#title #ubuntu-meeting Meeting Meeting started by slangasek at 15:00:41 UTC. The full logs are available at http://ubottu.com/meetingology/logs/ubuntu-meeting/2011/ubuntu-meeting.2011-10-19-15.00.log.html . == Meeting summary == *lightning round *Precise blueprints *Bugs *AOB Meeting ended at 15:46:39 UTC. == Votes == == Action items == * (none) == People present (lines said) == * slangasek (54) * ev (33) * cjwatson (25) * bdmurray (25) * mvo (17) * stgraber (16) * ubottu (16) * jhunt (15) * doko (12) * barry (11) * meetingology (3) == Full Log == 15:00:41 #startmeeting 15:00:41 Meeting started Wed Oct 19 15:00:41 2011 UTC. The chair is slangasek. Information about MeetBot at http://wiki.ubuntu.com/AlanBell/mootbot. 15:00:41 15:00:41 Available commands: #accept #accepted #action #agree #agreed #chair #commands #endmeeting #endvote #halp #help #idea #info #link #lurk #meetingname #meetingtopic #nick #progress #rejected #replay #restrictlogs #save #startmeeting #subtopic #topic #unchair #undo #unlurk #vote #voters #votesrequired 15:01:30 [TOPIC] lightning round 15:01:33 $ echo $(shuf -e cjwatson barry doko stgraber jhunt mvo ev bdmurray slangasek) 15:01:36 cjwatson mvo slangasek ev barry doko jhunt stgraber bdmurray 15:02:17 cjwatson: you win! 15:02:24 opening precise; lots of syncs, merges, NEW processing, build fixes, ... 15:02:24 main needs-merge graph looks really good so far, thanks to everyone who's been working on that 15:02:27 concentrating on the libjpeg transition, since that's big and caused several failures; also trying to keep on top of Pre-Depends additions for xz-compressed binary packages 15:02:30 looking into optimising cron.germinate; seems to be at least 7 minutes of room for improvement here; LP a bit non-committal on whether this would let us run the publisher */30, but might be possible 15:02:34 further optimisation of cdimage mirroring; now under 10 seconds for a no-op sync, and down to sensible times when we actually have to mirror anything 15:02:37 +1 maintenance team proposal 15:02:40 .. 15:03:10 apt: merge/upload new version to precise, work on cherrypick for #868977, #672314 to oneiric-proposed 15:03:10 software-center: add missing recommends on lzma for bug #868188, planning work, debug/fix steel storm2 desktop file bug, work on review languages and sorting, work on unicode_literals version (lp:~mvo/software-center/unicode_literals) to fight the str/unicode mess, add workaround for the white pixel border bug, look into startup time regression, commit one trivial workaround to trunk 15:03:10 update-manager: update for precise, SRU work, add postgresql to the removal blacklist and reupload (#871893), fix screensaver poke, fix crash when packages get downgraded (#873411, comment #8), honor dpkg desired install state (dpkg --set-selection state), some work on #859373, mostly diangose, cyhermox is the man 15:03:11 Launchpad bug 868188 in software-center (Ubuntu) ""Internal Error ...google-chrome-stable_current_amd64.deb could not be opened"" [Medium,In progress] https://launchpad.net/bugs/868188 15:03:14 apt-btrfs-snapshot: fix another issue with incorrect fstab lines (#873411, comment #7) 15:03:16 cups: test/upload fix for the upgrade issue #874835 15:03:18 misc: apt-clone, unattended-upgrades updates 15:03:20 review/sponsor compizconfig-backend-gconf upload for #207065 (yes, really low bugnumber) 15:03:22 apturl: debug/fix multiarch pkgname parsing (LP: #872146) 15:03:24 update-notifier: upload new -prposed version with fix for overlong menu text (#871691) 15:03:26 EOF 15:04:43 working on getting upstart into a state that it can be used in Debian (groundwork in sysvinit and lsb-base); maybe we can get a few more packages turned into syncs that way... 15:05:12 working through my own list of merges for precise 15:05:36 