QA/Meetings/20121203

From FedoraProject

Jump to: navigation, search

Contents

Attendees

  • adamw (118)
  • tflink (63)
  • Viking-Ice (45)
  • jreznik (17)
  • kparal (13)
  • jskladan (6)
  • zodbot (5)
  • maxamillion (4)
  • mel- (4)
  • pschindl (4)
  • Southern_Gentlem (2)
  • nirik (1)
  • mkrizek (1)
  • satellit (1)

Agenda

  • Previous meeting follow-up
  • Fedora 18 Final status/planning
  • Test case / criteria revision
  • Blocker Meeting Scheduling and Length
  • Open floor

Previous meeting follow-up

  • tflink to ensure some kind of upgrade documentation is ready for beta availability tomorrow - this was done
  • tflink to brief #fedora ops and fedora-user-list regulars on fedup - not done yet, passed to adamw
  • adamw to co-ordinate with anaconda team on TC1 date planning - not done yet but planned for same day, TC1 expected to land during the week
  • jskladan to review final criteria and test cases for obvious revision candidates - passed on to pschindl, he has completed work but needs to send email
  • viking-ice or tflink to try and get a fedup design document out of wwoods - closest thing we have is this blog post

Fedora 18 Final status/planning

  • Beta docs: There was a Beta release announcement which mentioned fedup and kparal and adamw worked on commonbugs
  • To request a release note for a bug, set fedora_requires_release_note flag to ?
  • Potential problem areas: fedup was noted as the big one, jreznik planned to check on it

Test case / criteria revision

  • No new criteria proposals at present, adamw, kparal and pschindl all had some in pipeline

Blocker Meeting Scheduling and Length

  • Suggestion of holding meetings at different times to encourage more participation tabled till F19
  • Some discussion of an idea of holding more but shorter blocker meetings, no definite decision raised
  • In-bug voting generally considered good to reduce blocker meeting burden, but generates BZ spam

Open floor

  • Enterprise storage to be covered next week

Action items

  • adamw to brief #fedora ops and fedora-user-list regulars on fedup
  • jreznik to draft a list of required functionality for fedup for Final
  • adamw to put 'enterprise storage support in newui' on next week's agenda

