15:00:15 <adamw> #startmeeting Fedora QA meeting
15:00:15 <zodbot> Meeting started Mon Aug 13 15:00:15 2012 UTC.  The chair is adamw. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:00:15 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
15:00:19 <adamw> #meetingname fedora-qa
15:00:19 <zodbot> The meeting name has been set to 'fedora-qa'
15:00:26 <adamw> #topic roll call and snacks
15:00:29 <adamw> happy, tflink? :)
15:00:40 * pschindl is here
15:00:43 * jskladan free food?
15:00:58 * mkrizek here
15:01:17 <adamw> jskladan: no, no, no, the idea is you all bring ME snacks
15:01:17 * Martix is ready for collisions
15:01:28 * kparal heard something about food
15:01:43 * spoore is here...lurking
15:01:45 * tflink is around
15:02:00 <tflink> kparal: that was for the docs meeting, unfortunately :'(
15:02:02 <jskladan> adamw: aaah, that's the trick - I'll have something for you prepaired on DevConf ;)
15:02:31 <Martix> jskladan: something spicy?
15:02:32 <tflink> adamw: wait a second, I think we've been tricked
15:02:38 <adamw> =)
15:03:25 <jsmith> tflink: Docs team is sneaky like that :-p
15:04:01 <adamw> #topic F18 status, TC testing preparation
15:04:11 * nirik lurks
15:04:12 <adamw> sooo...in case you didn't notice, we have this thing called fedora 18 to test...
15:04:51 <adamw> Alpha TC1 landed over the weekend, we don't have the validation matrix in place yet
15:04:55 <adamw> first order of business, does it work? at all?
15:04:56 <kparal> we have
15:05:02 <kparal> the matrix
15:05:14 <kparal> pschindl's done that
15:05:47 <kparal> he also claims it doesn't boot, and if you make it boot, it doesn't install
15:05:58 <adamw> ah, excellent, thanks for catching up on that
15:06:15 <tflink> it sounds like the anaconda devs are aware of the issue and are working on it, though
15:06:23 <tflink> unless there's another one that I'm not following
15:06:27 <adamw> ok, so alpha tc1 testing is complete, great job everyone =)
15:06:43 <adamw> #info F18 Alpha TC1 released over the weekend, initial testing indicates it fails to boot or install successfully
15:07:10 <adamw> #info pschindl updated installation results matrix template
15:07:39 <adamw> #chair tflink paral
15:07:40 <zodbot> Current chairs: adamw paral tflink
15:07:42 <adamw> grr
15:07:43 <Martix> and if you upgrade to F18 and boot with 3.6 kernel, you hit boot loop bug https://bugzilla.redhat.com/show_bug.cgi?id=847418
15:07:48 <adamw> #undo
15:07:48 <zodbot> Removing item from minutes: <MeetBot.items.Info object at 0x24152790>
15:07:54 <adamw> #info F18 Alpha TC1 released over the weekend, initial testing indicates it fails to boot or install successfully
15:07:59 <adamw> siiiigh. can someone please feed me coffee?
15:08:03 <adamw> #chair kparal
15:08:03 <zodbot> Current chairs: adamw kparal paral tflink
15:08:12 <adamw> hi, robatino
15:08:16 <nirik> Martix: fixed in recent systemd... make sure you have new systemd and re-gen your initramfs.
15:08:16 <tflink> adamw: you re-did the wrong #info
15:08:24 <adamw> #undo
15:08:24 <zodbot> Removing item from minutes: <MeetBot.items.Info object at 0x1cc9a710>
15:08:27 <Martix> nirik: noted, thanks
15:08:31 <adamw> #info pschindl updated installation results matrix template
15:08:35 <nirik> at least that worked for me. ;)
15:09:23 <adamw> #info bug #847418 is also causing issues for f18 users, recommended action: update systemd and re-generate initramfs
15:09:40 <adamw> ok, the other thing i had on my list for this topic is revising the criteria and test cases
15:09:59 <adamw> anyone noticed any required changes already?
