QA/Meetings/20110509

From FedoraProject

Jump to: navigation, search

Contents

Attendees

People present (lines said):

  1. jlaska (162)
  2. adamw (41)
  3. kparal (27)
  4. tflink (19)
  5. rbergeron (12)
  6. robatino (5)
  7. nirik (3)
  8. StylusEater_work (2)
  9. vhumpa (2)
  10. jskladan (1)
  11. fenrus02 (1)
  12. brunowolff (1)
  13. the_hydra (1)
  14. Viking-Ice (1)
  15. jsmith (1)
  16. michich (1)

Unable to attend:

  1. Rhe (hopefully sleeping)
  2. Hongqing (hopefully sleeping)
  3. Twu (hopefully sleeping)

Agenda

Previous meeting follow-up

  1. tflink to follow-up with cloud sig for test day recap

F-15-Final-RC1 status

  • STATUS: ETA - Wednesday, May 11 -- see rel-eng#4697
  • Continue to monitor for blocker bugs at https://fedoraproject.org/wiki/Current_Release_Blockers
  • Robatino and rhe prepared to send RC1 ISO announcements to test-announce@.
  • Robatino already created Final RC1 QA wiki pages at https://fedoraproject.org/wiki/Category:Fedora_15_Final_RC_Test_Results
  • The team conducted a *mini* review of 2 accepted and 1 proposed blocker bugs ...
    1. RHBZ #697834 - gnome-menus - Other menu appears in default installation (any .desktop entry with Category=Settings ends up here (even if there's also Category=System))
      • Unclear what this needs to move forward, jlaska will reach out to mclasen for guidance
      • AdamW researching and testing a patch to resolve the issue
    2. RHBZ #693809 - imsettings - Error message about missing input methods should be removed
      • Waiting to hear back from tagoh to respond to caillon feedback.
      • jlaska will reach out to caillon to confirm that recent comments are still inline with original bug report
    3. RHBZ #702650 - plymouth - Concurrency problem when unlocking partition
      • kparal retesting and plans to reach out to test@ for feedback
      • halfline is on PTO, lennart said he would provide feedback in his absence

AutoQA update

  • AutoQA trac milestones have been changed to reflect the new topic for 0.5.0 release - making AutoQA results more user-friendly (for package-maintainers). https://fedorahosted.org/autoqa/milestone/0.5.0
  • We held a meeting where we discussed different approaches related to 0.5.0 milestone. tflink was kind enough to create a record of it: https://fedorahosted.org/autoqa/wiki/Planning050series
  • We agreed to substantially lower the number of emails that are currently sent to package maintainers through Bodhi comments. We will also improve the structure and contents of the test log - it will be specific to a particular update, it won't contain many other updates any more. Every test will be documented on Fedora wiki and the most information will be provided there.
  • We tackled a problem with too low disk space on AutoQA server that resulted in massive deletion of very recent logs. And we will continue to work on it. The problem was caused by some stuck-in-a-loop tests that generated many GBs of test logs.
  • jlaska reported that the quick changes made on Friday dramatically reduced the size of depcheck test logs on the autoqa server. Special thanks to smooge for providing additional disk space for our guest

Upcoming QA events

  1. Wednesday, May 11 - Final Candidate testing starts
  2. Tuesday, May 17 - Go/NoGo meeting
  3. Tuesday, May 24 - Final (GA) release

Open Discussion - <your topic here>

Review proposed NTH bugs

  1. https://bugzilla.redhat.com/show_bug.cgi?id=674321 - exo shouldn't take x-scheme-handler/file mimetype
    • AGREED: 674321 - AcceptedNTH - impacts desktop criteria for non-default desktop
  2. https://bugzilla.redhat.com/show_bug.cgi?id=694900 - Missing icon in xfce4-panel 'Add new items' dialog
    • AGREED: 694900 - AcceptedNTH - obvious polish issue for non-default desktop
  3. For bugs related to F14->F15 upgrades where configured services are no longer starting under systemd - jlaska will file bugs and will raise/discuss after meeting.

