fedora-qa
LOGS
15:00:15 <jlaska> #startmeeting Fedora QA Meeting
15:00:15 <zodbot> Meeting started Mon Jun 14 15:00:15 2010 UTC.  The chair is jlaska. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:00:15 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
15:00:20 <jlaska> #meetingname fedora-qa
15:00:20 <zodbot> The meeting name has been set to 'fedora-qa'
15:00:32 <adamw> yo
15:00:40 <jlaska> #topic Gathering critical mass
15:00:42 <jlaska> adamw: howdy
15:01:26 <hicham> i wan't to talk about drafting mozilla addons packaging guidelines
15:01:32 <hicham> there are none for the moment
15:01:35 <jlaska> we'll be without wwoods today
15:02:10 * kparal waves
15:02:10 * jskladan is here!
15:02:15 <jlaska> hicham: You're welcome to discuss that topic in the open floor.  But just a heads up, this is the quality assurance meeting, you probably want to raise that topic with the package reviewers?
15:02:22 <jlaska> howdy kparal jskladan
15:02:38 <jlaska> wwoods is on PTO today, and hopefully public holiday for Liam and Rhe
15:02:41 <hicham> jlaska : is there another meeting for that ?
15:02:52 <adamw> i think that'd be devel group stuff? is there an official devel group meeting?
15:03:35 <jlaska> I was going to suggest sending your questions to https://admin.fedoraproject.org/mailman/listinfo/packaging
15:03:47 <jlaska> I've seen a lot of similar discussion topics there.
15:03:57 <jlaska> adamw: would you suggest the devel mailing list?
15:04:19 <jlaska> hicham: http://fedoraproject.org/wiki/PackageMaintainers/Join#Join_the_important_Mailing_Lists
15:04:49 <jlaska> hicham: specifically, I think you might be looking for http://fedoraproject.org/wiki/Packaging/Committee#Guideline_Change_Procedure
15:04:50 <adamw> no, you seem right with packaging list.
15:05:04 <jlaska> alright, let's get started
15:05:06 * adamw feels like a Kafka character
15:05:18 <jlaska> #topic Previous meeting follow-up
15:06:04 <jlaska> #info jlaska will discuss CommonBugs entry for RHBZ #552423 with halfline
15:06:12 <jlaska> #info Completed, see https://fedoraproject.org/wiki/Common_F13_bugs#gnome-panel-crash
15:06:18 <adamw> i believe you should take your issue to Committee 493, Subcommittee 2398-A
15:06:28 <jlaska> adamw: :)
15:06:55 <jlaska> #info adamw will propose a document 'What do proventesters test for?'
15:07:14 <adamw> done!
15:07:21 <adamw> https://fedoraproject.org/wiki/User:Adamwill/Draft_proventesters_instructions
15:07:24 <jlaska> saweeet!
15:07:47 <adamw> it's mostly just talking points for now
15:07:51 <adamw> i'm not happy with how word-y it is
15:08:04 <adamw> but if people think it's going in the right direction, i'll make it tighter with a second draft
15:08:19 <jlaska> adamw: awesome, thanks for drafting.  I'll queue up for a post-meeting read
15:08:40 <adamw> i'll post it to the ml after the meeting
15:08:45 <jlaska> thanks
15:08:54 <kparal> with quick read it sounds good
15:09:18 <jlaska> #info jlaska to check-in with lmacken on the status of https://fedorahosted.org/bodhi/ticket/424
15:09:27 <adamw> thanks kparal
15:09:41 <jlaska> I've sufficiently annoyed lmacken by email, so I'll wait to hear back.
15:09:54 <jlaska> Unless anyone else has updates on when bodhi karma will be re-enabled for critpath updates
15:10:09 <adamw> nup.
15:10:14 <lmacken> jlaska: sorry, i'm teaching professors all week.. I'll get back to you soon :)
15:10:30 <jlaska> lmacken: no worries, you're always quick to respond ... so I assumed you were swamped
15:10:55 <jlaska> lmacken: I'll turn off *annoy* mode for now! :D
15:10:57 <adamw> or velociraptored
15:11:06 <jlaska> heh
15:11:21 <jlaska> this next item was really a follow-up to the above
15:11:23 <jlaska> #info Someone needs to check with bodhi on when 'proventester' karma is required for critpath packages
15:11:24 * lmacken on a conference call with the NY State senate now... 230 conference call with Yahoo!... just hired 2 co-ops that we're mentoring today, and teaching RIT profs all week... a little busy :)
15:11:31 * Viking-Ice sneaks in a bit late as usual..
