=== Guest89941 is now known as Zic === dosaboy_ is now known as dosaboy === smb` is now known as smb === dosaboy is now known as dosaboy__ === Ursinha-afk is now known as Ursinha === jono is now known as Guest19501 === james_ is now known as Guest73337 === cjohnston_ is now known as cjohnston [13:01] * barry waves [13:01] hey [13:01] stgraber, ondra, ogra_: heyo [13:01] moo [13:02] barry: should we start with download service? [13:02] lool: sure [13:02] so, i uploaded a new version last night with dbus support [13:03] stgraber: pointed to a problem with config file updates: https://bugs.launchpad.net/ubuntu-system-image/+bug/1204090 [13:03] Ubuntu bug 1204090 in Ubuntu system image "Drop 'device' from client.ini" [Critical,In progress] [13:03] so i'll be getting an 0.8 out this morning [13:03] had a chat with seb yesterday. his team was going to start work on the update ui. i'll circle round to him today, though i am on patch pilot for at least half the day [13:04] barry: so even after deleting client.ini today, I'll have to repeat every day until this is solved? [13:04] we sketched out a very basic interaction pattern, so i think the minimal ui will be doable [13:04] lool: hey [13:04] lool: yeah, but hopefully "every day" will be just "today" [13:05] 0.8 will have only this fix and then you won't have to edit client.ini to set the device model any more [13:05] anyway, i think that's about it for the client [13:05] barry: so when do you think you would have a simple UI for an update? (even if partial) [13:05] d/l service. i have to circle back to mandel to see where/when/if that is landing [13:06] barry: download service should be all landed now [13:06] lool: ah, good. [13:06] lool: i'm still aiming for tomorrow [13:06] (on the ui [13:06] ) [13:06] https://launchpad.net/ubuntu/+source/ubuntu-download-manager [13:07] libubuntudownloadmanager1 and ubuntu-download-manager are in the images [13:07] barry: ok [13:07] excellent. i'm still not going to integrate that until after the demo most likely [13:07] done [13:07] barry: yeah; we had noted the download service as an action on you last week, just wanted to close this topic [13:07] since it's all landed now [13:07] +1 [13:07] oh wow, we even have unity-scope-click in the image too ! [13:08] lool: beuno was going to demo that afaik [13:08] sergiusens: yeah [13:08] so I think stgraber isn't around [13:09] he was to land gpg in recovery [13:09] but basically everything is landed, updater is announced for adventurous folks [13:09] isnt that n since days [13:09] ? [13:09] and we had various issues at the time of launch and since the launch [13:09] but getting resolved [13:09] though we havent released images since 5 days due to the new policies [13:10] ogra_: stgraber pulls from pending to rebuild the images [13:10] ogra_: yeah but update imgaes are generated from latest image rather than current, so it still works [13:10] ogra_: new policies? [13:10] (and i expect releases to not happen for a while until all bits ship full tests that utah can run) [13:10] barry, 100% test coverage on the dashboard now [13:10] ah [13:10] lool, well, we will need a full test of the bits that got added to recovery ... [13:10] barry: now that I read it more carefully, I don't understand how LP 1204090 is critical; is that in the case where the model detection would change over time?!? [13:11] Launchpad bug 1204090 in Ubuntu system image "Drop 'device' from client.ini" [Critical,In progress] https://launchpad.net/bugs/1204090 [13:11] theoretically we couldnt release atm [13:11] after asac's definition [13:11] ogra_: not sure what you mean [13:11] ogra_: you mean add new testsuites to cover the new bits? [13:11] lool, we add ubuntu code to recvery that has zero tests [13:11] by the new definition that isnt releasable [13:12] so testing gpg and the ubuntu updater scripts would be needed [13:12] and proper integration into utah [13:12] lool: no. if future changes are made to