Action items

  1. jlaska to follow-up w/ halfline on bug#702650
  2. kparal + jlaska to continue testing bug#702650
  3. adamw proposing patch for bug#697834
  4. jlaska - file proposed NTH bugs for F14->F15

IRC Log

jlaska #startmeeting Fedora QA Meeting 15:00
zodbot Meeting started Mon May 9 15:00:47 2011 UTC. The chair is jlaska. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00
zodbot Useful Commands: #action #agreed #halp #info #idea #link #topic. 15:00
jlaska #meetingname fedora-qa-meeting 15:00
zodbot The meeting name has been set to 'fedora-qa-meeting' 15:00
jlaska #topic Roll Call ... 15:00
* tflink is here 15:01
jlaska Hi folks ... let's get this party started 15:01
* StylusEater_work is here 15:01
jlaska wave your electronic hands so we know who is here 15:01
adamw yo 15:01
jlaska tflink, StylusEater_work, adamw: greetings 15:01
* the_hydra is newcomer...willing to learn 15:01
jlaska regrets for brunowolff ... we'll catch up with him later 15:01
* kparal is here 15:01
* fenrus02 waves bacon around to share 15:02
jlaska welcome the_hydra and kparal :) 15:02
* vhumpa say hey there 15:02
jlaska greetings vhumpa 15:02
rbergeron woop woop 15:02
vhumpa ^^^ says :-) 15:02
* jskladan lurks 15:02
rbergeron wave your hands in the aiiiiiir and grab some baaaaaconnnnnn 15:02
jlaska jskladan: rbergeron: hi folks 15:03
jlaska oh, bacon, where? 15:03
* StylusEater_work snags some bacon 15:03
adamw then what do we do with the bacon? 15:03
jlaska robatino: Viking-Ice? who else are we missing? 15:03
* robatino here 15:03
jlaska alright ... I think we are "critically massy" now 15:03
rbergeron adamw: insert bacon in mouth 15:03
rbergeron then enjoy 15:03
jlaska rbergeron: can you create a wiki test case for that please? 15:04
jlaska  : 15:04
jlaska  :) 15:04
jlaska #topic Previous meeting follow-up 15:04
jlaska I had only 1 item on my list from last week 15:04
jlaska #info tflink to follow-up with cloud sig for test day recap 15:04
jlaska I think this is stalled because a lot of folks were out @ summit last week? 15:04
jlaska tflink: any news/updates ... anything I can help with on this one? 15:05
tflink I heard back from mgoldmann and while he was busy @ summit/JBW last week, he's planning to send out a recap this week 15:05
jlaska eggsellent, thanks for the update. Let's keep this on the list for next week 15:06
rbergeron i think he literally just got back yesterday 15:06
jlaska and his arms are probably tired! 15:06
jlaska hey-yoooo! 15:06
jlaska you see, after he just flew in from. ... oh nm 15:06
jlaska anything else from last week that I failed to capture in the minutes? 15:07
adamw i'm still recovering from that last joke 15:07
jlaska haha, be sure to tip your waiter 15:07
jlaska alright ... let's dive into the main course ... 15:07
jlaska #topic F-15-Final-RC1 status 15:08
jlaska So, really this should be F-15-Final-TC1 status, but since the RC1 is slated for composition for rel-eng starting today ... I changed it 15:08
jlaska We aren't expecting delivery of F-15-Final-RC1 until Wednesday 15:08
jlaska this gives rel-eng time to work through any late-breaking compose issues 15:08
jlaska #link https://fedorahosted.org/rel-eng/ticket/4697 15:09
jlaska I just added a slew up updates to that ticket that I believe we need rel-eng to include in RC1 15:09
adamw we also have open blockers, right> 15:09
jlaska you got it! 15:09
jlaska if you don't mind, we'll walk through those in a moment (hopefully quickly) 15:09
jlaska #info Continue to monitor for blocker bugs at https://fedoraproject.org/wiki/Current_Release_Blockers 15:09
jlaska looks like robatino continues to rock ... 15:09
jlaska #info Robatino prepared wiki pages for RC1 test tracking (see https://fedoraproject.org/wiki/Category:Fedora_15_Final_RC_Test_Results) 15:10
jlaska I guess depending on when RC1 lands, either robatino or rhe will handle blasting the test announcement? 15:10
robatino i can assuming i'm around 15:11