15:11:38 <jlaska> lmacken: no kidding, very cool!
15:11:41 <jlaska> Welcome Viking-Ice :)
15:11:50 <jlaska> next ...
15:11:53 <jlaska> #info jlaska to update https://fedoraproject.org/wiki/QA/Join  to include link to proventester page
15:12:00 <jlaska> well, adamw did this for me already!
15:12:10 <adamw> oh, i thought that was on my list =)
15:12:15 <adamw> yeah, i've updated various places in the wiki
15:12:18 <jlaska> I was going to ask about some thoughts on cleaning up the [[QA/Join]] page
15:12:23 <jlaska> but adamw went to town on this
15:12:38 <adamw> https://fedoraproject.org/wiki/QA , https://fedoraproject.org/wiki/QA/Join , https://fedoraproject.org/wiki/QA/Updates_Testing are all updated
15:12:45 <jlaska> adamw: awesome, thank you :)
15:13:00 <adamw> i noticed on the way through we should probably update https://fedoraproject.org/wiki/Critical_Path_Packages_Proposal too
15:13:09 <adamw> currently it talks about
15:13:16 <jlaska> I'll send some thoughts I had on cleaning up QA/Join to the mailing list
15:13:16 <adamw> 'qa/releng' doing the testing
15:13:25 <jlaska> adamw: ah, good catch
15:14:22 <jlaska> should we keep the 'proposal unchanged' and move stuff to https://fedoraproject.org/wiki/Critical_Path_Packages ?
15:15:07 <jlaska> anyway, we can talk about that further post-meeting
15:15:15 <kparal> we should link rather to final page than to proposal
15:15:26 <kparal> which we do not :) https://fedoraproject.org/wiki/Special:WhatLinksHere/Critical_Path_Packages_Proposal
15:15:34 <jlaska> yeah, that was my sense
15:16:02 <adamw> the thing is, the 'final' page is completely different
15:16:04 <adamw> and has nothing useful on it
15:16:11 <adamw> all the interesting stuff is on the 'proposal' page
15:16:18 <jlaska> right, it needs to mature quite a bit
15:16:22 <kparal> yes, it should be redone
15:16:45 <jlaska> I'll be happy to take a stab at cleaning that up ... unless anyone else is anxious to have it
15:17:11 <jlaska> #action Clean-up [[critical path packages]] wiki to be a suitable home for critpath information
15:17:16 <adamw> mostly the 'final' page just links to bits of the 'proposal' page :)
15:17:28 <jlaska> yeah, that was my quick'n'dirty page a while back
15:17:31 <jlaska> I can do better
15:17:46 <adamw> ahh, okay, i thought it was the releng team i was subtly slagging off there ;)
15:18:00 <jlaska> nope, it's cranes maska
15:18:03 <jlaska> ;)
15:18:12 <jlaska> alright, 2 more on the list ...
15:18:15 <jlaska> #info adamw to forward wwoods a good explanation of the nss-softokn problem scenario to ensure autoqa catches it in future
15:18:31 <jlaska> not sure if we still need to track that ... and wwoods is out today
15:19:04 <adamw> oh yikes, i forgot to transfer that to my todo list so i didn't do it
15:19:21 <adamw> queued up for this week
15:19:27 <jlaska> adamw: cool, can you send it to the autoqa-devel@ list too?
15:19:48 <jlaska> while wwoods is out, a group of us were going to try and test post-bodhi-update in his absence
15:19:58 <jlaska> I don't know if we'll get to depcheck, but who knows
15:20:12 <jlaska> #info jlaska to add autoqa FIXME links for a wiki page describing how to update repoinfo.conf when branched release is available, linked from the existing branched SOP pages
15:20:22 <adamw> okay
15:20:40 <jlaska> I created https://fedoraproject.org/wiki/How_to_update_AutoQA_repoinfo.conf to discuss how to update the autoqa repoinfo.conf file ... will send to autoqa-devel list
15:21:02 <jlaska> okay, anything else from last week that I missed?
