From Fedora Project Wiki

< QA‎ | Meetings


  • adamw (111)
  • tflink_ (40)
  • j_dulaney (39)
  • kparal (34)
  • tflink (9)
  • zodbot (5)
  • jsmith (4)
  • nirik (2)
  • brunowolff (2)
  • robatino (2)
  • pschindl (2)
  • jskladan (1)


F16 retrospective task update

  • Two tickets are closed already and most of the others are being worked on actively

FUDCon Blacksburg notes

  • tflink, adamwill, j_dulaney and red_alert will all be present
  • j_dulaney is planning an autoqa hackfest on friday
  • The anaconda ui test hackfest is planned for all day sunday
  • We will try to meet up for lunch as a group on friday

AutoQA update

  • kparal added caching proxy support inside AutoQA
  • kparal fixed several issues with sending out opt-out and opt-in emails
  • autoqa-results list is back, now with fewer messages!
  • kparal proposed using an icon that rather resembles the version of Marvin from the recent H2G2 movie as AutoQA's icon, tflink and adamw raised legal concerns; kparal will check in with spot regarding this

Open floor: remote_install update

  • tflink notes the remote_install project is working at demo quality behind the RH VPN at present
  • public demo will be made available soon using a public cloud system, probably ohi


tflink #startmeeting Fedora QA meeting 16:00
zodbot Meeting started Mon Jan 9 16:00:45 2012 UTC. The chair is tflink. Information about MeetBot at 16:00
zodbot Useful Commands: #action #agreed #halp #info #idea #link #topic. 16:00
tflink #meetingname fedora-qa 16:00
zodbot The meeting name has been set to 'fedora-qa' 16:00
tflink #topic Roll Call 16:00
tflink Who's ready for some meeting fun time? 16:01
tflink #chair adamw 16:01
zodbot Current chairs: adamw tflink 16:01
adamw huh 16:01
adamw darn xchat and suspend/resume. 16:01
* jsmith lurks (during another conference call) 16:01
* kparal opts in for fun time 16:01
adamw i already started the meeting, in my own private bubble. :) 16:01
* jskladan is in 16:01
tflink adamw: a meeting of one! 16:02
* brunowolff is here 16:03
adamw tflink: keeps the riff-raff out! 16:03
tflink adamw: and I opened the flood gates :) 16:04
adamw alright, so - let's get going 16:06
adamw #topic f16 retrospective task update 16:06
adamw this is just my excuse to bug everyone about the stuff that came from the f16 retrospective, and remind us all it still needs doing :) 16:06
adamw 16:07
adamw I haven't updated the 'closed' statuses there, btw. 16:07
tflink do you want to go through the individual tickets or just have us pipe up on status? 16:08
adamw no need to review each one i don't think 16:08
adamw just note that out of nine qa tickets, two are closed 16:08
adamw most of the others look like they have some activity; #151 (pschindl) and #259 (twu) don't 16:09
tflink I'm thinking that the Xen ticket can be closed, I should have looked at the first proposal more closely 16:09
adamw anyone know where they're at? 16:09
adamw tflink: that's fine, if the current test is complete and comprehensive 16:09
kparal I pinged pschindl 16:10
adamw jskladan: kparal: any idea of pschindl's status on his ticket? 16:10
adamw ah, thanks 16:10
kparal I don't 16:10
kparal know 16:10
kparal let's come back to it when he responds 16:11
adamw yup 16:12
adamw overall looks like we're doing okay on those tasks, so that's good. remember, we have until alpha starts up in earnest to get them done. 16:12
adamw #info two tickets are closed already and most of the others are being worked on actively 16:13
adamw #topic FUDCon Blacksburg notes 16:13
adamw hey, jay 16:13
* j_dulaney appologizes for tardiness 16:13
adamw that's all good 16:13
adamw we've just been over the retrospective tickets so far 16:14
adamw anything you wanted to add on that topic? 16:14
j_dulaney I'll check the logs and comment on the ML 16:14
adamw okay 16:15
adamw so, for fudcon: doesn't look like there are any big issues 16:15
j_dulaney What's been said so far on FUDCon 16:15
adamw nothing, i'd just set the topic when you waltzed in :) 16:15
adamw good lord, the tflinks are spawning 16:15
adamw i think he's got us surrounded 16:15