client.ini, they most likely won't get installed (at least easily) since conffile will see a local edit. i think the point is to remove the need for almost everyone to edit their client.ini file. [13:12] barry: oh ok [13:12] barry: sorry, I get it now [13:12] barry: I thought it was entirely generated [13:12] but it's a conffile *and* gets updated after install, and wont get updated on upgrades anymore [13:12] ok [13:13] nope. maybe i'll port gentoo's conffile update system to debian :) [13:13] barry: there's ucf too [13:13] yeah [13:13] but this only concerns people who have renabled apt, right?> [13:13] it's better to query the system for the device name anyway [13:13] otherwise you get the new conffile overwritten by a system image update? [13:14] i think that's right [13:14] ogra_: so we do have tests in certain places, but not for everything [13:14] lool, right [13:14] ogra_: we also lack end-to-end integration tests [13:14] and new features need to ship tests alongside [13:14] yeah [13:15] currently all we have are the app tests and default installability [13:15] bit for the recovery changes we need tests too [13:15] I'm not sure how practical it would be to have them run daily; I don't really see how that would reasonnably be done, but we could e.g. run update tests from the latest release or something like that [13:15] and there should be a test that prevents stuff like above as well [13:15] (broken configs etc) [13:15] lool: the end to end stuff we can discuss in our meeting with QA later today [13:15] sergiusens: yeah [13:16] http://reports.qa.ubuntu.com/smokeng/saucy/ [13:16] thats what we depend on now [13:16] and the next daily will only go out if all is green [13:16] ogra_: right; but we need a base image to smoke test an update [13:16] (which i suspect might take until the weekend) [13:17] lool, right [13:17] I think we covered all actions from last week [13:17] i feel that daily-release should always use last /current + the proposed changes [13:17] in that way we foster a culture of delivering something that will work in one shot [13:17] asac, that makes only sense if /current is actually daily [13:17] rather than saying you can have multiple shots to get things right (e.g. we have an option to easily see results against /rpoposed) [13:17] ogra_: sure. it only makes sense if we go strict [13:18] so I guess the only remaining open topics are: a) barry adding UI (high) a') barry fixing LP 1204090 b) barry switching to download service (low) c) sergiusens / ogra_ / lool discussing integration tests [13:18] Launchpad bug 1204090 in Ubuntu system image "Drop 'device' from client.ini" [Critical,In progress] https://launchpad.net/bugs/1204090 [13:18] lool: seems reasonable [13:18] on /etc ... I think it makes sense to put a zero /etc in any touch image policy [13:18] lool, sounds fine [13:18] because its a half broken concept in the new world of system image updates [13:18] zero /etc ? [13:18] can you elaborate ? [13:18] at best defaults are never shipped in /etc/ anyway, and we can provide other locations to change defaults through .DOT files [13:19] asac: it's more of a template [13:19] lool: those should go to /usr/share/docs :) [13:19] asac: how does that work with convergence? [13:19] right now in the linux world /etc is not cleanly designed [13:19] barry, we'll find out in the 14.04 cycle :) [13:19] its used as one place for a) defaults (that should usually go into the app) and b) admininstrative/system wide changes [13:20] ogra_: :) [13:20] i think we hsould go clean about those two approaches in touch :) [13:20] thats all i am saying (now i am out again) [13:20] asac: I don't see a big gain in pushing defaults out of /etc [13:20] this is make-work [13:21] (like merging /usr/bin and /bin or renaming /etc to /config or something) [13:21] anyway [13:21] its not consisten behaviour [13:21] * ogra_ thought /etc is readonly anyway [13:21] well, at least let's say an irc comment leaves it a