QA/Meetings/20091102

From FedoraProject

Jump to: navigation, search

Contents

Attendees

People present (lines said)

  • adamw (72)
  • wwoods (32)
  • jlaska (15)
  • kparal (7)
  • Oxf13 (7)
  • tk009 (5)
  • zodbot (2)
  • buggbot (1)
  • thomasj (1)
  • Viking-Ice (1)
  • drago01 (1)
  • mchua (1)
  • seelchen (1)

Regrets:

Agenda

Previous meeting follow-up

See https://fedoraproject.org/wiki/QA/Meetings/20091026#Action_items

  • adamw will ask Milos to follow-up at next week's meeting or on the list on the FTBFS topic

Adam was in touch with Milos and asked for a follow-up on the FTBFS idea. So far no updates have been made. Milos asked when the next QA meeting would be, but doesn't seem to be available today.

  • Viking-Ice investigating creating a test case for bug#530452 ... and adding to F12 install matrix

No updates

  • wwoods to add roadmap/tickets for reworking shared autoqa code (watchers, test helper methods, etc) into a proper library

In progress. Will is working on a private git branch

Review beta test

James put this on to get a feel for how everyone thinks f12 beta testing is coming along with a view to whether we reckon we'll be ready to go RC on schedule. Adam asked the group for any particular thoughts/concerns.

TK009 asked how much time was left. Wednesday is the cutoff date (kparal directed the group towards the schedule). Adamw noted that ideally after Wednesday all the 'code bugs' should be fixed and further builds should only be addressing issues in the iso builds themselves.

Wwoods noted that after last Friday's EPIC blocker bug review (see recap), he felt there was a good handle on the scope of remaining issues.

With regards to blocker bugs, wwoods provided some data:

25 open bugs, none NEW, 12 ASSIGNED, 13 MODIFIED (see F12Blocker). Adam pointed out that 7, or 8, of the OPEN issues are still waiting for developer action. This includes some easy issues where the bug is waiting for tagging before being moved to MODIFIED. The group didn't think the remaining issues were cause for alarm. Wwoods and Adamw debated whether a community bug verification event was needed to help push through remaining MODIFIED bugs. Adam noted there were clear action items from the blocker review meeting. However, Adam reminded everyone that most of the bugs in MODIFIED need some sort of testing, so please take a minute to look through the MODIFIED items and provide any feedback you can, if you have time.

AutoQA update

Wwoods updated the team on his work to rebuild the autoqa code to provide a nice python library. For details see http://git.fedorahosted.org/git/?p=autoqa.git;a=tree;f=lib/python;hb=wwoods-autotest. The hope is to have this code merged into master by the end of the week. Packaging and testing of a newer version of autotest-0.11 is underway as well (see ticket#9).

Kparal indicated there haven't been new developments with rpmguard this past week. Several new tickets arrived to track enhancements recommended on his blog. The main focus is integrating rpmquard into autoqa.

Open discussion - <Your topic here>

Some general discussion around the upcoming release candidate and the friday blocker bug day. Jlaska took an action item to catch up with denise and the installer team for any issues they were tracking for inclusion into the RC.

Upcoming QA events

  • 2009-11-02 - End of beta testing
  • 2009-11-04 - Compose release Candidate
  • 2009-11-04 - Test release candidate
  • 2009-11-06 - Blocker bug day

Action items

  • (carry over) - Viking-Ice investigating creating a test case for bug#530452 ... and adding to F12 install matrix
  • jlaska to check-in with anaconda-devel team for status on any issues they are monitoring for the wednesday RC compose