15:10:23 <adamw> we'll likely have to edit a large number of the install validation test cases to reflect newui; we may need to adjust the criteria for missing functionality still
15:11:17 <tflink> I thought we were still waiting somewhat to see what all lands in newui in time for F18
15:11:24 <pschindl> adamw: I think that we should wait for working tc and then change everything
15:11:45 <adamw> pschindl: yeah, having a tc that vaguely works will be a help =)
15:12:03 * kparal also votes for changing "everything" :-)
15:12:19 * jskladan would do it real time soon now
15:12:30 <pschindl> I know about few changes, that will be needed in criteria, but I'm not sure about how it will work in new ui
15:12:47 <adamw> pschindl: are those the ones you already proposed to the list, or others?
15:13:39 <pschindl> adamw: those was only the most critical, which we should resolve firstly
15:13:53 <adamw> okay
15:14:21 <adamw> it'd be good to get the other proposals in soon
15:14:25 <pschindl> the other changes are mostly about rewording current criteria
15:14:41 <adamw> as for the test cases, i think we can probably just all chip in and update them as we go along
15:14:52 <adamw> or if someone wants to take a task for doing it systematically that's fine too
15:15:39 <adamw> #info pschindl has some further criteria change proposals to come
15:17:36 <adamw> ok, so to summarize our plan of attack: 1) figure out what's wrong with TC1, 2) get a TC2 that works, 3) revise test cases/criteria for newUI changes
15:17:52 <adamw> does that cover everything f18-y? anything I missed?
15:18:04 <jsmith> WORKSFORME
15:18:20 * nirik is a bit worries about lives... but hopefully we will have something that works by the end of the week
15:18:24 <pschindl> yep. I talked to anaconda team and fix could be done till tomorrow
15:19:03 <adamw> nirik: ah, right, lmc, that's the other thing
15:19:24 <adamw> nirik: so you tried to build lives with lmc and had issues?
15:19:40 <nirik> I have not yet...
15:19:41 <tflink> I thought it was the lack of koji support more than anything
15:20:00 <adamw> ah, okay. so we haven't even reached the 'trying' stage yet =)
15:20:15 <adamw> #info live image generation is still marked HERE BE DRAGONS
15:20:16 <tflink> not in the same env as the lives have always been built in, no
15:20:24 <nirik> there are no lives with tc1... I don't know if that is due to an issue dgilmore hit or if he didn't get to trying until the koji support is done.
15:20:48 <tflink> nirik: the TC1 ticket says that it's due to no koji support but he's hoping to get that fixed this week
15:20:54 <jsmith> I got the idea that we was still working on the koji support
15:21:00 <jsmith> but it wasn't super-clear
15:21:12 <nirik> ok.
15:21:56 <adamw> #info still waiting on koji support for livemedia-creator
15:23:09 <adamw> alrighty. well, that gives us lots of stuff to fix =)
15:24:01 <adamw> #topic multi-image validation
15:24:13 <adamw> so i just threw this on the agenda since it's something that's actively in draft at present
15:25:07 <kparal> I replied to your emails. current proposal looks good
15:25:10 <adamw> if anyone hasn't seen it, i put together a draft validation matrix for the multi-desktop and multi-boot images we've been building for the last couple of releases, and we've been trying to refine it: https://lists.fedoraproject.org/pipermail/test/2012-August/109284.html
15:25:15 <adamw> kparal: yup, thanks a lot for your feedback
15:25:33 <adamw> if anyone hasn't seen it, please take a look and contribute any ideas you have...anyone have any thoughts on it that aren't in the thread?
15:27:42 <adamw> okey dokey, then
15:29:02 <adamw> moving along
15:29:05 <jsmith> adamw: I read through it, but didn't have any spectacular ideas -- the proposal looked solid
15:29:05 <adamw> #topic AutoQA update
15:29:08 <adamw> just for old times' sake
15:29:10 <adamw> jsmith: thanks
15:29:17 <adamw> any big autoqa news lately, autoqa folks?
15:29:42 <kparal> I enabled F18 testing in AutoQA
15:29:58 <kparal> that mostly concerns repoclosure, fileconflict and rats_* tests
15:30:06 <adamw> cool. do we know the status of rats wrt newui?