IRC Log

adamw #startmeeting Fedora QA meeting 16:01
zodbot Meeting started Mon Dec 3 16:01:56 2012 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 #meetingname fedora-qa 16:02
zodbot The meeting name has been set to 'fedora-qa' 16:02
jreznik and a good afternoon/evening for the rest :) 16:02
adamw #topic roll call 16:02
* kparal still here 16:02
* mkrizek is here 16:02
* tflink is here 16:02
* pschindl is here 16:02
* jreznik is here 16:02
* nirik is lurking, ping if needed. 16:02
* jskladan still lurks 16:02
adamw morning everyone 16:03
* maxamillion is here 16:04
* Viking-Ice fetches coffee 16:04
adamw #topic previous meeting follow-up 16:05
adamw oooh. coffee. good idea. 16:05
maxamillion +1 16:05
adamw so we have a giant pile of stuff here 16:05
adamw "tflink to ensure some kind of upgrade documentation is ready for beta availability tomorrow" 16:05
adamw i believe that got done? 16:05
* jreznik thinks so too 16:06
Viking-Ice I'm starting to get a bit worried upgrading encrypted partitions 16:07
tflink yeah, it got mostly done 16:07
Viking-Ice mean upgrading + encrypted partitions 16:07
tflink done enough for beta, anyways 16:07
tflink Viking-Ice: yeah, I want to give that a test - it sounds like there may be dragons in there 16:07
kparal I didn't see any problems except for the timeout 16:08
adamw #info this was done - https://fedoraproject.org/wiki/FedUp#How_Can_I_Upgrade_My_System_with_FedUp.3F 16:08
adamw gr 16:08
adamw #undo 16:08
zodbot Removing item from minutes: <MeetBot.items.Info object at 0x2b6039d0> 16:08
adamw #chair kparal tflink viking-ice 16:08
zodbot Current chairs: adamw kparal tflink viking-ice 16:08
adamw #info "tflink to ensure some kind of upgrade documentation is ready for beta availability tomorrow" - this was done: https://fedoraproject.org/wiki/FedUp#How_Can_I_Upgrade_My_System_with_FedUp.3F 16:08
tflink there are some issues with the release notes, but I think those are getting taken care of 16:08
* tflink will check on that 16:09
adamw "tflink to brief #fedora ops and fedora-user-list regulars on fedup" - did that happen? 16:09
* satellit late and listening 16:09
tflink not so much and it shows - it looks like many people are using old instructions for testing 16:09
adamw okay 16:10
adamw do you want to take it again or should I? 16:10
adamw or anyone else? 16:10
tflink either way, I'm not so sure who the best people to ping are 16:10
Viking-Ice kparal, the timeout issue is present in current GA afaik ( unless it has been fixed have not tested it recently ) just wait entering the password for let's say 5 minutes ( cant remember what the default is ) and you get dropped to systemd shell 16:11
tflink I think that's known, though 16:11
tflink there are multiple bugs filed about it 16:11
adamw yeah, one for dracut and one for systemd. 16:12
tflink two for systemd, I think 16:12
kparal Viking-Ice: it's present in Beta, yes. and the timeout is much shorter, I think 1-2 minutes 16:12
jreznik that's #881670 at least 16:12
adamw #info "tflink to brief #fedora ops and fedora-user-list regulars on fedup" - this was not done yet 16:12
tflink one for regular, one for fedup 16:12
adamw #action adamw to brief #fedora ops and fedora-user-list regulars on fedup 16:12
tflink it shouldn't cause problems that aren't workaround-able by rebooting, though 16:12
adamw it's a bug party! 16:12
Viking-Ice the problem with fedup is that people start upgrade and go doing something else 16:12
adamw "adamw to co-ordinate with anaconda team on TC1 date planning" - yeah, so, oops. 16:13
adamw dgilmore just poked me about this this morning, so i guess we'll work it out after the meeting. but we probably should start doing TCs this week. any objections to that? 16:13
tflink none here 16:13
* jreznik is ok with TCs this week 16:13
Viking-Ice not the more the merrier  ;) 16:13
Viking-Ice mean no 16:13
adamw #info "adamw to co-ordinate with anaconda team on TC1 date planning" - not done yet, will do today, TCs likely to land this week 16:15
adamw "jskladan to review final criteria and test cases for obvious revision candidates" 16:15
adamw jskladan? 16:15
* jskladan is skilled in delegation 16:15
jskladan pschindl did it 16:16
pschindl adamw: I did it, but haven't yet sent the mail 16:16
jskladan he just needs to send the email IMHO 16:16
pschindl I'm going to send it after this meeting 16:16
adamw damnit, petr, stop slacking on jskladan's work ;) 16:16
pschindl  :) 16:16
adamw #info "jskladan to review final criteria and test cases for obvious revision candidates" - passed on to pschindl, he has completed work but needs to send email 16:17
* jskladan is good at training up good interns ;) 16:17
tflink or slacking off, depends on how you look at it :-P 16:18
adamw truly, you are on the road to project colada 16:18
adamw #info "viking-ice or tflink to try and get a fedup design document out of wwoods" - speaking of projects, how is bloodfromastone going? 16:18
adamw alternatively: project excalibur 16:18
tflink not a whole lot of change as of late 16:18