jlaska adamw: tflink: rbergeron: btw ... thanks for processing the bugs during the blocker review last Friday ... looks like some good progress was made 15:11
* rbergeron points at everyone else and takes absolutely no credit 15:11
jlaska robatino: thank you ... and I'm sure rhe will happily grab that if it lands during her daytime 15:11
jlaska alright ... who likes bugs? 15:12
jlaska good, good ... now who likes quickly reviewing bugs? 15:12
* jlaska raises hand 15:12
* tflink likes bug reviews to be done :) 15:12
adamw still tired...see...because of his arms... 15:12
tflink not so much on the doing part 15:12
jlaska heck yeah! Can you believe that's almost here :) 15:12
jlaska adamw: LOL! 15:13
jlaska adamw: get it, get it? 15:13
* adamw drools from one side of his mouth 15:13
jlaska alright ... so there are 2 AcceptedBlocker bugs that are still OPEN 15:13
* rbergeron adds a piece of bacon to block the drool 15:13
jlaska I'll take gnome-menus for $1000 please ... 15:13
jlaska #info bug#697834 - gnome-menus - Other menu appears in default installation (any .desktop entry with Category=Settings ends up here (even if there's also Category=System)) 15:13
adamw ...sigh. 15:14
jlaska so it seems that mclasen is proposing that we drop the 'Other' menu criteria for F15? 15:14
jlaska or is there something else preventing this from moving forward? 15:14
rbergeron this isn't a case of "and later we'll add the "other" menu criteria back in" is it? 15:15
jlaska For those not familiar with the blocker review format ... we are checking in on any unresolved AcceptedBlocker bugs and deciding what is needed to move them forward 15:15
adamw mclas: yes, but he's very short on reasons 15:15
adamw grr 15:16
adamw jlaska: he's very short on reasons 15:16
adamw it bugs me that they are seemingly putting their foot down on fixing this for no good reason 15:16
tflink yeah, I don't really understand the hangup either 15:16
adamw it's literally about a three-line fix in gnome-menus 15:16
adamw i guess i'll post a patch and see if that helps 15:16
jlaska ooh good approach 15:16
jlaska I was going to suggest we sync up w/ mclasen post meeting in #fedora-desktop to see if we can agree on something 15:16
jlaska in fact, let's do both 15:17
jlaska thanks adamw 15:17
jlaska anything else on this one? 15:17
* jlaska thinks that should help get resolution *today* on this 15:17
jlaska okay ... next up ... 15:18
jlaska #info bug#693809 - imsettings - Error message about missing input methods should be removed 15:18
jlaska I've reached out to tagoh on this one, but I don't expect a response until he's awake 15:18
jlaska I'd like to catch up w/ caillon too to see if there's anything we can do to help this along before tagoh is online 15:18
* jlaska open to any other suggestions 15:19
jlaska if there are none, I'll tell more jokes 15:19
tflink I don't think that there is much else that we can do 15:19
tflink at the moment, anyways 15:19
jlaska another thought, does the new stuff proposed by caillon change the intent of this bug 15:20
jlaska in other words, is this still the same issue that was AcceptedBlocker? 15:20
jlaska I'll follow-up with caillon with this q ... "Is the original reported issue resolved, but there are additional improvements that could be made?" 15:20
* jlaska prepares to move on ... 15:21
adamw they seem like significant issues to me 15:21
jlaska okay 15:21
adamw but they're different from the initial report 15:21
jlaska I'm willing to bet caillon understands the time crunch, and has a keen sense for what is required, vs what can be resolved later 15:22
jlaska I'll update the bz if anything surfaces that requires re-thinking the issue 15:22
jlaska adamw: seem sensible? 15:22
jlaska #info bug#702650 - plymouth - Concurrency problem when unlocking partition 15:23
jlaska Looks like kparal provided the requested plymouth.debug logs ... I think we just need input from halfline to proceed 15:23