15:21:57 <jlaska> alright, main agenda ...
15:22:03 <jlaska> #topic Fedora 13 QA Retrospective
15:22:33 <jlaska> I've compiled all the feedback in the retrospective and come up with a draft of recommendations
15:22:37 <jlaska> https://fedoraproject.org/wiki/Fedora_13_QA_Retrospective
15:22:52 <jlaska> I'm not ready to open to the list just yet, I'd like to clean it up a bit further
15:23:06 <jlaska> I plan to send this to the list for feedback by Wednesday of this week
15:23:07 <adamw> cool
15:23:52 <jlaska> so no action yet here, but I hope to have this out for consumption
15:24:07 <jlaska> and then we can go through what's next etc...
15:24:15 <jlaska> many of you are already working the next steps, which is good :)
15:24:28 <jlaska> and speaking of ...
15:24:35 <jlaska> #topic Proventester status
15:24:59 <adamw> we sorta covered this above, really?
15:25:02 <jlaska> adamw gave an update in the previous meeting follow-up, but I had this on as a standing item
15:25:19 <jlaska> adamw: right on
15:25:29 <jlaska> in terms of next steps ... I don't have any changes here
15:25:43 <adamw> me either, we're waiting on releng for the most important bit, and we're working on our process
15:25:48 <jlaska> #info review adamw's draft document guiding proventesters
15:26:11 <jlaska> #info waiting for re-enabling 'proventester' karma for critpath packages
15:26:28 <jlaska> I think that captures it, anything else?
15:26:37 <adamw> nup
15:26:58 <jlaska> adamw: kudos for tackling the QA/Join updates.  I was struggling to figure out how best to incorporate the link
15:27:42 <jlaska> alright, moving on ...
15:27:49 <jlaska> #topic AutoQA initscripts test validation
15:27:56 <jlaska> jskladan: I left this on as a topic from last week
15:28:07 <jskladan> jlaska: sure thing
15:28:14 <adamw> see my comment on the change - ideally i wanted to incorporate it into the existing mention of updates-testing, but couldn't think of a good way.
15:28:14 <jlaska> jskladan: were there any updates you wanted to give, or next steps to discuss?
15:28:35 <jlaska> adamw: yeah, that's what I was struggling with too!
15:28:47 <jskladan> so as you all probably know, I created a document <https://fedorahosted.org/autoqa/wiki/initscripts> on how to review, and sent an email to fedora test list
15:28:57 <adamw> i keep meaning to mention this in a blog post, didn't get around to it yet - sorry!
15:29:34 <jskladan> for now, we have one new guy - Scott M Fergusson, who's working on the task
15:29:44 <jlaska> yay Scott, thanks for helping!
15:30:02 <adamw> awesome
15:30:21 <jskladan> and two more contacted me via email today, so there is a great possibility of extending the possible AutoQA contributors even further :)
15:30:41 <jlaska> moohahahaha
15:30:56 <jskladan> so it seems like it had some effect after all :)
15:31:03 <jlaska> nice work jskladan!
15:31:22 <jskladan> and might be a great starting point for new contributors/people interested in joining QA forces :-D
15:31:24 <adamw> that's great
15:31:27 <adamw> yep for sure
15:31:32 <adamw> thanks for backing the project
15:32:13 <jlaska> jskladan: how do you want to handle this topic going forward?  Shall I take it off the agenda, or leave it on for weekly updates?
15:32:37 <jskladan> i think you can leave it in the agenda
15:32:51 <jlaska> jskladan: okay, will do
15:32:59 <jskladan> we all like to hear about others joining and contributing, don't we? :)
15:33:04 * adamw does
15:33:06 <jlaska> YES!