broke ia32-libs in precise - it'll fix itself when everything is multiarched 15:06:40 tracking various SRUs for oneiric (bug #874130, bug #871083, et al) 15:06:42 Launchpad bug 874130 in krb5 (Ubuntu Precise) "Canonicalize fallback only works for different realm (MITKRB RT #6917)" [High,Triaged] https://launchpad.net/bugs/874130 15:06:43 Launchpad bug 871083 in pam (Ubuntu Oneiric) "package libpam-modules 1.1.3-2ubuntu1 failed to install/upgrade: ErrorMessage: './usr/share/man/man8/pam_shells.8.gz' is different from the same file on the system" [Medium,Fix committed] https://launchpad.net/bugs/871083 15:06:47 EOF 15:07:31 - Release week sprinting. 15:07:32 - Fixed some interesting bugs in ubiquity found during smoke testing and 15:07:32 pushed out a new version into Precise. 15:07:32 - Chat with mvo on downloading external resources in dpkg (flash, 15:07:32 msttcorefonts), and queueing package installations in 15:07:33 software-center/update-manager (language packs at install). 15:07:33 - Meeting with Kate to discuss her requirements for the crash database and 15:07:33 metrics. 15:07:40 - Post-release ubiquity bug triage. 15:07:40 - Still lots of disk errors. Wubi will help here, but I wonder if we should 15:07:40 start looking for squashfs errors in syslog or if there's a better API for 15:07:40 this. 15:07:41 - Wrote Precise blueprints for uploading to trunk and rdepends testing. 15:07:41 - Created USB disks for the Ubuntu Shop and wrote up checksumming instructions 15:07:41 for the supplier, given Robbie's recent disaster. 15:07:42 - Investigated how and where we're using the gstreamer automatic codec 15:07:42 installation for Christian. 15:07:43 - Started researching Cassandra and other bits and pieces for crash and 15:07:43 metrics. 15:07:44 - Wrote a test for ubiquity to verify that we have a completely translated 15:07:44 interface. Annoyingly this requires a whitelist for a small set of widgets 15:07:45 that we translate on the fly, as the page they're associated with is run. 15:08:50 ev: I would love to hear more about the verify that the interface is translated test, I want this too for s-c :) 15:08:53 glad to see that NM regression fixed so quickly in SRU 15:09:02 s/regression/mid-upgrade breakage/ 15:09:09 so then it didn't get cut off? 15:09:11 translated or translateable? 15:09:15 IYSWIM 15:09:37 translated - it overrides translate_widget adn does the lookups itself 15:09:44 I assume you use a fake language like "mirror" or something that can be "auto-translated"? or how does it work? 15:09:54 I'd call that translateable I think, but OK, good :) 15:10:01 err yes 15:10:02 indeed 15:10:07 i often use a fake 'rot13' language for stuff like that :) 15:10:12 haha 15:10:14 nice as well 15:10:53 detection of missing translations - very nice :) 15:11:04 mvo: that's what they speak in Feedonia 15:11:12 er, Freedonia 15:11:41 barry: you're up 15:11:51 ah, sorry didn't see ev's "done" 15:11:52 syncs, merges, blueprints; investigated didrocks i18n crasher bug & wrote blog post; review rfc 6377; pypi testing experiment & TIP mailing list post; bug 813146 debugging; done. 15:11:53 15:11:55 Launchpad bug 813146 in linux (Ubuntu Precise) "kernel panic when running Python test suite on ecryptfs" [Critical,In progress] https://launchpad.net/bugs/813146 15:12:24 slangasek: it kept biting me (and writing the test even found one more missing one, yay) 15:13:06 ev: right, good demonstration of the value of automatic testing :) 15:13:17 :) 15:13:42 slangasek: one nice thing to the pypi testing work is that a few other related projects have come out of the woodwork. we're discussing merging/collaboration 15:14:37 pypants? 15:14:41 barry: great! 