From Fedora Project Wiki

< QA‎ | Meetings
Revision as of 17:34, 11 February 2013 by Tflink (talk | contribs) (Post-meeting stuff (logs, action items etc.))
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)

Attendees

  • tflink (131)
  • Viking-Ice (49)
  • kparal (23)
  • dan408- (15)
  • nirik (8)
  • zodbot (6)
  • satellit (6)
  • misc (2)
  • jsmith (2)
  • jskladan (1)
  • akshayvyas (1)
  • pschindl (1)

Agenda

  • Previous Meeting Followup
  • Criteria Presentation Revision
  • Fedora 19 Test Days
  • Retrospective Review
  • Open Floor

Previous Meeting Followup

  • adamw to draft up a proposal for revising the presentation of the release criteria
    • Work has been started on revising the presentation of the release criteria, not ready for proposal yet
    • Topic for discussion later in meeting
  • adamw to draft up a kickstart criterion for f19
    • Has not been done

Criteria Presentation Revision

Fedora 19 Test Days

  • We need to kick off the F19 Test Day cycle - who wants to own it? Ideas for Test Days?
    • No volunteers to own the F19 test day cycle
  • Viking-Ice stated that he plans to host an iscsi test day for F19 (date TBD)

Retrospective review

  • Let's take a look through the Fedora_18_QA_Retrospective to see if anything jumps out
    • Nothing constructive added
    • Please add constructive thoughts to the retrospective
  • Viking-Ice requested discussion about the blocker review process (tabled to another meeting)