15:33:08 <jlaska> :D
15:33:31 * kparal votes for
15:33:56 <jlaska> I wager Viking-Ice would vote for it as well
15:34:10 <adamw> or shoot an arrow at it
15:34:24 <jlaska> that counts as a positive vote, right
15:34:26 <jskladan> and i'm still thinking about approaching Ray Chen - i'll probably write him some "hey, i saw your blog <http://blog.chinaunix.net/u2/68938/showart.php?id=2250919> come join us"
15:34:30 <jskladan> or smt like that
15:35:00 <adamw> that's excellent, it's always a good idea to take a personal approach like that
15:35:01 <jlaska> jskladan: so 3 new leads, cool
15:35:59 <jskladan> so thats about it on initscripts review effort, if noone has anything more, i think we can go on with the agenda
15:36:11 <jlaska> alright, thanks for the update
15:36:26 <jlaska> similar theme, but different focus ...
15:36:29 <jlaska> #topic AutoQA
15:37:04 <jlaska> so Kamil led us through a priority discussion last week (https://fedorahosted.org/pipermail/autoqa-devel/2010-June/000668.html) and helped get things focused
15:37:47 <kparal> tried to :)
15:38:00 <jlaska> kparal: are you able to guide us through current tasks?
15:38:16 <kparal> alright
15:38:26 <kparal> so we decided our current priorities are:
15:39:00 <kparal> * testing and finishing the Bodhi hook, because it is essential for many of our test cases
15:39:36 <kparal> * having quick support of running selected test cases in virtual machines, because we start to have some test cases which we don't really believe they are absolutely safe
15:40:07 <kparal> * getting testing instance of autotest-server and clients, so we can play with it and test new features rather than breaking our stable server
15:40:14 * Viking-Ice shoots the vote high up in the air in stead of deep in the ground..
15:40:33 <kparal> * getting publicly accessible machine for resultdb if possible
15:41:17 <kparal> * working on resultdb. We need to store the data and have some kind of simple form to explore it. that will allow larger audience to get familiar with our testing
15:41:47 <kparal> I think that's all. these tasks have the highest priority, everything else may wait a while
15:41:53 <jlaska> kparal: what happened to all the easy tasks! :)
15:42:24 <kparal> some of them are not so hard, but others consist of many subtasks, that's true :)
15:42:30 <jskladan> jlaska: you know the stuff, once you get to think about the simple tasks for a while, they become more complicated :-D
15:42:41 <jlaska> hehe, so true
15:42:42 <kparal> anyway, at least it helps us keep our focus
15:42:46 <adamw> i find not thinking about it helps
15:42:49 <jlaska> definitely, thanks kparal
15:43:40 <kparal> yrw
15:44:14 <jlaska> kparal: I should have an update on my action item for getting a new autotest-server shortly
15:44:33 <kparal> great
15:44:57 <jlaska> kparal: mmcgrath has been working hard to make available our 6 systems (and 1 virt guest (aka the autotest-server)).  It appears to be mostly working, I just need to figure out how we'll use this for the time being
15:44:57 * kparal looks forward and hopes he will not break it immediately :)
15:45:26 <jlaska> kparal: anything else you'd like to track on this topic for next week?
15:45:41 <kparal> I don't think so
15:46:02 <jlaska> alright, thanks for the update kparal
15:46:06 <jlaska> who wants open floor time?
15:46:43 * adamw is good
15:46:44 <adamw> oh
15:46:46 <jlaska> #topic Open floor - <your topic here>
15:47:03 <adamw> i should just briefly mention i'm working on expanding desktop testing for f14
15:47:08 <adamw> to cover kde, xfce and lxde as well as gnome
15:47:18 <jlaska> #topic Open floor - adamw and F14 desktop testing
15:47:19 <adamw> in the early planning stages on that one, something should hit the mailing lists soon
15:47:48 <jlaska> #info i'm working on expanding desktop testing for f14 to cover kde, xfce and lxde as well as gnome
15:48:00 <jlaska> #info something should hit the mailing lists soon
15:48:15 * jlaska gets friendly with meetbot
15:48:58 <jlaska> adamw: nice, I gather that will be a welcome improvement
15:49:07 <adamw> i hope so
15:49:24 <jlaska> #topic Open floor - <your topic here>
15:49:48 <jlaska> any other thoughts/concerns/haiku?