j_dulaney LOL 16:16
tflink_ woah, that's strange 16:16
* j_dulaney is reminded of the British being attacked by Zulu 16:16
adamw ah, a classic pincer movement 16:17
adamw watch out for tflink__ on the right flank 16:17
j_dulaney LOL 16:17
adamw :) 16:17
adamw so as i was saying... 16:17
adamw looks like there's no problems with fudcon NA 16:17
adamw for those playing along at home, it's happening this weekend in blacksburg, which is in virginia (at least i hope it is, or i'm going to the wrong darn airport) 16:18
jsmith adamw: If only all of us could say that :-) 16:18
adamw tflink and I will be on hand to ruin QA's reputation and create gigantic bar bills for jsmith 16:18
j_dulaney LOL 16:18
adamw red_alert will also be there 16:19
* j_dulaney is going to be there 16:19
adamw and j_dulaney...yep 16:19
j_dulaney I'll try to run a bit of an AutoQA hackfest on Friday 16:19
adamw awesome 16:19
adamw i guess tflink may be able to help out there 16:19
* j_dulaney needs to get new headphones; these temp cheapos suck 16:19
adamw i thought sunday was for hackfests and such 16:19
adamw ? 16:19
kparal j_dulaney: what's the agenda of that hackfest? 16:20
j_dulaney Friday, too 16:20
j_dulaney kparal: Self testing 16:20
adamw okay 16:20
* j_dulaney keeps meaning to blog about it 16:20
adamw yeah, blog is always a good idea 16:20
kparal I'll be looking out for it 16:20
kparal thanks 16:20
adamw i also need to remember to blog about it :) 16:20
adamw so aside from that, we have the anaconda ui testing hackfest planned for monday, red_alert is taking care of that 16:21
j_dulaney Sunday 16:21
adamw erf. yeah. 16:21
adamw it would be a lonely hackfest on monday... 16:21
j_dulaney Well, probably be some students there 16:21
adamw and i think various of us have presentations to propose. 16:21
adamw yeah, wondering where the Major Battles of the Zulu Wars class went wrong ;) 16:22
tflink_ yeah, the barcamp proposal list looked a bit sparce last night 16:22
adamw tflink_: usually a lot of people don't bother listing talks ahead of time. 16:22
adamw or writing them ahead of time, in some cases (koff koff) :P 16:22
adamw but looks like we more or less have our plans in order, so that's good. 16:23
adamw i recommend making sure to eat with jsmith as much as possible so that red hat pays for your food. protip! 16:23
j_dulaney Indeed, especially as broke as I am 16:24
adamw awesome. so, any outstanding issues / questions on fudcon? 16:24
j_dulaney Where/how/when should we meet up? 16:25
j_dulaney QA party 16:25
j_dulaney With hot dogs 16:25
jsmith adamw: Thanks man :-) 16:26
adamw j_dulaney: good question...maybe we should aim to do a group lunch on one of the days? 16:26
j_dulaney Indeed 16:26
adamw i'm going to try and get a t-mobile prepaid sim this time, so i should have phone comms 16:27
* j_dulaney admits that he doesn't know what y'all look like 16:27
adamw j_dulaney: tflink and I and red_alert all know each other 16:27
j_dulaney So, find one, find all? 16:27
adamw yeah 16:27
brunowolff I am off to a work meeting. 16:27
adamw and usually it's not THAT hard to find people at fudcon; it's not so big 16:27
adamw alrighty bruno, cya later 16:27
nirik (just hang out at the bar... people will come to you. ;) 16:28
adamw pretty much everyone will be in one place during the initial keynote and meeting proposal period for e.g 16:28
j_dulaney Indeed 16:28
adamw we could aim to try and at least identify each other then, and then meet up again for lunch after the first session of talks, perhaps. 16:28
adamw sound good? 16:29
tflink_ sounds like a plan to me 16:29
j_dulaney Indeed 16:29
* j_dulaney notes that it is likely he'll be the biggest there 16:30
j_dulaney With a beard 16:30
adamw ah, that'll make you stand RIGHT out in a meeting of *nix geeks ;) 16:30
j_dulaney LOL 16:30
adamw alright, we'll go with that plan. i'm sure it'll work out somehow. 16:31
j_dulaney Too bad there won't be any jlaska 16:31
* j_dulaney knows him 16:31
adamw yeah, it would've been good if james was there 16:31
adamw wwoods will be there, though. with an awesome badge. 16:32
adamw alrighty, moving on 16:33