15:30:21 <adamw> or is it hampered by the fact we don't appear yet to have any recent newui-based images that work? =)
15:30:27 <tflink> no text mode == no rats
15:30:30 <kparal> rats_install is currently broken, I'll work on it once we have usable TC
15:30:32 <adamw> oh, yeah.
15:30:42 <kparal> text mode should be included eventually, shouldn't it?
15:30:48 <adamw> yeah, we don't have a date yet though.
15:30:55 <tflink> maybe but there is no guarantee that it'll make F18
15:31:00 <kparal> hrm, so that's a blocker
15:31:10 <kparal> but I'll look at it anyway
15:31:13 <adamw> thanks.
15:31:21 <kparal> rats_sanity should work ok, but doesn't have a maintainer
15:31:30 <adamw> #info kparal enabled f18 testing in autoqa
15:31:38 <adamw> #info rats_sanity is without a maintainer
15:31:44 <kparal> I hoped twu would take over from hongqing, but he doesn't seem to be receiving my emails
15:32:07 <kparal> so still looking for volunteers
15:32:41 <kparal> another news, autotest package review is very close to an end
15:32:45 <adamw> yaaay
15:32:47 <kparal> so kudos to mkrizek
15:32:51 <adamw> #info autotest package review is almost complete
15:32:56 <adamw> thanks mkrizek!
15:33:23 <kparal> also kudos to garretraziel (jan sedlak), who wrote an autotest_results_cleanup.py script for us
15:33:43 <kparal> that makes the results directory half in size and inodes
15:33:55 <adamw> hey, that's neat.
15:34:20 <kparal> and that's it for the last week I suppose
15:35:15 <adamw> coolness.
15:35:33 <adamw> on a more general topic: what's our progress like towards the 'enforce package tests' goal?
15:36:16 <tflink> not much, unfortunately
15:36:38 <kparal> I was about to reply 'none at all'
15:36:41 <kparal> at least lately
15:37:02 <adamw> i think tflink was putting a positive spin on 'none at all' =)
15:37:19 <tflink> eh, it's more than 0 ... just not much more than 0 :)
15:37:27 <adamw> well, that sucks. let's do not that in future. i have now discharged my managerial obligations!
15:37:28 <kparal> we're terrible project planners
15:38:06 <adamw> #info not much progress recently towards the 'enforce package tests in bodhi' goal
15:38:18 <adamw> okay, anything else for autoqa, anyone?
15:39:39 <kparal> nope
15:40:08 <adamw> alrighty
15:40:11 <adamw> #topic open floor
15:40:43 <tflink> Just a reminder: The blocker review meetings will be on Wednesday at 16:00 UTC starting this week
15:40:55 <adamw> so, anyone have anything else to talk about? the weather, who should be the drummer for the who, cat grooming tips...
15:41:06 <adamw> okay, OR you could be boring and say something on-topic and to the point, yeesh.
15:41:17 <adamw> #info The blocker review meetings will be on Wednesday at 16:00 UTC starting this week
15:41:52 * kparal likes
15:42:02 <tflink> at some point (hopefuly soon), the blocker tracking app will be moved to fedora infra and the source will be released
15:42:29 <kparal> have we decided on the hostname already?
15:42:37 <adamw> #info the new blocker tracking page/webapp will soon be moved to fedora infra
15:42:57 <tflink> there was some question on whether using qa.fedoraproject.org would mess with the internal subdomain
15:43:36 <tflink> no decision was reached, no
15:43:52 <tflink> too many nos in that sentance
15:43:53 <kparal> I don't personally like using subdomain in this case
15:44:11 <tflink> kparal: not sure I follow you there, maybe I'm not being clear
15:44:31 <tflink> there is no external qa.fp.o subdomain - it only exists internally
15:44:33 <nirik> it should/will be reachable via qa01.dev.fedoraproject.org... but if you want another cname going to it, we just need to decide what. ;)
15:44:59 <nirik> and setting it to qa.fedoraproject.org might work, or might blow up the internal only qa.fedoraproject.org zone.
15:45:16 <tflink> apparently, nirik types faster than I do :)
15:45:27 <kparal> so we don't know whether it would work?
15:45:31 <adamw> what's an 'internal only qa.fedoraproject.org zone' when it's at home?