Open floor

IRC Log

tflink #startmeeting Fedora QA Meeting 16:01
zodbot Meeting started Mon Feb 11 16:01:02 2013 UTC. The chair is tflink. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:01
zodbot Useful Commands: #action #agreed #halp #info #idea #link #topic. 16:01
tflink #meetingname fedora-qa 16:01
zodbot The meeting name has been set to 'fedora-qa' 16:01
tflink #topic Roll Call 16:01
tflink #chair kparal 16:01
zodbot Current chairs: kparal tflink 16:01
* kparal waves 16:02
* satellit listening 16:02
* jskladan tips his hat 16:02
dan408- im not here 16:03
* kparal nudges pschindl 16:03
* tflink waits another minute before getting things started 16:04
tflink bit of a small group today but let's get started 16:05
dan408- no adamw? 16:05
tflink holiday in canada 16:06
dan408- snowboarding holiday 16:06
kparal adamw is slacking off 16:06
dan408- okay continue please 16:06
tflink and a public holiday, but either way 16:06
tflink #topic Previous Meeting Followup 16:06
* pschindl is here 16:07
tflink #info adamw to draft up a proposal for revising the presentation of the release criteria 16:07
tflink #info adamw has been working on the release criteria presentation, no proposal has been sent out yet. will be covered later in meeting 16:08
tflink #info adamw to draft up a kickstart criterion for f19 16:08
tflink I don't think this was done, either 16:08
tflink did I miss something? 16:09
dan408- nope 16:09
dan408- +1 16:09
tflink #info this hasn't been done yet 16:10
tflink anything else from meeting followup? 16:10
tflink I don't see anything from the minutes 16:10
* tflink assumes silence == nothing else 16:11
tflink #topic Criteria Revision 16:11
* Viking-Ice joins in 16:12
dan408- hello Viking-Ice 16:12
tflink We started the conversation a bit last week, but it sounded like there was general agreement that the presentation of the release criteria could be improved 16:12
tflink #info adamw has been working on a draft update to the presentation of the release criteria 16:12
tflink #link https://fedoraproject.org/wiki/User:Adamwill/Draft_alpha_criteria_sandbox 16:13
kparal interesting 16:13
tflink I don't think that he's done yet but it shows a re-thinking of the current presentation 16:13
dan408- where can we make suggestions here? 16:13
kparal dan408-: test list 16:14
tflink dan408-: test@ 16:14
tflink it turns out that figuring out how to present the information is not an easy problem 16:14
dan408- sigh 16:14
tflink one thought that I had was to take the criteria out of the wiki so that we aren't restricted by wiki formatting 16:15
dan408- there's nothing wrong with the wiki 16:15
dan408- maybe we can use talk? 16:15
dan408- test is way too busty 16:15
dan408- i dont even pay attention to it 16:15
Viking-Ice tflink, yeah if we can put it on the qa criteria page you are working on 16:15
dan408- https://fedoraproject.org/w/index.php?title=User_talk:Adamwill/Draft_alpha_criteria_sandbox&action=edit&redlink=1 16:15
tflink I took a snapshot of what adamw was working on, translated it to rst and generated a really basic proof-of-concept 16:15
Viking-Ice I dont think nesting this is a way to go as adam is doing there 16:16
tflink #info tflink made a proof-of-concept of taking the release criteria out of the wiki 16:16
tflink #link http://tflink.fedorapeople.org/criteria_display/ 16:16
tflink I haven't done _any_ theming or admonition work yet 16:16
Viking-Ice since it requires the user to perform additional click to get that information 16:16
tflink so the layout is really rough 16:16
dan408- this is prettier.. 16:16
tflink but my thought is that it would allow for easier integration with the blocker proposal app, would give us _much_ more flexibility in presentation and wouldn't make it too much more difficult for people to contribute to the criteria writing 16:17
kparal I have some comments about it, but I think this would be best discussed on the list 16:17
Viking-Ice should really that many people be editing the criteria 16:18
tflink the whole thing would live in a git repository, criteria revisions could be discussed like code reviews and the static html would be generated automatically on push to git master 16:18
Viking-Ice tflink, your page is better presented then what adam has been working on 16:18
tflink the question becomes how long it would take to translate everything 16:19
Viking-Ice should it really be translated ? 16:19
tflink but either way, we need to figure out how to present the information 16:19
tflink Viking-Ice: translation from mediawiki to rst 16:19
tflink poor choice of words on my part, sorry 16:19
Viking-Ice ok took it as language translation 16:19
tflink yeah, I should have phrased it differently 16:20
tflink one thought was to talk with the design team about how to present the information 16:20
kparal I don't think we need the design team for everything. this page should be mainly useful for us, because general users/testers won't read it anyway 16:21
kparal we should have a simplified version for them, and _that_ could be discussed with the design team 16:21
tflink I'm jumping around a bit but the main concern that I have about moving the criteria out of the wiki is the time needed to write all the templates/css and the work needed to set up reviews and the automatic generation of html 16:22
kparal tflink: how do you imagine the review process? 16:22
tflink kparal: agreed, but I also think that it would be good to see if we can get some outside ideas on how to present the information in a relatively easy-to-read fashion 16:23
Viking-Ice tabs? 16:23
tflink if we want people to cite criteria when proposing blockers, they need to be relatively easy to understand 16:23
tflink kparal: reviewboard 16:23
kparal that's yet another project to polish :) 16:24
tflink I finished setting up a reviewboard instance that we're using for the blocker tracking app 16:24