bit underspecified :) [13:21] if apt is there [13:21] seems we're getting a bit out of the topic [13:21] variable data needs to go into /data [13:21] ogra_: not if apt is back i am sure [13:21] asac, ah, no, then not indeed [13:21] ogra_: exactly; except in convergence, or if you swithc to developer mode [13:22] lool, i dont see why convergence would have to differ :) [13:22] but thats 14,04 stuff [13:22] not for now [13:22] ogra_: because of the installer user bsae [13:22] as in, today my /etc is r/w on my laptop [13:22] ogra_: if /etc is read only, where does network mananger setup it's configs? [13:23] sergiusens: /dev/mmcblk0p9 on /var/lib/NetworkManager type ext4 (rw,relatime,user_xattr,acl,barrier=1,data=ordered) [13:23] sergiusens, in a tmpfs overlay ? no idea [13:23] sergiusens: bind mounts to writable space [13:23] * ogra_ needs to look into the setup this week [13:23] * sergiusens needs to check [13:23] i had planned to do an install this week [13:23] sergiusens: sorry, I meant: /dev/mmcblk0p9 on /etc/NetworkManager/system-connections type ext4 (rw,relatime,user_xattr,acl,barrier=1,data=ordered) [13:23] yeah [13:23] thats /data [13:23] lool: and I thought you changed the location :-) [13:24] the other is for more ephemeral stuff [13:24] like dhcp leases and stuff [13:24] they go into /run anyway, no ? [13:24] sergiusens: so basically anything which used to write to /var or /etc needs to be diverted to /data [13:24] yeah [13:24] yeah :) [13:25] ogra_: not that ephemeral [13:25] and everything that is in /etc should rewally go to doc (template) or /usr/share/ (default settings) [13:25] ogra_: you could remove it, but it's best to keep across reboots [13:25] :() [13:25] e.g. which ssids you've seen and such [13:25] yeah [13:25] asac, whats the benefit ? [13:25] * ogra_ doesnt see it [13:25] asac: I don't think we need to spend engineering cycles moving stuff around in gazillions of packages [13:26] if /etc is readonly anyway, why would you want to have apps read from a non std place [13:26] yeah, lool expressed that better [13:26] seems like a waste [13:26] it's kind of funny to have this argument in this direction with asac [13:26] well, i will cover this topic elsewhere... short term its not on topic [13:26] heh [13:27] asac: you're supposed to be more pragmatic than me! and I ought to advocate beauty of the file hierarchy :-) [13:27] i am pragmatic [13:27] alright, almost out of time [13:27] * ogra_ sees no beauty in moving configs from /etc to /doc :) [13:27] barry, ondra, ogra_, sergiusens: Anything else we want to cover? [13:27] * barry is good [13:27] also, do we want to keep this meeting? [13:27] lool, do we know who will write tests ? [13:27] perhaps until it's used by everyone? [13:27] (thats why i brought it up_ [13:28] i know stgraber will not be around (vacations, etc) [13:28] lool: i think we're far enough along that it most likely won't be necessary for much longer [13:28] and we need to switch at some point [13:28] lool: not much from me, I haven't done much, only recent activity was to help stgraber to resolve gpg problem, but that was 1 tas [13:28] but we cant without having tests in place [13:28] ogra_: I would think stgraber would write upgrade tests when he's back? [13:29] dunno, i think i remember something about three weeks [13:29] barry: I'm thinking we should cancel next week then have one more meeting to plan enabling for everyone [13:29] with parts of that being on/off and parts of that being gone completely [13:29] barry: or would you be interested in writing the upgrades tests in the next weeks? since stgraber is on leave, you might be a good candidate since you know the upgrade logic inside out :-) [13:29] barry: would you want to join the call we have with QA folks later? [13:30] ogra_: yeah; release sprint, debconf and other leaves [13:30] lool: when's the call? [13:30] lool, right [13:30] barry: 4pm UTC [13:30] barry: adding you [13:30] barry: feel free either