kparal jlaska: I did, but not reproducing the bug 15:23
kparal jlaska: it just didn't manifested with plymouth.debug on 15:23
jlaska Oh, so everytime you boot with plymouth debugging, the bug doesn't reproduce? 15:24
kparal and, I don't know how to disable plymouth. no documentation whatsoever 15:24
michich plymouth.enable=0 15:24
kparal jlaska: yes 15:24
kparal michich: thanks. unfortunately it seems to be written nowhere 15:24
* kparal will try that asap 15:24
jlaska michich: Huh, I don't see that as a supported boot option 15:25
kparal jlaska: I'll try a few more rounds after the meeting 15:25
jlaska kparal: okay, thanks 15:25
jlaska This issue is still only proposed, not accepted 15:25
adamw see comment #6 15:25
jlaska yeah, just read 15:26
jlaska Worth a call to test@ for feedback from people with non-root encrypted partitions? 15:26
jlaska I can queue up a few installs post-meeting to test 15:26
tflink probably 15:26
kparal I think it should be pretty easy to reproduce. I did several installations 15:26
kparal I hit the bug in at least 50% of system boots on those installations 15:27
jlaska fascinating 15:27
tflink kparal: did that increase? your original report said 33% 15:27
jlaska and concerning we haven't seen this before 15:27
* jsmith pays closer attention 15:27
kparal tflink: yeah, I underestimated a little, not to be an alarmist :-) I also did more runs after that 15:27
jlaska anything else we can do with this issue in _this_ meeting? 15:28
jlaska sounds like we're going to continue retesting (kparal and jlaska) 15:28
adamw kparal: it'd be good to add that to the report 15:28
kparal yep 15:28
adamw (info that you hit it in several re-installs) 15:28
kparal adamw: ok 15:28
jlaska and I'll see if I can get help from halfline post-meeting 15:28
jlaska okay ... I think we are done on this topic 15:29
jlaska #action jlaska to follow-up w/ halfline on bug#702650 15:30
jlaska #action kparal + jlaska to continue testing bug#702650 15:30
jlaska #action adamw proposing patch for bug#697834 15:30
jlaska #topic AutoQA update 15:31
jlaska Okay, quite a busy week for AutoQA ... I'm still behind on autoqa-devel@ mails 15:31
tflink we aim to keep you on your toes :) 15:31
jlaska Any exciting news to share? 15:31
jlaska haha, </success> :D 15:31
* kparal is the man for sharing updates 15:31
kparal but just shortly, if you don't mind 15:31
* kparal likes short updates 15:32
kparal #info AutoQA trac milestones have been changed to reflect the new topic for 0.5.0 release - making AutoQA results more user-friendly (for package-maintainers). 15:32
kparal #link https://fedorahosted.org/autoqa/milestone/0.5.0 15:32
kparal #info We have held a meeting where we discussed different approaches related to 0.5.0 milestone. tflink was kind enough to create a record of it: 15:32
kparal #link https://fedorahosted.org/autoqa/wiki/Planning050series 15:32
kparal #info We have agreed to substantially lower the number of emails that are currently sent to package maintainers through Bodhi comments. We will also improve the structure and contents of the test log - it will be specific to a particular update, it won't contain many other updates any more. Every test will be documented on Fedora wiki and the most information will be provided there. 15:33
kparal of course, if you have some related proposals, don't be afraid to talk to us at autoqa-devel mailing list 15:33
kparal and one more thing from the last week: 15:34
kparal #info We have tackled a problem with too low disk space on AutoQA server that resulted in massive deletion of very recent logs. And we will continue to work on it. The problem was caused by some stuck-in-a-loop tests that generated many GBs of test logs. 15:34
kparal ok, that's all I have in my notes. did I forget anything? 15:35
tflink nothing that's coming to mind ATM 15:35
jlaska Regarding the disk space ... we briefly discussed on #fedora-qa earlier that it looks like the corrected tmpwatch command-line used by cron, and the log compression script are working nicely 15:36
