bug report affecting Ubuntu Budgie!! https://bugs.launchpad.net/ubuntu/+source/budgie-desktop/+bug/1631745  ... any chance you can have a quick glance over my proposed patch to gnome-menus please?
[15:46] <ubottu> Launchpad bug 1631745 in gnome-menus (Ubuntu) "hostname-panel crashed with SIGSEGV in g_slice_alloc()" [High,In progress]
[15:48] <andyrock> mmm the fix is not a real fix
[15:49] <andyrock> we need a proper bt or valagrind log to solve it
[15:54] <fossfreedom_> k - my intention is not to affect any other desktop like Unity - just needed to get budgie-desktop to run through the original code because the problem does not occur with that codebase
[15:54] <fossfreedom_> valagrind doesnt help here - it doesnt crash - I guess its a race condition of somesort and the slow running through valgrind doesnt exercise the issue.
[16:17] <andyrock> gnome-panel does not crash
[16:19] <andyrock> it can be that the problem is in budgie-desktop main menu code
[16:19] <andyrock> fossfreedom_ i'm ok with that patch
[16:20] <andyrock> but I'm wondering why you can't just distro patch directly in budgie-desktop
[16:20] <fossfreedom_> andyrock: yeah  - discussed with upstream - they couldn't reproduce this so would not deal with the issue (understandably)
[16:22] <fossfreedom_> budgie-desktop is vala based.  I don't have any vala knowledge myself so can't patch it.  I only know C and python :/
[16:23] <andyrock> oki
[16:23] <andyrock> seb128 is not online
[16:23] <andyrock> we need to talk with him
[16:30] <juliank> Laney: I just received the email that apt was accepted :)
[16:31] <juliank> Exciting times!
[16:31] <Laney> juliank: nice to know we have solid tests :P
[16:32] <Laney> when rmadison agrees, we can fire the rebuild
[17:13] <nacc> cpaelzer: around?
[17:23] <cjwatson> fossfreedom: I think if you know C and Python then modifying existing Vala code should be pretty straightforward (maybe harder if you have to write significant new code)
[17:24] <cjwatson> to simplify, it's basically C with Javaish syntax layered over the top and approximately Python-like library bindings
[19:06] <coreycb> doko, a fix for python-glanceclient is in the trusty review queue.  bug 1404227
[19:07] <coreycb> https://bugs.launchpad.net/python-glanceclient/+bug/1404227
[19:07] <fossfreedom> cjwatson: Myself and upstream have looked at the budgie-code, there doesnt seem to be anything wrong in the way it is coded - proper locking to stop reentrant type issues etc.  The menu code really does assume gnome-menus does all the heavy lifting - it only uses the output.  So given where this issue is I dont see a way to intercept/integrate any code without rewriting masses of gnome-menu code :/
[19:07] <ubottu> bug 1404227 in python-glanceclient (Ubuntu) "Unit test failures due to changes in python 2.7.9" [Undecided,New] https://launchpad.net/bugs/1404227
=== salem_ is now known as _salem
=== _salem is now known as salem_
=== salem_ is now known as _salem
[22:01] <camako> Anyone know why I'm getting this?
[22:02] <camako> /usr/bin/ld: cannot find -lpthreads
[22:02] <camako> This is on a silo
[22:02] <camako> .. trying to build qtmir
[22:02] <camako> https://launchpadlibrarian.net/303893883/buildlog_ubuntu-zesty-amd64.qtmir_0.5.1+17.04.20170125.1-0ubuntu1_BUILDING.txt.gz
[22:05] <mwhudson> camako: that should be -lpthread surely
[22:05] <mwhudson> camako: without the s
[22:06] <camako> mwhudson, oh I wonder why it's with 's'
[22:06] <camako> mwhudson, I am blind... thanks for spotting it
[22:07] <mwhudson> camako: np :)
[22:39] <bdmurray> juliank: These apt autopkgtest failures for the SRU are false positives right?
[22:40] <juliank> bdmurray: ack
[23:59] <gpiccoli> Hi folks, greetings! I found a weird "bug" on libgtk/gtk2-engines-pixbuf, I manage to fix in my system, but worth the report I guess