Viking-Ice we need booze lot of booze to get that information 16:19
tflink someone from design (I don't remember who off the top of my head) has started to look @ the gui for gedup-client 16:19
Viking-Ice not the cheap stuff I might add ;) 16:19
tflink as far as a design document goes, the current "design document" is ... (searching for a link) 16:19
tflink http://ohjeezlinux.wordpress.com/2012/11/13/fedup-a-little-background/ 16:20
adamw #info "viking-ice or tflink to try and get a fedup design document out of wwoods" - currently rejoicing in the title of 'design document' is http://ohjeezlinux.wordpress.com/2012/11/13/fedup-a-little-background/ 16:20
tflink there has been some initial work to change mirror manager such that --instrepo won't be required anymore and the .treeinfo will be signed 16:20
* tflink is a little behind on fedup bugs right now - too much blocker bug happy fun time 16:21
jreznik btw do we have a list of required work that has to be done for fedup for final? 16:21
tflink nothing official, no 16:21
* jreznik will start working on it as we really need it (and also opinion from FESCo what they require for final - gui is known...) 16:22
adamw sounds like a plan 16:22
jreznik to avoid late surprises... 16:22
adamw #action jreznik to draft a list of required functionality for fedup for Final 16:22
adamw fffff 16:23
adamw okely dokely 16:23
adamw #topic Fedora 18 Final status/planning 16:23
tflink jreznik: late surprises? we never get those :) 16:23
adamw not sure what tflink meant by 'beta docs status'? 16:23
adamw we have a release announcement which mentioned fedup (yay) and kparal and I worked on commonbugs 16:23
tflink followup from last week to make sure we got everything 16:24
jreznik adamw: thanks for help with the announcement! 16:24
tflink at some point, we might want to think about coordinating better with docs about release notes - there was a mention of a bz flag for highlighting issues but that doesn't need to happen today 16:25
tflink better/differently 16:25
adamw yeah, there is one, i kinda assume people know about it, but maybe not. 16:26
adamw i use it now and again. 16:26
* tflink had never heard of it before that devel@ thread 16:26
adamw welp, the 'fedora_requires_release_note' flag is it, folks. 16:27
adamw i think you set it to ? . 16:27
adamw okay, the other thing was "Final: potential sore points, areas that need testing?", which i guess we've kind of been covering :) 16:31
adamw anything else we didn't cover yet? 16:31
Viking-Ice bootup + black screen 16:32
tflink fedup is mostly what I'm worried about 16:32
Viking-Ice I think I saw people mentioning that it was not only happening with the ati drivers 16:32
Viking-Ice i'm free of it with 3.7 rc kernels 16:33
adamw Viking-Ice: well, i mean, as i explained in a mail, 'it boots to a black screen' is one of the most generic symptoms we *have*. 16:34
adamw a few people seeing that alone doesn't tell us much useful - could be five different bugs. 16:34
Viking-Ice yeah they need to ssh into the machine and grap the log 16:35
Viking-Ice adamw, anyway I dont think those are 5 different bugs thou I think they all relate to that grub gfxpayloud stuff that we changed between releases 16:36
adamw what changed there? 16:38
adamw don't think i'm up on that one 16:38
Viking-Ice the gfxpayload settings 16:39
adamw i mean, what changed specifically 16:39
adamw or are you just talking about that we put in the theming for grub2? 16:40
adamw we were modesetting in f17, i think, but without theming 16:40
tflink I thought that there was a theme in F17 but it might have been added post-release 16:40
adamw okay, anyhow, in general: if we have people hitting black screens, we need more data. 16:42
Viking-Ice hm looks like I never mentioned that on the bug report 16:44
Viking-Ice well I mentioned it to airlied 16:44
Viking-Ice in anycase the "it boots to a black screen" is a regression in my case 16:45
adamw oh, you're hitting it yourself? well, should be easy enough to mess with the grub config and see if that fixes it? 16:45
Viking-Ice F16/F17 ( and pre 3.6 kernel ) worked just fine 16:45
Viking-Ice adamw, I know it fixes it 16:46
Viking-Ice well removing that line atleast does 16:46
adamw ah, okay. 16:46
adamw well, if it looks like other people have the same problem, elevate it to proposed blocker... 16:46
adamw moving on, in the interests of time 16:47
adamw #topic Test case / criteria revision 16:47
adamw do we have much here? 16:47
adamw i don't see any proposals since memory test 16:47
adamw i'm still not getting to the partitioning criteria :( 16:47
tflink it sounds like we're waiting a bit on petr's email 16:48
* kparal didn't manage to write up the kickstart proposal 16:48
kparal somewhere in the queue 16:49
adamw #info no new criteria proposals at present, adamw, kparal and pschindl all have some in pipeline 16:49
adamw #topic Blocker Meeting Scheduling and Length 16:51
adamw tflink, want to take this one? 16:51
tflink sure 16:53
tflink there have been some complaints/suggestions around the blocker meetings recently 16:53