adamw #topic AutoQA update 16:33
adamw kparal, tflink - what's the news? remember it's been like a month since the last meeting. dec 11th, I think. 16:33
kparal I have some summary 16:33
* kparal taking the mic 16:33
kparal The last meeting was a long time ago, so I'll include also some items that happened before Christmas. 16:34
kparal 1. I added support for caching proxy inside AutoQA. This is useful mainly in devel environments where we don't have too much of bandwidth. Subsequent test executions are very fast. We can also share these files between multiple developers in an office by deploying the proxy on a shared computer. 16:34
adamw #chair kparal 16:34
zodbot Current chairs: adamw kparal tflink 16:34
kparal 2. There has been several problems in sending out (optin and regular) emails from our test clients. Hopefully I corrected all of them and optin email should be flowing once again. 16:34
kparal big thanks to jlaska here 16:35
adamw #info kparal added caching proxy support inside AutoQA 16:35
adamw #info kparal fixed several issues with sending out opt-out and opt-in emails 16:35
kparal 3. autoqa-results mailing list is revived. I cut down the amount of traffic substantialy by only sending full logs from test results. Currenly only emails from staging server are routed there. Production server still generates too much traffic, I'm working on it. I'll probably implement a configurable blacklist on these emails. 16:35
kparal 4. After re-enabling autoqa-results I spotted (too many) problems. Most of them should be reported as Hot issues in our Trac, free to take :) 16:36
adamw #info autoqa-results list is back, now with fewer messages! 16:36
tflink_ my mail client can open that folder again! 16:36
kparal :) 16:36
tflink_ for some reason, it doesn't handle folders with >10k messages all that well 16:36
adamw sounds like a productive period, thanks kparal 16:37
adamw tflink_: weak! 16:37
kparal very weird email client 16:37
kparal 5. AutoQA logo changed to Marvin the Paranoid Android, yay! 16:37
kparal #link 16:37
kparal It was about time for some optimistic imagery! 16:37
j_dulaney Win 16:37
adamw did we address the, er, legal concerns there? 16:38
kparal it's GPL icon 16:38
* tflink_ is still a little concerned about copyright on that one 16:38
kparal tflink_: shoot 16:38
tflink_ kparal: true but I don't think that means we can use a copyrighted character from a secondary source 16:38
tflink_ I sincerely doubt that the person who uploaded that icon had the rights to assign copyright 16:39
adamw even if the person who drew the image licenses it GPL or whatever, if it resembles the movie Marvin too much, the movie studio would have rights. 16:39
kparal I assumed it was uploaded by the author 16:39
kparal meaning I can't draw _my own_ mickey mouse? 16:39
adamw we couldn't use Mickey Mouse as the logo just because we used a drawing that wasn't actually by Disney - if it looks too much like Mickey, Disney has the rights. 16:39
adamw kparal: ..exactly. 16:39
adamw you can *draw* it 16:40
tflink_ kparal: probably not 16:40
kparal is this the correct universe I live in? 16:40
adamw but using it 'commercially' (which is a very broad stroke) is troublesome. 16:40
kparal alright, I'll ask spot about it 16:40
kparal he's the license guru in Fedora, right? 16:40
j_dulaney Indeed 16:40
adamw he's our fake lawyer, yeah. 16:40
adamw and he knows how to talk to real lawyers :) 16:41
kparal it was such a pretty logo! :/ 16:41
adamw life sucks, huh 16:41
kparal alright, moving on 16:41
kparal 6. Upgradepath displayed wrong results for multi-build updates. The fix is comitted but not deployed yet. 16:41
kparal and that's all the news I know of 16:41
adamw #info kparal proposed using an icon that rather resembles the version of Marvin from the recent H2G2 movie as AutoQA's icon, tflink and adamw raised legal concerns 16:42
adamw #action kparal to follow up with spot on the legal questions around the marvin logo 16:42
adamw kparal: cool, thanks 16:42
adamw tflink_: any news to add? 16:42
tflink_ is remote_install autoqa related? 16:42
kparal nope 16:42