15:45:37 <nirik> I don't. I guess we could try it, or pick another name.
15:45:48 <kparal> let's push the red button
15:45:54 <kparal> and try it
15:46:02 <nirik> adamw: there is a 'qa.fedoraproject.org' zone... it's not public on the net tho, it's only available to servers in phx2.
15:46:17 <adamw> oh, i see.
15:46:19 <tflink> adamw: it's all of the autoqa machines, mostly
15:46:22 <nirik> so autoqa01.qa.fedoraproject.org is an internal dns name...
15:46:31 <nirik> but that won't resolve in the public internet.
15:46:34 <kparal> the internal zone can be renamed, if needed?
15:46:43 <nirik> sure, it would be a hassle, but sure.
15:46:51 <kparal> hmm
15:47:03 <nirik> or you could just do 'qa-blockers.fedoraproject.org' or something?
15:47:12 * nirik doesn't know it needs too much bikesedding.
15:47:13 <kparal> but public qa.fedoraproject.org is so pretty.....
15:47:27 <nirik> if thats what you guys prefer, we can try it/do it. ;)
15:47:28 <kparal> and I imagine more apps will run there in the future
15:47:34 <kparal> not just blocker bugs app
15:47:54 <tflink> kparal: the vm we'll use isn't meant to be a permanent solution
15:48:21 <kparal> no, but the VM can be changed and the hostname can stay, right?
15:48:27 <nirik> yeah, it would/should move to a qa01.stg then a 'qa01' in production when it's ready.
15:48:37 <nirik> yes, changing it is easy, it's just dns. ;)
15:48:59 <kparal> if it's not much of a problem, I'd like to try whether public qa.fp.o works
15:49:07 <nirik> sure, we can.
15:49:11 <kparal> if it doesn't, let's decide on another name
15:49:22 <tflink> kparal: says the guy who's about to sign off and won't be around to fix any explosions :-P
15:49:29 <tflink> :-D
15:49:41 <nirik> hum.
15:50:07 <kparal> hiding in a bunker is the best way to survive, you know
15:50:16 * tflink is also for trying it, anyways
15:50:18 <nirik> we already seem to have a qa.fedoraproject.org record.
15:50:26 <nirik> but I am unsure where it points.
15:50:44 <adamw> it doesn't work out here.
15:50:46 <kparal> doesn't resolve for me
15:50:59 <Martix> qa.fedoraproject.org has address 209.132.181.30
15:51:08 <nirik> yeah, thats what I get... ^
15:51:25 <tflink> same here
15:51:40 <nirik> I'll try and track it down.
15:51:53 <nirik> I don't know what would break by repointing it. ;)
15:52:16 <tflink> any other suggestions for hostname if the qa.fp.o doesn't work out?
15:52:41 <kparal> let's discuss it when it really doesn't
15:52:48 <tflink> works for me
15:52:55 <nirik> f18blockers.fedoraproject.org ? ;)
15:53:01 <nirik> thats fine too.
15:53:08 * nirik will investigate, get back to you all
15:53:15 <kparal> thanks
15:53:15 * tflink needs to learn him some pgsql admin before the app comes up, anyways
15:53:19 <adamw> OK
15:53:31 <adamw> so, anything else?
15:54:04 <jsmith> tflink: I'm happy to help in that regard, if you have specific questions or need some help
15:54:47 <tflink> jsmith: I just have no idea how to set up pgsql, to be honest. I'm trying to work through the docs ATM on a local VM before I set up the dev vm
15:56:06 <adamw> alrighty, sounds like we're about done
15:56:16 <jsmith> tflink: OK, just general install pgsql, set up access, create users and database -- that sort of thing?
15:56:17 <adamw> thanks for coming out everyone...looks like we have lots of stuff to do =)
15:56:23 <jsmith> tflink: I'll help you after the meeting
15:56:31 * adamw sets fuse for X minutes
15:56:39 <tflink> jsmith: yeah, this sounds like something non-meeting related
15:57:19 * jsmith cuts the fuse in half
15:58:16 <adamw> meeting explodes in jsmith's face
15:58:19 <adamw> #endmeeting