tflink one was the time is inconvenient for some people and it might better to not always use the same time if we want more participation 16:54
tflink which may be true, but I'm tempted to leave that one alone for now 16:54
tflink ie, leave the discussion around that for post-f18 16:54
tflink the other is about frequency and duration 16:54
tflink any thoughts on whether the current format of 3 hour meetings at least once a week vs. 1 hour meetings several times per week? 16:55
adamw i prefer getting it done in one go 16:55
adamw there's quite a bit of 'overhead' which gets multiplied with multiple meetings 16:55
tflink yeah, but we tend to lose people after a while 16:55
adamw obviously, though, 3x3 hour meetings is the worst of the worst :) 16:56
jskladan ^ :) 16:56
Viking-Ice dont we have criteria that hits "* Put advanced storage (filtering, multipath/iscsi/zfcp dialogs) back in." item ( from post-f18 newui TODO on anaconda list ) 16:56
Viking-Ice is anaconda in f18 in good shape for "enterprise storage" 16:56
Viking-Ice  ? 16:56
Viking-Ice  ? 16:56
Viking-Ice dont we have criteria that hits "* Put advanced storage (filtering, multipath/iscsi/zfcp dialogs) back in." item ( from post-f18 newui TODO on anaconda list ) 16:56
Viking-Ice is anaconda in f18 in good shape for "enterprise storage" 16:56
jreznik but seriously - is that even possible? looking on proposed blocker bugs list? 16:56
Viking-Ice adamw, do you have any clue on the enterprise storage part of anaconda 16:57
tflink yeah, I don't think anyone likes the current method 16:57
tflink Viking-Ice: that seems to be a bit off topic 16:57
Viking-Ice you where speaking of the criteria to begin with 16:57
tflink but installing to iSCSI, FC, FCoE etc. is not in F18, will return in F19 IIUC 16:57
tflink I was? 16:58
adamw Viking-Ice: like half an hour ago? 16:58
adamw so we tried this thing this week where tflink categorized the bugs for on-bug voting 16:58
adamw what did everyone think about that? 16:58
Viking-Ice adamw, more like 10 minutes 16:58
Viking-Ice and I'm the actual one that was proposing we go for one hour meetings 16:58
adamw Viking-Ice: still, we moved on to a new topic since then...there's always open floor if you want to bring up something from before 16:58
adamw otherwise we just get confused 16:58
Viking-Ice adamw aha 16:58
* maxamillion is almost always confused anyways 16:59
adamw =) 16:59
Viking-Ice more frequently 16:59
adamw Viking-Ice: since we're discussing the length of blocker meetings now, talking about enterprise storage criteria seems a bit out of place :) 16:59
* maxamillion is just getting worse at multi tasking ... $day_job is more busy than $old_day_job 16:59
Viking-Ice adamw, you weren't when I asked those questions 16:59
Viking-Ice you conveniently ignored it 16:59
adamw Viking-Ice: er - i set the topic at xx:51 to "Blocker Meeting Scheduling and Length" 17:00
adamw you asked your questions at xx:56 17:00
adamw after there had already been several minutes of discussion on the blocker meeting length topic 17:00
Southern_Gentlem move on and deal with this later 17:00
tflink yep 17:00
Viking-Ice adamw, no I re-asked those question at that time 17:00
tflink the blocker meeting is supposed to be starting soon 17:00
kparal (now) 17:01
Viking-Ice yeah I proposed for 3x1 hour or 5x1 meeting instead of 3 hours meeting 17:01
adamw Viking-Ice: oh. i never got the originals. maybe they were affected by that netsplit i see in the history. sorry 17:01
tflink I didn't see them either 17:01
* jreznik does not have that question neither... 17:01
Southern_Gentlem perfer we dont have that many blockers so 1 -1 hr meeting can deal 17:01
adamw that's obviously the best 17:02
adamw but it seems unrealistic 17:02
tflink especially right now 17:03
adamw how can we have such a low blocker count without evaluating proposed blockers and rejecting some? which is...what we do in the meeting? :) 17:03
jreznik clean-up in tickets should help a little - /me voted in several bugs today 17:04
adamw yeah, i think that was a good idea 17:04
adamw is anyone worried about covering at least 'obvious' bugs with in-bug voting? 17:04
adamw it seems like the best way to reduce the load a little 17:04
tflink yeah, I need to go through and modify the ones that have enough -1s or +1s 17:04
adamw ah, i was about to ask if you'd done that 17:04
tflink I'm making the list for today's meeting from bugs that weren't on the 'more obvious' list 17:05
tflink we have enough to go through that it shouldn't be an issue 17:05
adamw sounds good 17:05
adamw shall we take a vote on the 'many short meetings' proposal? 17:05
adamw or more discussion on it? 17:05
Viking-Ice should we have it 1 hour or 1 and half hour 17:06
kparal if the obvious list is compiled by someone and sent to the list, I'm OK. just going randomly though blocker list doesn't seem great 17:06
Viking-Ice ( takes 10 minutes to start ) 17:06
tflink kparal: I sent the list out to test@ on friday 17:06
kparal tflink: yeah, I know 17:06
tflink ok, you meant in general 17:06