IRC Transcript

adamw #startmeeting 2009-11-02 QA meeting 16:01
zodbot Meeting started Mon Nov 2 16:01:47 2009 UTC. The chair is adamw. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:01
zodbot Useful Commands: #action #agreed #halp #info #idea #link #topic. 16:01
adamw #topic taking attendance 16:01
adamw who's here? 16:02
* tk009 16:02
* kparal first :) 16:02
kparal damn 16:02
tk009 not =P 16:02
* jlaska lurking 16:02
* wwoods hereish 16:02
* seelchen here 16:03
jlaska I'm double booked for meetings right now, so I'll be slower than usual to respond 16:03
adamw i shall be running the meeting due to our Glorious Leader's inability to manage his schedule ;) 16:03
adamw #topic previous meeting follow-up 16:04
jlaska adamw: bitten by DST :( 16:04
* mchua lurking 16:04
adamw hi mel 16:04
adamw let's see - I did indeed ask Milos to follow-up on his FTBFS idea, but he obviously hasn't yet. he did ask when the next meeting was going to be, but I don't see him here. 16:05
adamw "Viking-Ice investigating creating a test case for bug#530452 ... and adding to F12 install matrix" 16:05
buggbot Bug https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=530452 high, low, ---, jmccann, CLOSED RAWHIDE, Gnome sets the keyboard layout to USA after every log in 16:05
adamw Viking-Ice: ping, any news? 16:06
adamw I guess he's not here 16:07
adamw wwoods to add roadmap/tickets for reworking shared autoqa code (watchers, test helper methods, etc) into a proper library - anything on that, will? 16:07
wwoods yeah! 16:08
wwoods I'm working in a git branch in autoqa 16:08
wwoods all of the repo info used by the watchers has been moved to a shared library and a config file 16:08
wwoods so we can update repo info without needing to replace the watcher code 16:08
adamw excellent 16:08
adamw let's save more detail for the autoqa topic 16:09
wwoods indeed 16:09
adamw so, onto the next... 16:09
adamw #topic Review beta test 16:09
wwoods but It's Coming Along Nicely. 16:09
adamw so, James put this on to get a feel for how everyone thinks f12 beta testing is coming along 16:09
adamw with a view to whether we reckon we'll be ready to go RC on schedule 16:10
adamw does anyone have any particular thoughts? 16:10
wwoods how much time is left? 16:10
* Viking-Ice jumps in late 16:10
tk009 2 days? 16:10
adamw i believe wednesday is basically the cutoff 16:10
kparal http://poelstra.fedorapeople.org/schedules/f-12/f-12-key-tasks.html 16:11
wwoods after friday's Epic Blocker Review I feel like we have a pretty good handle on the scope of the remaining issues 16:11
kparal 2 days 16:11
adamw at that point all the 'code bugs' should be fixed and further builds should only be addressing issues in the iso builds themselves 16:11
wwoods eep. 16:11
adamw wwoods: yeah, that was my feeling. i'm hoping to knock off several of those today... 16:11
* tk009 didn't look out of fear 16:12
adamw as far as the blocker bugs go, the list is at https://bugzilla.redhat.com/showdependencytree.cgi?id=473303&hide_resolved=1 for reference 16:12
adamw it still looks quite long but as I said I hope to close rather a few of those today...hopefully 16:12
wwoods AFAICT: 25 open bugs, none NEW, 12 ASSIGNED, 13 MODIFIED 16:13
wwoods in theory all MODIFIED bugs should be closeable as soon as we confirm the fix 16:13
adamw some of the ASSIGNED are somewhat misleading 16:14
adamw i haven't got an exact count but I think off the top of my head there's only 7 or 8 where we're really still waiting on developer action 16:14
adamw and some of them are pretty easy stuff, like the bug that's just davidz reminding himself to do a tag request for an updated DeviceKit-disks build 16:14
wwoods gotcha 16:15
adamw overall i'm fairly hopeful we'll actually be able to more or less knock everything off in time 16:15
wwoods later today can we run through the list and close/move bugs to MODIFIED where appropriate, to make this less scary? 16:15
wwoods I don't feel like there's any hair-on-fire-panic issues but that list is worry-making 16:16
adamw i'll be doing that myself anyway, but if you want to make it an event we could 16:16
adamw obviously will just didn't get enough blocker bug review meeting action on friday =) 16:16
wwoods nah, just wanted to make sure someone was directly responsible for managing the list 16:17
* wwoods hungers for it 16:17
wwoods just don't have myself CC'd on enough bugs anymore 16:17
adamw there's clear action items here, of course, btw - most of the bugs in MODIFIED need some sort of testing, so please everyone do take a minute to look through the MODIFIED items and provide any feedback you can, if you have time 16:17
adamw so that's the blocker bug perspective...anyone have perspective from any other angles? i haven't been doing / following much install testing for e.g. 16:17
adamw i guess not! 16:20
drago01 fwif did two installs yesterday one from dvd one livecd both where fine 16:21
adamw well, my feeling is we may well be able to make it for wednesday but it's hard to say for sure at this point. i'm reasonably optimistic. 16:21
adamw thanks, drago. 16:21
adamw ok, onto the next one... 16:21
adamw #topic AutoQA update 16:22
adamw step forward our brave autoqa warriors! 16:22