tflink it will do email notifications of code reviews on top of supporting inline and overall comments 16:24
kparal great 16:24
tflink the only thing it doesn't do is integrate with FAS 16:24
tflink and that's due to lack of support for openid upstream 16:25
Viking-Ice but so we are clear we are only talking about revisiting the presentation of the criteria not the criteria itself ? 16:25
Viking-Ice that topic can be a bit misleading 16:25
tflink Viking-Ice: at the moment, yes 16:25
tflink I assume that's what adamw had in mind when he wrote the agenda, anyways 16:25
kparal if we don't have FAS integration, we can use github comments, for instance. is there any further benefit of reviewboard? 16:26
tflink #info as a clarification, this topic is about the presentation of the release criteria, not revising the criteria themselfs 16:26
tflink kparal: I'd really rather not rely on github if we don't have to but it sounds like I might about the only one who thinks that 16:27
* kparal shrugs 16:27
Viking-Ice no we should keep this inhouse 16:27
kparal it might be easier than to maintain another app 16:27
tflink kparal: yeah, it would be 16:27
kparal if github explodes, it doesn't hurt much 16:27
tflink and it's not like we would be the only ones doing it - pretty much all of fedora infra is on github now 16:28
Viking-Ice interesting 16:28
kparal except some of the past discussion could be inaccessible 16:28
tflink kparal: it's still a hosted service outside our control and not open source 16:28
kparal I don't have a clear opinion on this, I just sometimes feel like having a NIH syndrome 16:29
misc tflink: mhh, no, I think the fedora infra is still in house 16:29
akshayvyas oops i am late :-( 16:29
misc ( and pushing stuff to git hub make it more complex since this is not tied to FAS ) 16:29
tflink misc: I'm pretty sure they moved everything after FUDCon Lawrence 16:29
tflink https://github.com/fedora-infra 16:30
kparal misc: yeah, that's true 16:30
tflink to clarify, I mean most of the apps that fedora infra uses 16:30
nirik we did not move everything. ;) We allowed projects that wished to use it to use it. 16:30
tflink nirik: outside of koji, which projects didn't move? 16:30
tflink but I digress a little bit 16:31
kparal tflink: I think you should post the links (when it's somewhat ready) to test list, there will be surely lots of interesting comments 16:32
nirik lots of things I'm sure. I don't have a list. 16:32
tflink #info one large concern about moving the criteria out of the wiki is the theming work and the work needed to re-generate the HTML on git push 16:33
tflink #info both reviewboard and github are possibilities for reviewing criteria changes if we moved out of the wiki 16:33
tflink either way, we wouldn't be the first group that did stuff outside of fedorahosted 16:34
tflink kparal: yeah, I imagine that there would be some interesting discussion around any proposed changes 16:35
kparal :) 16:35
tflink but we don't have a whole lot of time left before F19 16:35
tflink any other thoughts on this for today or is it OK to move on to the next topic? 16:36
kparal proposal: let's skip testing F19 and hack on infra instead 16:36
kparal let's move on 16:36
tflink kparal: as much as I'd like to +1000 that ... 16:36
tflink #topic Fedora 19 Test Days 16:37
tflink now that the F19 schedule has been accepted, we have test days to organize 16:37
tflink organize/oversee/whatever-you-want-to-call-it 16:37
tflink is anyone interested in volunteering to coordinate the F19 test days? 16:38
tflink don't everyone volunteer at the same time :-P 16:39
kparal this is something I don't really enjoy :) 16:39
Viking-Ice just ensure all the *D 16:39
* tflink hears a volunteer :-D 16:39
Viking-Ice just ensure all the *DE testing takes place after alpha 16:40
tflink any ideas for test days that haven't already been proposed? 16:40
tflink I think that KDE and GNOME test days have already been requested for F19 16:40
Viking-Ice I'm just referring to the schedule 16:40
Viking-Ice I'm planning on hosting a iscsi test day 16:40
Viking-Ice we fully migrated it to systemd sometime before christmas 16:41
tflink power management and NM test days have also been requested 16:41
tflink #info Viking-Ice suggested an iscsi test day 16:42
tflink if there's nothing else, we can move on 16:43
Viking-Ice tflink, I did not suggest I *will* be hosting a iscsi test day 16:43
tflink #undo 16:44
zodbot Removing item from minutes: <MeetBot.items.Info object at 0xb87ed50> 16:44
tflink #info Viking-Ice will be hosting an iscsi test day 16:44
Viking-Ice date scheduled on later date ;) 16:44
tflink #undo 16:44
zodbot Removing item from minutes: <MeetBot.items.Info object at 0xba8ced0> 16:44
tflink #info Viking-Ice will be hosting an iscsi test day. Exact date TBD and proposed later 16:45
tflink ok, anything else on this? 16:45
Viking-Ice we need to test "enterprise storage" in anaconda it's expected to be back ( gui wize if I'm not mistaken ( 16:45
Viking-Ice uhum ) 16:45
Viking-Ice ;) 16:45
Viking-Ice this release cycle 16:45
tflink Viking-Ice: yeah but that's hard to do in a test day - few people have access to "enterprise storage" 16:46
tflink well, access to "enterprise storage" they can test fedora installs with 16:46
Viking-Ice puff how does not have a san rack at home 16:46
Viking-Ice ;) 16:46
* tflink hasn't touched a SAN since his days in storage testing 16:47
Viking-Ice nirik, what kind of storage infrastructure does infra have 16:47
Viking-Ice local disk only or something more "advanced" 16:47
tflink mostly netapp filers, I think but testing fedora installs in the fedora infra sounds like a really bad idea to me 16:47
nirik we have local disk/lvm and some iscsi stuff. Nothing too odd 16:48
Viking-Ice tflink, why is that a bad idea? 16:48
tflink Viking-Ice: what part of testing installs of a pre-release OS on production infrastructure doesn't sound like a bad idea? 16:49