15:15:09 ev: those folks haven't spoken up, but pyti is one and pycheeseshop is another 15:15:15 cool 15:15:19 (pyti being a gsoc2011 project) 15:15:51 (i'm done btw) 15:16:10 - precise opening with some initial changes 15:16:10 - fix ARM libobjc and libjava bootstrap failures on GCC trunk 15:16:10 - work on component-mismatches, pester some people, fixed -perl packages 15:16:10 - some MIRs 15:16:10 - syncs and merges 15:16:12 done 15:16:24 In London last week for sprint. Spent a lot of time working on bug 15:16:24 849414 / bug 553745. Still unable to create. Reason this problem "comes 15:16:24 and goes" I think is due to us switching off apport post release. I've 15:16:24 requested users with the problem re-enable and send us some logs if they 15:16:24 can grab one. Now switched back to Upstart. I now understand the 15:16:27 TEST_ALLOC_FAIL issue I was seeing (TEST_ALLOC_FAIL is an evil but very 15:16:27 Launchpad bug 553745 in plymouth (Ubuntu Maverick) "plymouthd crashed with SIGSEGV in ply_event_loop_process_pending_events()" [High,Triaged] https://launchpad.net/bugs/553745 15:16:29 useful invention). Currently going through Upstart job logging test code 15:16:32 and enabling TEST_ALLOC_FAIL for it (this is slow as you need to track 15:16:34 each alloc call and the logger has various failure paths that vary the 15:16:37 number of alloc calls). Plan: finish Upstart tests and get this feature 15:16:39 into Precise "real soon now", get NIH multi-arched (started with 15:16:42 stgraber last week). Out this Friday. Wednesday Wish: that someone would 15:16:44 fix bug 861268 :) 15:16:46 Launchpad bug 861268 in xorg (Ubuntu) "text corruption in terminals (xterm, urxvt) and emacs" [Undecided,New] https://launchpad.net/bugs/861268 15:16:47 ☯ 15:17:12 - Was in London last week for the release sprint, testing/breaking/fixing the world 15:17:16 - Took Monday off 15:17:18 - Started looking at bug 876829 15:17:20 Launchpad bug 876829 in ifupdown (Ubuntu Precise) "Oneiric's ifupdown breaks ip aliases" [High,Triaged] https://launchpad.net/bugs/876829 15:17:21 - Registered a few specs for UDS, still quite a few more to discuss and register 15:17:24 - Spent a bit of time talking about app isolation (arkose, apparmor, arb-wrapper, ...) 15:17:27 - Got my first Precise container running 15:17:30 - Started discussing what's needed to make Edubuntu 12.04 an LTS release 15:17:32 - TODO 15:17:35 - Patch pilot this afternoon 15:17:37 - Go through the merges where I'm touched-it-last and hopefully grab a few from Colin's list 15:17:41 - Initial upload of ltsp and ldm in Precise so we can do some work on it at the LTSP hackfest next week (as the Debian maintainer will also be there) 15:17:44 - More pre-UDS planning (ISO tracker, recovery mode, local DNS resolver by default, Edubuntu, ...) 15:17:47 - Publish the automated IPv6 testing and get it running on Precise. Prepare a short demo for the testing in Ubuntu event at UDS. 15:17:50 - See how I can help James getting libnih ported to multi-arch so we can get non-x86 containers running on x86 hardware 15:17:53 - Try to spend some time on the ARB queue, we have quite a few apps in there that need packaging/review 15:17:54 jhunt: nice "done" character :) 15:17:56 (done) 15:18:03 doko: nice job with the MIRs; components-mismatches looks much better than the flurry of post-opening emails would have suggested 15:18:05 reassignment of no package apport-bug reports to the correct package 15:18:05 wrote a bug bot function to take care of these going forward 15:18:05 descrption based no package assignment testing 15:18:05 bug bot fix for private people and patch subscription 15:18:16 wrote and uploaded and apport-pacakge hook for synaptic 15:18:16 modified cheese apport package hook to file bugs about linux (LP: #793950) 15:18:19 data visualization brainstorming 15:18:22 fixing of bug stats scripts to take into account Incomplete