kparal it should have its own section I believe 16:43
tflink_ then nope, nothing to add - I've been spending most of my time on that 16:43
adamw no, but we can talk about it in open discussion 16:43
adamw anyone else have any autoqa news / qs? 16:44
adamw okay then 16:44
adamw #topic open floor 16:44
adamw sounds like tflink would like to update us on remote_install progress? 16:45
tflink_ sure 16:45
adamw #topic open floor - remote_install 16:45
adamw take it away! 16:45
tflink_ I've been working on a demo for the remote installation system proposed last year 16:45
tflink_ #link 16:45
tflink_ the demo system is mostly ready - I've been trying to figure out hosting details for a public demo 16:46
tflink_ #link 16:46
tflink_ #link 16:46
adamw right now it's running behind the RH firewall, as it was easy for us to find resources there. 16:46
tflink_ the current plan is to run a demo using public cloud infra once I write code to interface with the cloud we're going to use 16:47
j_dulaney You might could check with rbergeron 16:47
tflink_ if we go forward with the idea as more of a demo, we'll work out something using fedora infra 16:47
j_dulaney Maybe figure out something using ec2 or the like 16:48
adamw j_dulaney: yup, that's 'public cloud' 16:48
tflink_ j_dulaney: ec2 is too expensive for this 16:48
j_dulaney Ah 16:48
tflink_ their billing and infra setup doesn't lend well to installation testing 16:48
tflink_ ohi is a KVM based public cloud that is ~ 50% cheaper and less hacky 16:49
tflink_ 50% cheaper for this use case, anyways 16:49
tflink_ that and I really didn't want to estimate iops for EBS 16:49
tflink_ :) 16:50
j_dulaney Cheaper sounds better 16:50
tflink_ EC2 billing is complicated :-/ 16:50
robatino will it be possible to do size and checksum tests on the ISOs? (handy for avoiding download, especially for large spins) 16:50
tflink_ robatino: with the remote_install system? not sure, to be honest 16:51
tflink_ I don't know what would be needed for that off the top of my head 16:51
tflink_ but my instinct says that it would be a sub-optimal way of doing things 16:51
tflink_ and that AutoQA would be a better way to do that 16:51
tflink_ but I'll be sending out an announcement when we have timing more solid for the demo 16:52
adamw coolbeans. 16:52
robatino right now i can do checksum tests using my account on secondary01 without downloading, AutoQA plans to just download them last i heard 16:52
tflink_ if you have access to the RH VPN and want to try it out, let me know 16:52
adamw is there anywhere you could do with help, or is it rolling along nicely? 16:53
tflink_ I think that things are going well, I just need to get the cloud interface written 16:53
* tflink_ wishes that there was a python interface for the elastichosts API 16:54
adamw please don't get sidetracked into writing one. :) 16:54
tflink_ a compete one? no. Just as much as I need for the demo 16:54
adamw cool. 16:54
tflink_ I've already spent more time on this than I'd like for a demo 16:55
nirik tflink_: you might ping skvidal... he has some scripts to build/tear down/run things on cloud instances. 16:55
tflink_ nirik: will do, thanks 16:55
* pschindl is finaly here (sry, I stuck on diploma thesis and forgot to join) 16:55
j_dulaney What's up, pschindl 16:55
tflink_ I think that's it for remote_install, though 16:56
j_dulaney tflink_: Maybe, for long term, could we hook this to AutoQA somehow? 16:56
tflink_ j_dulaney: let's cross that bridge if/when we get there 16:56
j_dulaney Righteo 16:56
tflink_ for now, it's just a demo and may not live past the initial trial 16:57
pschindl j_dulaney: deadline of thesis is this week and it's not done 16:57
adamw okay, back to... 16:57
adamw #topic open floor 16:57
adamw do we have any other open floor topics? anything anyone's concerned about? 16:57
* j_dulaney is good 16:58
adamw okey dokey 16:59
adamw setting the fuse for one minute... 17:00
adamw thanks for coming, all! 17:02
jsmith Go QA! 17:02
adamw man, these lousy defective ACME fuses... 17:02
adamw #endmeeting 17:02

Generated by 2.8 by Marius Gedminas - find it at!