Viking-Ice tflink, hows exporting an empty share to on an pre-release os 16:49
Viking-Ice it's not like they would export something with mission critical data on it to the clients 16:50
Viking-Ice just and empty filesystem/partition 16:50
Viking-Ice s/and/an 16:50
nirik we possibly could, but don't want it to interfere with any production setups... would have to look if we have spare/isolated space for such a thing 16:51
tflink that's a discussion you/we can have with infra, if they're OK with it, I'm not going to stop attempts - I'm just really cautious about addressing storage 16:51
* nirik nods. I don't want prod interfered with. 16:52
* tflink has seen far too many "enterprise" storage systems explode 16:52
tflink but we digress 16:52
tflink this is a discussion that can happen outside of the meeting, on test@ and with infra 16:52
jsmith tflink: :-) 16:53
tflink since we're almost out of time ... 16:53
tflink #topic Fedora 18 Retrospective 16:53
tflink This is something that we haven't talked about too much post-F18 16:53
tflink a few things have been added to the retrospective page 16:53
Viking-Ice it's simply really never do anaconda/upgrade rewrites again 16:53
tflink #link https://fedoraproject.org/wiki/Fedora_18_QA_Retrospective 16:53
tflink Viking-Ice: while I don't think that we'll see another major rewrite like that anytime soon, I don't think it's as simple as "never rewrite again" 16:54
tflink just go about doing it differently _if_ it ever happens again 16:55
tflink but either way, constructive additions to that page would be appreciated 16:55
Viking-Ice tflink, no really if anaconda needs another rewrite in the distant future it should just be thrown away altogther and replaced with something written from scratch 16:55
tflink Viking-Ice: I don't see how it's our place to tell devs what to do 16:56
* satellit require install on hardware plus in Virtualiztion 16:56
Viking-Ice tflink, it is when their mess ends up at our door step ;) 16:56
tflink #info Constructive additions to the retrospective are appreciated 16:56
jsmith Viking-Ice: Most of the anaconda code *has* been rewritten from scratch 16:56
Viking-Ice hurray! 16:56
tflink Viking-Ice: we'd be naive to pretend that it was all their fault 16:57
tflink there is plenty of blame to go around 16:57
tflink blame/fault 16:57
Viking-Ice fesco has the other half 16:57
tflink but blamestorming doesn't help anything 16:57
Viking-Ice in anycase "I wish, that f18 alpha was not declared GOLD, knowing that it will destroy data, by default. Even after such a decision was made, i wish the announcement informed us clearly that f18alpha *will destroy data*, and not wait one more week until the release notes become available. " 16:57
Viking-Ice I thought it was common knowledge that pre-release testing did make the requirement to throw away their current installs at release time 16:58
tflink #info Constructive additions to the retrospective page would be appreciated, tickets will be created from retrospective items shortly 16:58
tflink Viking-Ice: I don't see that on the retrospective, where did it come from? 16:59
Viking-Ice it's there straight under "Could have been better" 16:59
tflink oh, I skipped over it due to formatting 16:59
Viking-Ice the rest seem to be just adamw with criteria bits which we are already working at and Wutao85 with personal retrospective 17:00
Viking-Ice as in himself could do better at replying to bugs 17:01
tflink hence the request for more ideas :) 17:01
tflink anyhow, we're over time already 17:01
tflink anything else that people want to cover re: F18 retrospective 17:01
tflink ? 17:01
Viking-Ice well what did we learn about the blocker bug meetings 17:02
tflink that we didn't already know before F18? 17:02
tflink they don't scale all that well 17:02
tflink but the hope is that F18 was very much an outlier 17:02
Viking-Ice well we should set a fixed channel and keep with the 3hour max limit 17:03
Viking-Ice we kinda ended up doing that 17:03
tflink sure, but I think that the blocker review process is a little outside the F18 QA retrospective 17:03
tflink especially when we're already over time 17:04
* tflink isn't trying to say that the process is perfect or that it shouldn't be discussed 17:04
Viking-Ice yeah sure whatever fits your meeting needs 17:04
* satellit added comment to retrospective 17:05
tflink #info discussion around the blocker review process for F19 would be wise before we get into testing 17:05
tflink #topic Open Floor 17:06
tflink Anything else that should be discussed today? 17:06
satellit when will koji try again for lives? 17:06
tflink I think there were some problems with rawhide over the weekend 17:06
tflink something about getting comps, I think but I didn't read the details 17:07
tflink either way, I'm not sure what the decision was regarding the frequency of rawhide compose attempts 17:07
* nirik restarted rawhide compose for today, should land in a few hours. 17:07
tflink does anyone else know what the plan is? 17:07
satellit thanks 17:08
tflink s/else// 17:08
nirik lives I can do after that. for install there was talk of doing weekly... but thats not yet started that I know of. 17:08
tflink I think that dgilmore was working with anaconda devs when they requested a compose 17:09
satellit FYI I installed anaconda from http://kojipkgs.fedoraproject.org//work/tasks/2340/4912340/Fedora-19-Nightly-20130129.10-x86_64-Live-xfce.iso 17:09
tflink anyhow, if there are no other topics, I'll set the fuse for ~ 5 minutes 17:10
Viking-Ice yeah it kinda serves no purpose to start composing images until anaconda says go 17:10
* nirik nods 17:10
tflink Viking-Ice: IIRC, they were one of the groups requesting rawhide composes 17:10
Viking-Ice yup 17:11
tflink Thanks for coming, everyone! 17:14
tflink #endmeeting 17:14

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

Retrieved from "https://fedoraproject.org/w/index.php?title=QA/Meetings/20130211&oldid=323003"