jlaska Looks like our scrambling paid off so far 15:36
jlaska oh, and extra kudos for smooge for helping us get more disk space 15:36
tflink jlaska: do you know how many multi-GB log files we've had since thursday? 15:36
tflink jlaska: on second thought, that is a better conversation for not here 15:37
jlaska tflink: okay 15:37
kparal jlaska: ah, I mixed it up a little. yes, the second problem was bug in tmpwatch 15:37
jlaska tflink: I'm looking for any bug reports against RHEL5 tmpwatch ... if not, I'll log something (or check upstream changelogs) 15:37
jlaska kparal: thanks for the updates ... if nothing else, we'll move on to the Open Mic portion 15:38
kparal that's all 15:38
* jlaska hoping for another productive AutoQA week ... in the face of RC1 15:38
* jlaska gives RC1 the evil eye 15:38
jlaska okay, I lied ... let's do a quick look at the QA calendar 15:39
jlaska #topic Upcoming QA events 15:39
jlaska #info Wednesday, May 11 - Final Candidate testing starts 15:39
jlaska #info Tuesday, May 17 - Go/NoGo meeting 15:39
jlaska #info Tuesday, May 24 - Final (GA) release 15:40
robatino would like to be sure that the compose is scheduled for the 11th - in the bugzappers meeting it was stated it would be the 10th 15:40
jlaska robatino: I think we give rel-eng a window to get things lined up ... by Wednesday 15:41
jlaska I think that gives dgilmore a chance to start composing nwo to work out any compose-related problems ahead of time 15:41
jlaska and once the proposed and accepted blocker list is cleared ... I think we can green light the RC1 compose 15:41
* rbergeron wonders if the schedule was confusing to someone or if something is wrong or if someone just read something incorrectly.. ? 15:41
adamw did it change? 15:41
rbergeron it shouldn't have 15:42
rbergeron maybe there's confusion about when it starts and when it is delivered 15:42
robatino in the past the scheduled date for the compose has always been the same as the testing start date 15:43
jlaska adamw: I've tried to accumulate a list of NTH and other important bodhi updates needed for RC1, but I don't know if I'm following the same process you've done in the past 15:43
robatino so when it was stated in the bugzappers meeting that it would be on the 10th, i was confused since the schedule says that testing starts on the 11th 15:43
adamw if you're looking at updates submitted for nth, yes. 15:43
adamw 'other important' updates worries me, though. the whole point of nth is there should be no 'other important' updates. 15:43
adamw anything you want to get in compose after freeze, should go through nth or blocker process. 15:43
jlaska adamw: yup ... they should be NTH ... but no bugs were filed for each of them 15:44
adamw then we should file bugs and jump through the hoops 15:44
jlaska adamw: I don't want to bypass the process, so I ... yes :) 15:44
adamw otherwise we're back to arbitrary freeze breaks and there's no point having a process 15:44
adamw okay :) 15:44
jlaska agreed 15:44
adamw but, i see only two bugs on the 'proposed NTH' list 15:44
adamw (which we should probably review) 15:44
jlaska good call 15:45
jlaska I'll save that for open discussion in a sec ... 15:45
jlaska Anything else related to upcoming schedule? 15:45
rbergeron robatino: so the compose is scheduled to take a full day - which is why testing doesn't start till the next day - i'll go back and see if that was changed at some point or alternately if I broke something at some point (though I don't remember touching this) 15:45
jlaska okay, moving on then ... 15:46
jlaska #topic Open Discussion - <your topic here> 15:46
jlaska Adam reminded me that we have 2 proposed NTH bugs on the list for review ... if no other discussion topics, I'll dive into that 15:46
adamw i'm hoping there'll be three in a minute 15:47