way [13:31] alright, closing this meeting them; have a good day everyone! [13:31] lool: sure, np [13:31] *then [13:31] cheers! [13:31] bye all [13:31] bye [13:31] see you on the channel to the left (or right) (irssi) [13:31] bye [13:31] sergiusens, pfft ... below [13:31] (xchat) [13:32] lol [15:32] o/ [15:34] oops [16:00] o [16:00] o/ [16:00] hello [16:00] o/ [16:01] \o [16:01] o/ [16:01] lets strt [16:02] #startmeeting [16:02] Meeting started Tue Jul 23 16:02:00 2013 UTC. The chair is jamespage. Information about MeetBot at http://wiki.ubuntu.com/meetingology. [16:02] 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 [16:02] #topic review previous actions === meetingology changed the topic of #ubuntu-meeting to: review previous actions [16:02] rbasak to land delta report to lp:ubuntu-reports, Daviey to deploy [16:02] rbasak, Daviey- hows that going? [16:03] zul: [16:03] zul has willfully volunteered to review and document what is blocking server related saucy-proposed migration excuses, with joy. [16:03] I think thats all sorted now right [16:03] ? [16:03] bah [16:03] almost apache2 again i think [16:03] rbasak: ^^^ [16:04] anyone actually here? [16:04] lol [16:04] anyway [16:04] #action rbasak to land delta report to lp:ubuntu-reports, Daviey to deploy [16:04] ACTION: rbasak to land delta report to lp:ubuntu-reports, Daviey to deploy [16:04] my bot is... [16:04] carried forward in their absence [16:04] #topic Saucy Development === meetingology changed the topic of #ubuntu-meeting to: Saucy Development [16:05] #link https://wiki.ubuntu.com/SaucySalamander/ReleaseSchedule [16:05] Debian Import Freeze is on the 25th July - manual syncs after that point will be required [16:05] alpha2 same week - we will be opting out again [16:06] also worth noting that 12.04.3 is just under a month away [16:06] the QA team now have some dkms auto-testing setup [16:06] openvswitch and iscsitarget are going to need some attention in the next two weeks. [16:06] ? [16:06] (as are lots more I guess) [16:07] hallyn, hardware enablement kernels [16:07] Oops, sorry. ubuntu-reports is almost ready for an MP. apache2 is done. I'm preparing a merge with an upgrade path bug and SSL passphrase prompting fixed. [16:07] 12.04.3 will get rarings 3.8 kernel [16:07] rbasak, marvellous! [16:07] jamespage: gotcha [16:07] next [16:07] #subtopic Release Bugs [16:07] #link http://reqorts.qa.ubuntu.com/reports/rls-mgr/rls-s-tracking-bug-tasks.html#server [16:08] I still need to look at asm4 transitions [16:08] and quantum service respawns [16:09] * jamespage berates himself again [16:09] lua5.2 is pending from review [16:09] ack [16:09] there's a yucky memleak in libvirt i need to look into [16:09] rrdtool and rubyluabridge [16:09] bug 1202653 should be targetted for Saucy. It's in hand though - I just want to fix SSL certificate prompting at the same time, write a dep8 test for that, and in any case proposed migration will be blocked as it's seeded in alpha-2 opt-in flavours. [16:09] zul: https://bugs.launchpad.net/ubuntu/+source/nova/+bug/1199791 [16:09] bug 1202653 in apache2 (Ubuntu) "apache2 2.4.4-6ubuntu4 failed to upgrade: Cannot load /usr/lib/apache2/modules/libphp5.so into server: undefined symbol: unixd_config" [High,In progress] https://launchpad.net/bugs/1202653 [16:09] Ubuntu bug 1199791 in nova (Ubuntu Saucy) "nova-compute-xcp misses nova-compute.conf" [High,Triaged] [16:09] rbasak, done [16:09] in progrress [16:09] thanks [16:09] zul, great - thanks [16:10] looks like adam_g's cheetah fixup is almost done [16:11] yup, should make its way out today [16:12] rbasak, whats the latest on bug 1031680? [16:12] bug 1031680 in nagios-plugins (Ubuntu Raring) "check_apt always report 0 critical updates" [High,Triaged] https://launchpad.net/bugs/1031680 [16:13] and that covers release bugs.... [16:13] #subtopic Blueprints [16:13] No progress. [16:13] #link http://status.ubuntu.com/ubuntu-s/group/topic-s-servercloud-overview.html [16:13] https://blueprints.launchpad.net/ubuntu/+spec/servercloud-s-mongodb [16:13] use of libv8 in the mongodb shell is causing some security concerns re supportability [16:13] which is understandable [16:14] looking at a shell disabled variant of the package - but I suspect thats worse than useless [16:14] so we might be inthe position where mongodb is not suitable for main inclusion [16:14] https://blueprints.launchpad.net/ubuntu/+spec/servercloud-s-juju-2-delivery [16:14] juju2 release going much smoother now [16:14] two in the last week [16:15] (one for OSCON) [16:15] local provider has also landed if no-one noticed - works well (thanks thumper) [16:15] roaksoax, https://blueprints.launchpad.net/ubuntu/+spec/servercloud-s-openstack-charms-ha-v2 [16:15] whats happening with HA stack for saucy? I know we have had a few conversations? [16:16] https://blueprints.launchpad.net/ubuntu/+spec/servercloud-s-ceph reappears [16:16] utlemming, https://blueprints.launchpad.net/ubuntu/+spec/servercloud-s-webscale [16:16] utlemming, are you likely to have capacity to work on this blueprint this release? its already a carry over from raring [16:17] zul: https://blueprints.launchpad.net/ubuntu/+spec/servercloud-s-openstack-qa [16:17] needs some updates [16:18] jamespage: ack [16:18] quite a few activities need to be worked on still this cycle there - might be work allocating out and postponing anything we know we won't get to this cycle [16:18] adam_g, ^^ [16:19] s/work/worth/ ? [16:19] agree [16:19] hallyn, yes [16:19] jamespage: sorry i'm supper lagged [16:19] k [16:19] roaksoax, ack [16:19] roaksoax, wanna update now? [16:19] * jamespage gives roaksoax a few to respond [16:20] zul: https://blueprints.launchpad.net/ubuntu/+spec/servercloud-s-openstack-havana [16:20] lots of things actually done there I think [16:20] good point to catchup now that havana2 is landing everywhere [16:20] ill upate it today [16:21] arosales, quite a few juju related ones with minimal progress - do you want to go through those now? [16:21] https://blueprints.launchpad.net/ubuntu/+spec/servercloud-s-openstack-charms [16:21] adam_g, roaksoax and I have been working on the python redux of the openstack charms [16:22] looking quite good now - hopefully we should have a full set of initial versions this week [16:22] im hoping to have all of the bash -> python rewrites ready for testing this week [16:22] \o/ [16:23] OK - anyone got any other BP related stuff they wanted to raise/query? [16:23] .... [16:23] ... [16:23] .. [16:23] mine of server banners is stuck, working on other issues [16:23] . [16:24] yolanda, is that just pending sponsorship or are you getting blocked? [16:24] jamespage, sorry I was in another window [16:24] jamespage, i stopped working on that for a while, working on syncs and merges, and now in charms [16:24] yolanda, OK [16:25] arosales, juju blueprints? [16:25] i got some of them uploaded, bind9 is pending on review [16:25] jamespage, a couple of the owners are at OSCON so I can take an action to review juju related ones this week [16:25] #action arosales to review Juju blueprints [16:25] ACTION: arosales to review Juju blueprints [16:25] #action arosales to review juju related blueprints with owners after OSCON [16:25] ACTION: arosales to review juju related blueprints with owners after OSCON [16:25] ooops [16:25] oh well - we will remember twice now! [16:25] lets move on [16:25] #topic Ubuntu Server Team Events === meetingology changed the topic of #ubuntu-meeting to: Ubuntu Server Team Events [16:26] something called OSCON I think [16:26] anyone going? [16:26] i'm guessing if they are, they are there already [16:26] lol [16:26] yes [16:26] any other events folk want to raise now? [16:26] m_3 and jcastro @ OSCON [16:26] giveing a Charm School talk as we chat here [16:27] excellent - hope its packed.... [16:27] #topic Weekly Updates & Questions for the QA Team (plars) === meetingology changed the topic of #ubuntu-meeting