adamw right, you're saying it should always be done that way, make it a process? 17:06
kparal yes, something like that 17:07
kparal it's better to have a list of obvious blockers, and then people can vote in the bugzilla or say "no this is not obvious" 17:07
adamw maybe we should have some kind of threshold at which the 'formal on-bug voting' process kicks in - >20 proposed blockers or something 17:07
kparal if we don't have the list, each person have a different opinion what is obvious 17:07
adamw sure 17:08
jreznik kparal: but you can still vote and other people can say no 17:08
tflink yeah, but I think that's a bit unavoidable for now 17:08
kparal jreznik: without the list I don't know they voted 17:08
adamw Viking-Ice: i'd prefer 1.5 to 1, yeah, the 10 minute overhead is significant 17:08
tflink the point of going through and doing some sorting is to reduce the number of bugs to discuss in meetings - I don't see a way to do that without one person doing the initial sorting 17:08
jreznik would it be possible to parse the bug for "-1/+1 blocker" in the current blocker bug list and show it? 17:09
tflink which is unavoidably biased to a certain point 17:09
Viking-Ice so try 3x1.5 ( monday/wednesday/friday ) 17:09
jreznik tflink: but yeha, someone has to do the initial sort 17:09
tflink jreznik: yeah, that wouldn't be too hard in principle - the hard part is making sure to catch all the minor variations in +/-1 17:10
tflink it might be interesting to add some support for flagging "obvious" bugs in the tracker app 17:10
tflink but there is no way I'm going to get to that until after F18 17:10
adamw okay, so sounds like we're broadly on board with the in-bug voting, i'm not hearing much discussion of 'multiple short meetings' 17:11
adamw and we're 10 minutes over time 17:11
tflink I think it's an interesting idea but I also think we need to get through the monster list sooner than later 17:11
Viking-Ice so should we discuss in-bug the gray area we might be hitting 17:11
tflink so for now, I'm -1 on the idea of shorter meetings 17:12
* tflink emphasizes "for now" as in at least until we get through the initial list 17:12
Viking-Ice I'm not so sure that maintainers will be happy about the bug spam we introduce by voting in the bugs themselves 17:12
* adamw is +/-0 - personally i prefer longer-but-fewer, but i certainly acknowledge the problem of losing people as the meetings go on 17:12
adamw Viking-Ice: that's a good point, actually, hadn't thought of that 17:13
adamw maybe we should check on devel@ 17:13
adamw #info viking-ice points out that a drawback of in-bug voting is bugzilla spam 17:13
jreznik let's go through the current list and we will see how many left - we can be flexible 17:13
adamw #info aside from that, general support for in-bug voting on 'obvious' blockers when the blocker count is high, but it should be a defined process 17:13
adamw #info no-one seems to have strong feelings either way on the multiple-short-meetings plan, but we don't have time to thrash it out further today 17:14
tflink I think that the blocker process could use some work, but that doesn't help for now 17:14
adamw anyone want to take an action item for considering a formal in-bug-review process further? 17:14
tflink I'm not against the idea, but I don't really want to do it right now 17:15
* tflink won't stop anyone else from doing it, though 17:15
adamw i guess everyone's a bit overloaded at present 17:15
adamw let's go on to open floor so we can discuss viking's missed question and get to blocker review 17:16
adamw #topic open floor 17:16
kparal I'm not really against in-bug voting, but I really like meeting voting more 17:16
adamw Viking-Ice: sorry your criteria question got missed earlier, what was it again? 17:16
Viking-Ice adamw, let's just add enterprise storage support in anaconda ( if any ) to next meeting item and start working on the blocker bugs 17:17
adamw okay 17:17
adamw #info viking-ice is concerned about storage support in newUI but meeting has overrun so we'll cover it next week 17:17
adamw #action adamw to put 'enterprise storage support in newui' on next week's agenda 17:17
Viking-Ice *enterprise* 17:17
adamw yeah, got it in the action item :) 17:17
Viking-Ice  ;) 17:17
adamw anything else for open floor? 17:18
mel- what does 'open floor' mean? 17:19
adamw topics that weren't covered elsewhere in the meeting 17:19
adamw though we'd like to finish quickly to get on to the blocker review meeting 17:19
adamw did you have something to bring up quickly? 17:19
jreznik blocker fun! any other topic could be re-raised after it... if not serious one 17:20
mel- adamw: well, i need to fill a fedup bug. dunno of that is appropriate here 17:20
mel- s/of/if/ 17:20
adamw mel-: are you having trouble filing it in bugzilla? 17:21
mel- adamw: no, i think will be fine :) 17:23
adamw okay, let's move on to blocker review, we can help mel outside of the meeting 17:23
adamw thanks for coming everyone! 17:23
adamw #endmeeting 17:23

Generated by irclog2html.py 2.11.0 by Marius Gedminas - find it at mg.pov.lt!