adamw as there's a sugar update we want in but no bug filed for the issues it fixes yet... 15:47
jlaska adamw: btw ... these other bugs I listed are all clones of 699198 (just against different services) 15:47
jlaska it's the same deal ... there are updates that resolve the problems, but no bugs yet ... I'll get those filed post-meeting 15:47
adamw are they all caused by the rpm trigger issue? 15:48
jlaska #action jlaska - file proposed NTH bugs for F14->F15 15:48
jlaska adamw: it's the same end-result ... some have different fixes 15:48
jlaska a default enabled service in F14, isn't enabled post-F15 upgrade 15:48
jlaska #topic https://bugzilla.redhat.com/show_bug.cgi?id=674321 15:49
jlaska #info exo shouldn't take x-scheme-handler/file mimetype 15:49
adamw okay, but different causes. 15:49
jlaska adamw: yup 15:49
adamw these are both Xfce 'blocker' bugs 15:49
jlaska some are incorrect package deps etc.. etc... I'll get them on the list and we can review post-meeting 15:49
nirik this causes problems with people with Xfce + other desktop they use. 15:50
* jlaska reminds our listeners about https://fedoraproject.org/wiki/QA:SOP_nth_bug_process 15:50
nirik they get exo handling links in gnome and such. 15:50
jlaska by definition, non-default desktop issues are automatic NTH bugs 15:50
jlaska "bugs which constitute infringements of the desktop-related Fedora_Release_Criteria as applied to non-default desktops " 15:50
jlaska +1 NTH 15:50
jlaska ack/nak/patch? 15:50
adamw ack 15:51
* nirik nods. 15:51
tflink ack 15:51
* adamw will secretary-ize 15:52
jlaska #agreed 674321 - AcceptedNTH - impacts desktop criteria for non-default desktop 15:52
jlaska adamw: thx 15:52
jlaska #topic https://bugzilla.redhat.com/show_bug.cgi?id=694900 15:52
jlaska #info Missing icon in xfce4-panel 'Add new items' dialog 15:52
tflink isn't that blocking the wrong bug? 15:53
jlaska hmm, this seems border-line ... I'm not sure if our desktop criteria require that *all* menus have icons 15:53
tflink nvm 15:53
jlaska but either way, given this is for a non-default desktop ... this seems fitting for NTH 15:53
adamw jlaska: borderline for the blocker criteria, but i'm happy taking it as nth as an obvious polish issue 15:53
adamw so +1 nth 15:54
tflink and a fix is available 15:54
tflink +1 nth 15:54
jlaska oh, I didn't think borderline for blocker at all 15:54
jlaska +1 nth 15:54
jlaska okay, I think we have enough NTH votes ... 15:54
adamw yup 15:54
jlaska #agreed 694900 - AcceptedNTH - obvious polish issue for non-default desktop 15:54
adamw the sugar thing is a bit complex as the update doesn't specify exactly what triggers the bugs in question, so i'm having trouble filing 'retroactive' bugs for it...i'll do it later and ping you folks 15:54
jlaska btw ... anyone can vote on these issues 15:55
tflink jlaska, adamw: who's doing the bz updates? 15:55
adamw jlaska: do you have bugs ready for the initscript issues? 15:55
adamw tflink: i am, 15:55
jlaska tflink: adamw is 15:55
jlaska adamw: no I don't 15:55
jlaska adamw: that's going to take a bit :) 15:55
jlaska #chair adamw 15:55
zodbot Current chairs: adamw jlaska 15:55
jlaska adamw: can you #topic your sugar bug? 15:55
jlaska stand-by .... 15:58
Viking-Ice drops in late 15:58
jlaska okay, if no other bugs then ... let's close this out and continue on #fedora-qa 15:59
jlaska #topic Open Discussion - Last call 15:59
brunowolff I just got back if you did have something for me. I am just starting to catch up on the log. 15:59
jlaska Welcome Viking-Ice and brunowolff 15:59
adamw jlaska: i don't want to do sugar yet, see above 15:59
jlaska adamw: ah, sounds like you're in the same boat as me :) 15:59
jlaska welcome 15:59
jlaska okay, I'll #endmeeting in 1 minute unless there are any topics 16:00
jlaska 15 seconds until #endmeeting ... 16:00
jlaska Thank you for your time ... I'll follow-up to the list with minutes 16:01
jlaska #endmeeting 16:01

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