to: Weekly Updates & Questions for the QA Team (plars) [16:27] plars, floor is yours [16:27] hi, a couple of bugs are affecting server testing at the moment [16:27] both on the virtual host part [16:27] https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1203694 and https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1203211 [16:28] Ubuntu bug 1203211 in linux (Ubuntu) "duplicate for #1203694 Modprobe doesn't recognize any parameters on 3.10.0-4" [Critical,Fix committed] [16:28] Ubuntu bug 1203211 in linux (Ubuntu) "Modprobe doesn't recognize any parameters on 3.10.0-4" [Critical,Fix committed] [16:28] should be fixed by -5 in proposed [16:28] bot seem to be well on their way though :) [16:28] ack [16:28] AIUI the release team are working on landing -5. [16:28] excellent [16:28] sounds good [16:28] that's it from me [16:28] thanks plars [16:28] anyone got anything else qa related while we are here? matsubara maybe? [16:29] .... [16:29] ... [16:29] .. [16:29] . [16:29] not really [16:29] ack [16:29] #topic Weekly Updates & Questions for the Kernel Team (smb) === meetingology changed the topic of #ubuntu-meeting to: Weekly Updates & Questions for the Kernel Team (smb) [16:29] There seem also to be some problem with qemu and vmvga crashing the host which apw is working on. [16:30] The bug number slipped my brain though [16:30] ah [16:30] bug 1204005 [16:30] bug 1204005 in linux (Ubuntu) "[saucy] kvm host hangs of guest boot with 3.10.0-5" [Critical,Triaged] https://launchpad.net/bugs/1204005 [16:31] Apart from that I cannot think of anything hot right now (exccept the weather) [16:31] Are there questions? [16:31] smb, thanks [16:31] sound like a no for questions... [16:31] #topic Weekly Updates & Questions regarding Ubuntu ARM Server (rbasak) === meetingology changed the topic of #ubuntu-meeting to: Weekly Updates & Questions regarding Ubuntu ARM Server (rbasak) [16:32] Nothing new to report. Any questions for me? [16:32] not from me [16:32] anyone else? [16:33] OK [16:33] #topic Open Discussion === meetingology changed the topic of #ubuntu-meeting to: Open Discussion [16:33] anything anyone wants to raise? [16:33] UDS fastly approaching [16:33] discuss or otherwise argue about? [16:33] http://uds.ubuntu.com/ [16:33] yeah i raised this [16:33] jamespage: basically, I'm currently updating the cluster stack, which means the removal of 'redhat-cluster' from the archives. All cluster will be based on pacemaker+corosync2.0, gfs2 + dlm (a new package that needs to be uploaded) [16:33] for distributed storage [16:33] /filesystem [16:33] roaksoax, man - that is some lag - thanks for the update [16:34] https://bugs.launchpad.net/ubuntu/+source/quantum/+bug/1203766 [16:34] Ubuntu bug 1203766 in quantum (Ubuntu) "Please remove from Ubuntu Archive" [Undecided,New] [16:34] and all packages that used to depend on redhat-cluster related dependencies need to be update [16:34] ah sorry for the dup zul [16:34] good point about UDS _ only a month away [16:34] good point to recheck on progress for saucy so far [16:34] that should be interesting [16:35] just to be clear - quantum is now called neutron if anyone missed that [16:35] its not another new core project in openstack [16:35] :-) [16:35] anyway [16:35] #topic Announce next meeting date and time === meetingology changed the topic of #ubuntu-meeting to: Announce next meeting date and time [16:36] Tuesday 30th July - 1600 GMT [16:36] see you then [16:36] #endmeeting === meetingology changed the topic of #ubuntu-meeting to: Milestone Targeted Work Items (ogasawara) [16:36] Meeting ended Tue Jul 23 16:36:25 2013 UTC. [16:36] Minutes (wiki): http://ubottu.com/meetingology/logs/ubuntu-meeting/2013/ubuntu-meeting.2013-07-23-16.02.moin.txt [16:36] Minutes (html): http://ubottu.com/meetingology/logs/ubuntu-meeting/2013/ubuntu-meeting.2013-07-23-16.02.html [16:36] thanks folks! [16:51] jamespage, thanks for chairing [17:00] #startmeeting [17:00] ## [17:00]