15:50:16 <jlaska> If no other topics, I'll close the meeting in 2 minutes
15:51:15 * adamw is short on poetic inspiration
15:51:41 <jlaska> #topic Open floor - AutoQA media kit sanity tests
15:51:59 <kparal> do we have robatino?
15:51:59 <jlaska> robatino asked about this, so it might be worth noting in case others are interested
15:52:23 <robatino> here
15:52:29 <jlaska> robatino: howdy
15:52:40 <jlaska> Liam has code in his private AutOQA branch to automate the mediakit sanity tests
15:53:01 <jlaska> The tests he is automating can be found at https://fedoraproject.org/wiki/Category:Installer_Image_Sanity_Test_Cases
15:53:07 <robatino> i was wondering if a test should be added to compare the actual ISO size to the ISO header size
15:53:30 <adamw> oh, yeah, you found problems with that didn't you?
15:53:42 <robatino> the live and netinst images currently fail that
15:53:47 <jlaska> robatino: I don't know, to be honest.
15:53:56 <adamw> what's the impact of the problem again?
15:53:59 <robatino> not something to block, but just to track
15:54:15 <jlaska> robatino: sure ... if you have any script or code to help validate, please add to https://fedorahosted.org/autoqa/ticket/186
15:54:35 <robatino> besides being inconvenient, we don't really know whether the images are just oversized, or corrupted as well
15:55:04 <jlaska> robatino: did this first start with F-13 ISOs?
15:55:15 <robatino> F12 Alpha, I think
15:55:31 <jlaska> I think you have a bug tracking this already, right?
15:55:34 <robatino> https://bugzilla.redhat.com/show_bug.cgi?id=585006
15:55:55 <robatino> it's against livecd-tools because currently that's the only way i know to reproduce it
15:56:06 <robatino> it's probably a genisoimage bug though
15:56:41 <jlaska> huff: you own livecd-tools bugs now?
15:57:01 <robatino> someone good at python could probably throw in some test statements and track it to the source, then reassign to the correct component
15:57:35 <pjones> robatino: are the images larger or smaller than the size listed in the headers?
15:57:46 <robatino> larger
15:58:08 <pjones> that's potentially valid
15:58:14 <pjones> (though still an error)
15:58:33 <robatino> of course, without knowing the cause, it
15:58:38 <jlaska> pjones: do you know what writes out that ISO header?  is that genisoimage as robatino suggests?
15:58:45 <robatino> it's possible they could be smaller as well
15:58:45 <pjones> that's genisoimage, yes
15:58:49 <jlaska> okay
15:59:31 <jlaska> well, sounds like more work is needed to pinpoint this
15:59:50 <adamw> seems like we have enough time we should be able to pinpoint it, though
15:59:55 <jlaska> right
16:00:35 <jlaska> not sure what else is needed, I think this just needs a subject matter expert to review
16:01:07 <adamw> this actually reminds me of something else, if we have a minute
16:01:16 <jlaska> adamw: new topic?
16:01:22 <adamw> yeah
16:01:28 <jlaska> to close out this ...
16:01:43 <jlaska> robatino: can you suggest the new test to Liam in the ticket
16:02:05 <jlaska> and we'll continue monitoring this bug, perhaps Oxf13 or pjones have ideas to further isolate
16:02:17 <jlaska> adamw: go for it
16:02:18 <robatino> yes, i was planning to mention that, and also to make sure to distinguish between SI and IEC units for different media
16:02:24 <Oxf13> I'm here.
16:02:26 <jlaska> robatino: cool, thanks
16:02:26 <Oxf13> what's the bug?
16:02:32 <jlaska> 585006
16:02:34 <jlaska> .bug 585006
16:02:35 <zodbot> jlaska: Bug 585006 livecd-creator creates i386 and x86_64 ISOs which are larger than indicated by the ISO header - https://bugzilla.redhat.com/show_bug.cgi?id=585006
16:02:50 <Oxf13> ah that.
16:03:05 <Oxf13> We were thinking about migrating over to libisofs to see if that changes anything, or if it's easier to work with
16:03:11 <Oxf13> shell execution within python is ugly
16:04:08 <jlaska> Oxf13: anything you can think of to help robatino further isolate the bug?