changes 15:18:24 fixing bugstats-complete on cranberry to count private bugs 15:18:26 research into flashplugin-downloader bug 870281 and out of date mirrors 15:18:26 wrote a bugpattern for flashplugin bug 870643 15:18:26 wrote a bugpattern for 873397 (ubiquity adobe acrobat and archive.canonical.com) 15:18:28 Launchpad bug 870281 in flashplugin-nonfree (Ubuntu Oneiric) "installer crash when user choose to install additional software: http://archive.canonical.com/pool/partner/a/adobe-flashplugin/adobe-flashplugin_10.3.183.10.orig.tar.gz doesn't exist" [High,Fix released] https://launchpad.net/bugs/870281 15:18:29 sru of bug fix for bug 207065 for Lucid 15:18:30 Launchpad bug 870643 in flashplugin-nonfree (Ubuntu) "package flashplugin-downloader 11.0.1.152ubuntu1 failed to install/upgrade: wget: unable to resolve host address `archive.canonical.com'" [High,Confirmed] https://launchpad.net/bugs/870643 15:18:32 Launchpad bug 207065 in compizconfig-backend-gconf (Ubuntu Lucid) "Bad Compiz Bindings Bug" [Low,Fix committed] https://launchpad.net/bugs/207065 15:18:32 barry: ☺ 15:18:34 ubiquity bug cleanup of install failures due to media errors on hard disks (not just cd drives) 15:18:37 wrote a udd query for ubuntu bug reports with the same bug watch 15:18:40 done 15:19:56 861268 - eep 15:21:33 slangasek: not to minimise doko's work, certainly, though some of the component-mismatches stuff was because I'd been a bit too hasty in telling it to look at armhf 15:21:49 cjwatson: ah 15:21:59 turning that back off avoided some of the insanity, after doko pointed it out to me earlier today 15:22:17 well, still ~30 -perl left, and ~25 -java (resolved by jamespage) 15:22:42 bdmurray: could we have http://reports.qa.ubuntu.com/reports/foundations-bugs/foundations-bugs-next-milestone-important-bug-tasks.html pointed at oneiric-updates for a bit? 15:22:44 but it's indeed understandable again 15:23:07 people really should get an email, when a package enters into dep-wait 15:23:13 slangasek: sure 15:23:17 doko: are you dealing with promotions as you go along, or should I be watching the mir-track list? 15:23:18 bdmurray: thanks! 15:23:27 doko: does that not happen? I thought it did 15:23:29 cjwatson, yes doint so 15:23:42 slangasek, did you see one for your cron merge? ;-P 15:23:56 doko: great, thanks 15:23:59 doko: I saw mail for it, but that might be just because I receive the component-mismatches diffs! :D 15:24:15 ahh, I don't get those 15:24:36 I'll MIR that soon, fwiw 15:24:53 [TOPIC] Precise blueprints 15:25:00 (not to be confused with "precise blueprints") 15:25:27 public service announcement... please get blueprints filed by the end of this week, so we can get them on the UDS schedule in time to let the schedule *settle* 15:25:32 any questions? :) 15:25:34 * mvo nods 15:25:59 slangasek: I've sent you an email about that 15:26:21 ev: will reply right after the meeting 15:26:25 cheers 15:26:28 [TOPIC] Bugs 15:26:53 bdmurray: any interesting bugs popping out of the woodwork post-release? 15:27:10 slangasek: this report has quite a few on it http://reports.qa.ubuntu.com/reports/foundations-bugs/foundations-bugs-recently-important-bug-tasks.html 15:28:47 bug 836378 for example 15:28:48 Launchpad bug 836378 in grub2 (Ubuntu) "Oneiric upgrade, boot hangs at error: symbol not found: grub_divmod64_full" [High,Confirmed] https://launchpad.net/bugs/836378 15:28:52 hmm, bug #874518 looks like a duplicate of bug #874130 15:28:54 Launchpad bug 874518 in openssh (Ubuntu) "ssh with kerberos fails after upgrade to 11.10" [High,New] https://launchpad.net/bugs/874518 15:28:54 Launchpad bug 874130 in krb5 (Ubuntu Precise) "Canonicalize fallback only works for different realm (MITKRB RT #6917)" [High,Triaged] https://launchpad.net/bugs/874130 15:29:03 bdmurray: yet another case of misconfiguration, not new 15:29:18 cjwatson: you have a bucket for that bug then? 15:29:22 basically *every* "symbol not found" error in grub2 is a local misconfiguration 15:29:40 I'll reply to it 15:30:01 heh, I guess those would be hard to bugpattern 15:30:05 yep 15:30:10 since they tend to be filed by hand 15:30:36 well the bug bot could do something though 15:31:53 cjwatson: can you work with bdmurray to figure out what the bug bot should do? 15:33:37 bug #877194 looks like a system that hasn't been properly upgraded to natty in the first place.. 