wwoods bam! 16:22
wwoods so yeah: reworking autoqa code to provide a nice python library 16:22
wwoods #link http://git.fedorahosted.org/git/?p=autoqa.git;a=tree;f=lib/python;hb=wwoods-autotest 16:23
wwoods oh wait. #link is unnecessary. heh. 16:23
wwoods anyway, other shared (or shareable) code from the various tests is being moved in there 16:24
wwoods so e.g. the virtguest code that rats_install uses is now available for any other tests to use 16:24
wwoods there's a koji_utils module that can be used to get the NVR of previous released versions of a given package 16:25
wwoods etc. 16:25
wwoods I hope to have this merged into master by the end of the week 16:25
wwoods and the existing tests modified to use the shared code 16:25
wwoods and to start actually wiring up the post-koji-build watcher and get rpmguard running 16:25
kparal which is also my task :) 16:26
wwoods one other note 16:26
wwoods we're packaging and testing a slightly newer autotest version 16:26
wwoods so there may be some weird emails and stuff coming from that as we set up the new autotest version and work on the new hook and adapt the tests 16:26
adamw thanks for the warning 16:27
adamw anything of interest on the rpmguard front, kparal? 16:27
kparal unfortunatelly there has not been any development in rpmguard front in the past week. I have only created a few tickets with possible enhancements. 16:28
kparal the main task now will be autoqa integration. when that works, I can work more on improving the code 16:29
adamw alright, thanks 16:29
adamw so onto... 16:29
adamw #topic upcoming QA events 16:29
adamw see the list in the agenda: https://www.redhat.com/archives/fedora-test-list/2009-November/msg00043.html 16:30
adamw end of beta testing is listed as today, which is clearly a sick joke ;) 16:30
adamw but yeah, RC composing action begins wednesday. we have a blocker bug day listed for friday, which seems slightly odd to me as - by my understanding - we can't fix any more blocker bugs after wednesday. 16:31
adamw without slipping. 16:31
adamw Oxf13: any clarifications? 16:31
Oxf13 adamw: we can fix bugs that show up from the RC 16:31
Oxf13 up until Monday without slipping 16:31
Oxf13 the following Monday is the go/nogo date, where it's either fixed or we're slipping 16:32
adamw ah, ok 16:32
adamw but we want to fix as much as possible for the RC start date anyway 16:32
Oxf13 we have to have the blocker list clear before we can enter RC 16:32
Oxf13 so having an RC on Wed is not a forgone conclusion 16:32
adamw alright 16:33
adamw thanks for the clarification 16:33
adamw so yeah, not much new here: now until wednesday is full-steam-ahead blocker bug bashing mode 16:33
adamw if you're not working on blocker bugs you'd better have a note from your doctor explaining that you have a serious case of 'dead' or you'll have me to answer to ;) 16:34
adamw anything you wanted to mention about the upcoming week, oxf13? 16:34
* jlaska no longer distracted 16:34
adamw jlaska: feel free to add anything i'm missing 16:35
* thomasj late lurking 16:35
jlaska Liam is planning to kick off a RC install test run on wednesday 16:35
Oxf13 We have seriously tiny amount of time between the RC compose and the go/nogo vote 16:35
jlaska ideally this would be against the RC ... but Oxf13 mentioned if we haven't reached closure on the F12Blocker list, he would compose a "Test Compose" for us to track updated install results 16:36
Oxf13 so we need to be hitting rawhide with a vengeance leading up to the RC, and also hitting RC with a vengeance to make sure it's up to snuff before we vote on it. 16:36
adamw right 16:36
jlaska denise: are there any bugs in progress on the anaconda side that would be good to have land in the wednesday compose 16:36
jlaska whether it's RC or 'test compose' ? 16:37
jlaska we can work that outside the meeting, but just something to coordinate for that compose milestone 16:37
adamw all the anaconda bugs are currently in MODIFIED state according to the blocker list 16:37
adamw so the onus there seems to be on us 16:37
jlaska sort of 16:37
adamw (pun not intended) 16:37
jlaska there are a few inprogress, before we do the compose, I think it would be helpful to check-in with her team 16:38
jlaska to avoid spinning another compose the next day with a new anaconda 16:38
adamw all the anaconda bugs currently marked as blocker bugs are intended to be fixed in anaconda 12.42, which was tagged today 16:38
adamw so we definitely need all of those re-tested with today's rawhide 16:39
jlaska #action jlaska to check-in with anaconda-devel team for status on any issues they are monitoring for the wednesday RC compose 16:39
jlaska that's it from me 16:41
adamw alright 16:41
adamw anything we didn't cover on the topic of upcoming events? 16:41
adamw there's no test day this week 16:41
adamw in that case, moving on to... 16:43
adamw #topic open floor 16:43
adamw anything at all anyone wants to share? come on, don't be shy! 16:43
jlaska nothing here 16:43
adamw alright, banging gavel in 3... 16:45
adamw 2... 16:45
adamw 1... 16:45
adamw thanks for coming, everyone 16:45
tk009 =) 16:45
adamw #endmeeting 16:45

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!