16:04:30 <Oxf13> not really.
16:04:58 <jlaska> is this libisofs change, that's with livecd-tools or something else?
16:05:45 <jlaska> alright, let's move on ... we can talk further on this in the bug or in #fedora-qa
16:05:48 <jlaska> adamw: take it away
16:05:55 <adamw> okay
16:06:09 <jlaska> #topic Open floor - adamw
16:06:16 <adamw> so there's been a few forum threads where people have said they couldn't get a DVD burned from the f13 x86-64 ISO to boot, at all
16:06:33 <adamw> their systems actually don't see it as a bootable disc
16:06:44 <adamw> in each case, they could boot from the i686 DVD, or the x86-64 live
16:07:08 <adamw> threads - http://forums.fedoraforum.org/showthread.php?t=245321 , http://forums.fedoraforum.org/showthread.php?p=1364323 , http://forums.fedoraforum.org/showthread.php?t=245844
16:07:20 <adamw> but obviously in many cases the dvd does boot fine
16:07:34 <Oxf13> jlaska: mostly pjones and I were looking at it for anaconda buildinstall and pungi, but lessons learned there should carry over to livecd-tools
16:07:39 <adamw> i've no idea what could be causing this, the whole process by which a system identifies and boots from a bootable device is voodoo to me
16:07:45 <jlaska> Oxf13: okay
16:08:08 <adamw> just thought i'd mention it in case anyone knows more than me, or knows who could take a stab at figuring it out
16:08:27 <Southern_Gentlem> adamw,  and to majority of those that came into #fedora hadnt verified the iso and had bad downloads or where burning incorrectly
16:08:50 <Oxf13> adamw: x86_64 doesn't boot, but i386 does... that tickles something in my brain
16:08:56 <adamw> there's also a bug report for this - https://bugzilla.redhat.com/show_bug.cgi?id=597283
16:09:00 <Oxf13> I wonder if we're putting EFI info on x86_64 media, but not i386 media.
16:09:06 <adamw> which oxf13 has commented on, so that may be the tickle point =)
16:09:12 <Oxf13> oh...
16:09:23 <adamw> would the efi info also be on x86-64 live in that case?
16:09:30 <adamw> the reporters say x86-64 live boots, iirc
16:09:37 <Oxf13> adamw: no, efi wouldn't be there.
16:10:08 <Oxf13> since it's DVD media, and that's my territory, I don't generally do anything with EFI so I don't think that's me
16:10:10 <Oxf13> or it
16:10:48 * adamw didn't quite parse that last line
16:11:54 <jlaska> comparing i386 and x86_64 "efi" files on the DVD's -- http://fpaste.org/fBu6/
16:12:00 <jlaska> i386 has a #
16:12:05 <jlaska> /EFI/BOOT/BOOT.conf file, but x86_64 does not
16:12:33 <jlaska> anyhow, should we take the details to #fedora-qa
16:12:58 <adamw> or the bug report, sure
16:13:01 <jlaska> right on
16:13:14 <jlaska> adamw thanks for raising this topic ... has me perplexed :)
16:13:16 <Oxf13> adamw: pungi is what creates the DVD isos, and it's during that creation one would have to make them efi bootable or not IIRC.  And since I don't do that in pungi, I don't think it could be EFI that is causing issues.
16:13:28 <adamw> Oxf13: ah, okay.
16:13:46 <jlaska> for the logs ... this is
16:13:48 <jlaska> .bug 597283
16:13:49 <zodbot> jlaska: Bug 597283 Fedora 13 x86_64 DVD does not boot in some computers - https://bugzilla.redhat.com/show_bug.cgi?id=597283
16:13:56 <jlaska> okay folks ... thanks for your time
16:14:02 <adamw> Southern_Gentlem: in the cases i'm looking at, the reporters have verified the burns and in some cases have actually booted the same disc successfully on other machines
16:14:19 <adamw> Southern_Gentlem: i'm painfully familiar with bad burns, this doesn't look like that =)
16:14:38 <jlaska> If folks want to continue, feel free to discuss on #fedora-qa or the mailing list
16:14:42 <jlaska> thanks all!
16:14:57 <jlaska> #endmeeting