15:33:38 I guess 15:33:39 Launchpad bug 877194 in update-manager (Ubuntu) "do-release-upgrade in console to oneric fails: SystemError: E:Could not perform immediate configuration on 'multiarch-support'. Please see man 5 apt.conf under APT::Immediate-Configure for details. (2)" [High,Incomplete] https://launchpad.net/bugs/877194 15:33:54 I don't know if I'm comfortable with them being auto-closed though 15:34:04 I think people would prefer to interact with a human here rather than a bot 15:34:10 cjwatson: your call 15:34:44 just auto mark them as incomplete, then it doesn't have to be on your conscious when they're closed ;) 15:34:50 cjwatson: I don't think closing is necessary but giving people some direction to head in or tagging the bugs so they can be dealt with in batches may help 15:36:01 how about tagging any grub2 bug whose summary matches "symbol not found" as grub-modules-skew, then? 15:36:39 that sounds fine to me 15:37:23 and maybe ask for the output of 'sudo debconf-show grub-pc | grep install_devices' 15:37:25 what about bugs on grub1 (which is where this one was filed)? 15:38:19 there's also bug #858122 which is a major pain for upgrading... we really need to deploy some workaround for these users that have had insserv run on their systems against their will. someone want to volunteer to look at that? 15:38:21 Launchpad bug 858122 in Release Notes for Ubuntu "incomplete migration to /run (shutdown script order has been demolished)" [High,In progress] https://launchpad.net/bugs/858122 15:39:50 slangasek: might there be duplicates of that bug lying around / hidden elsewhere? 15:39:57 certainly 15:40:06 and what would I look for? 15:40:09 probably not easily autodetectable 15:40:09 grub(1)> yes 15:40:24 I've followed up to bug 836378 now with detailed explanation/directions, and marked it invalid 15:40:26 Launchpad bug 836378 in grub2 (Ubuntu) "Oneiric upgrade, boot hangs at error: symbol not found: grub_divmod64_full" [High,Invalid] https://launchpad.net/bugs/836378 15:40:36 I haven't looked at all yet into 858122 but if noone else will I can give it a go 15:40:58 bdmurray: well, if you can figure out a search term that matches "user rebooted after upgrade and their system didn't come up because dbus was missing and replacing /var/run with a symlink fixed it"... :) 15:41:19 mvo: appreciate it! 15:41:47 mvo: fwiw, I think the most appropriate workaround here is probably to shove the standard symlinks in /etc/rc6.d back to their intended sequence numbers by hand on upgrade 15:42:04 so that, y'know, sendsigs, unmount, and reboot all happen in the intended order :P 15:42:30 if we fix *that*, the migrate-on-shutdown code should Just Work 15:42:38 slangasek: heh, that makes sense :) 15:43:00 we also need some fix for insserv itself breaking things, though :/ 15:43:22 not sure what kind of fix would be SRUable there... I almost want to move insserv off the path, or make it a no-op 15:44:13 mvo: anyway, you can have a look and see what you think :) 15:44:27 any other bugs that aren't getting the attention they need? 15:45:09 not that I know of 15:45:28 [TOPIC] AOB 15:45:39 who knows a joke? 15:46:39 #endmeeting Generated by MeetBot 0.1.5 (http://wiki.ubuntu.com/AlanBell/mootbot)