14:02:54 #startmeeting 14:02:54 Meeting started Mon Jun 22 14:02:54 2015 UTC. The chair is tflink. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:02:54 Useful Commands: #action #agreed #halp #info #idea #link #topic. 14:02:54 #meetingname fedoraqa-devel 14:02:54 #topic roll call 14:02:54 The meeting name has been set to 'fedoraqa-devel' 14:03:04 * mkrizek is here 14:03:31 * kparal is here 14:04:00 cool, who wants to go first with the status updates? 14:04:08 I'll go 14:04:15 * kparal creating notes 14:04:46 #topic mkrizek status report 14:05:18 er, tflink could you add us as chairs? 14:05:24 whoops 14:05:29 #chair mkrizek kparal 14:05:30 Current chairs: kparal mkrizek tflink 14:05:36 #topic mkrizek status report 14:05:48 #info sent a new proposal for logging rework 14:05:48 #link https://phab.qadevel.cloud.fedoraproject.org/D382 14:05:48 #link https://phab.qadevel.cloud.fedoraproject.org/T495 14:05:48 #info investigation on branch detection for distgit directive 14:05:48 #link https://lists.fedoraproject.org/pipermail/packaging/2015-June/010693.html 14:05:51 #link https://phab.qadevel.cloud.fedoraproject.org/D383 14:05:53 #info now working towards finishing D382 and D383 14:06:38 that's it from me this week 14:06:40 cool 14:07:12 i thought that D382 was waiting for an updated patch - did I miss something? 14:08:02 yeah it is, I was on PTO on Fri so it's not finished 14:08:44 ok, just making sure I wasn't slacking too much :) 14:09:06 * tflink is reading the packaging list thread about dist tags 14:09:26 any other questions/comments? 14:09:34 not here 14:09:48 * tflink can go next 14:09:53 go ahead 14:09:56 #topic tflink status report 14:10:05 #info helped get testCloud 0.1.0 release out the door 14:10:05 #info trying to fix mariadb setup issues with qadevel.stg 14:10:31 not much from me last week 14:10:35 * mkrizek didn't know we use mariadb 14:10:42 phab and beaker require it 14:10:48 ah ok 14:11:08 * tflink would rather not have 2 database types, though 14:11:59 the summary is short, but there were quite a few patches to testcloud, I was getting a lot of notifications :) 14:12:13 yeah, good progress on that front 14:12:21 now libtaskotron work can continue :) 14:12:28 \o/ 14:12:50 tflink: are you still working on booting just cloud images, or have you look also at booting the virt-builder generated images? 14:13:04 kparal: not sure I understand the difference 14:13:09 i thought they were pretty much the same 14:13:24 the generated images don't need cloud-init, and you don't need to extract kernel + initrd 14:13:52 ah, then no, we haven't looked at that 14:14:13 ok. those images should be just ready to boot and use 14:14:34 just clarifying 14:14:51 ok, I was just assuming that they were pretty much cloud images - thanks for clarifying 14:15:10 we can decide not to use them, but it seemed simpler to me than to fuss around with cloud extra complexities 14:15:35 * tflink adds testing with those images under testcloud to his todo list 14:15:49 great, thanks 14:15:58 testCloud still needs a lot of work but most of it is behind-the-scenes and isn't critical for the moment 14:16:12 * kparal can go next 14:16:24 kparal: there isn't anyone else left :-P 14:16:35 if there are no further questions, I mean :) 14:16:45 fair enough 14:16:57 * tflink assumes no more questions/comments 14:17:10 #topic kparal status report 14:17:14 #info many reviews done 14:17:19 #info still waiting for more feedback for cli/conf/formula changes for disposable clients 14:17:19 #link https://phab.qadevel.cloud.fedoraproject.org/T408 14:17:53 I don't want to start writing patches if we decide to do it in a different way 14:18:04 I have some preliminary feedback from mkrizek and jskladan 14:18:05 whoops - i missed that ticket change 14:18:20 tflink: if you can look at it, that would be great :) 14:18:27 will do 14:18:31 thanks 14:18:43 #info together with jskladan found out that execdb was missing db indices once again 14:18:52 #info discussing distgit directive and its implementation on the packaging mailing list 14:18:52 #link https://lists.fedoraproject.org/pipermail/packaging/2015-June/010693.html 14:18:52 sorry for the delay, not sure how I missed that in my ticket notifications 14:19:08 #info learned from lmacken that there should be an efficient way to query Bodhi for multiple build->update associations in a single request, need to look at it soon 14:19:39 that's it from me 14:19:56 kparal: if you end up using it, fakefedorainfra may need an update as well 14:20:21 good call, I'll check it as well 14:20:22 but that code should be less useful once bodhi2 is up and running 14:20:30 even though all we do is forward requests, I think 14:20:46 yeah but IIRC, it has handlers for the specific requests 14:21:01 if the json bits aren't implemented, fakefedorainfra may just come back with a 404 or 5XX 14:21:18 tflink: taskotron wasn't responsible for the infra overload last week, right? I just skimmed the logs 14:21:44 I don't think so but I haven't read through IRC logs and email to see if they found a root cause 14:22:02 we were affected by it, though - taskotron production wasn't accessible for a while 14:22:15 * tflink has filed a ticket to get that fixed in the future 14:22:25 * kparal noticed it 14:22:26 as usuall, fun with firewalls :) 14:22:59 it seems to be working now, though 14:23:16 which makes me suspect that the issue is at least not so much of a problem ATM 14:24:03 any other questions/comments? 14:24:14 nothing here 14:24:24 * tflink has one quick topic if you guys don't mind 14:24:31 go ahead 14:24:39 #topic taskotron build/deployment status 14:24:44 https://phab.qadevel.cloud.fedoraproject.org/T505 14:25:06 * tflink wishes he had pestered jskladan about this before he left last week 14:25:46 I think it is just a standard upgrade procedure, nothing special required, to get the artifacts going 14:25:46 are either of you aware of anything else that's missing for deployment? 14:26:03 yeah, just wanted to make sure we weren't waiting on any other code 14:26:12 we will of course need to check very well that the new bodhi comments are correct 14:26:35 if it's ready, I want to get it into dev today or tomorrow 14:26:41 it's not yet ready 14:26:53 two more tickets in https://phab.qadevel.cloud.fedoraproject.org/T399 14:27:25 one of them has approved diff, so that can be committed 14:27:31 but we still need to adjust task formulas 14:27:43 I think I could do it right away, if we want to have it asap 14:28:12 i just don't want to get too far behind so we're not deploying in huge chunks 14:28:21 wow, another thing I missed last week 14:28:46 * tflink is full of fail for acting on or noticing notifications last week 14:28:49 also more eyes on the diff (D395) can help as well :) 14:29:39 IIRC, there are some other unrelated revisions waiting for merge in resultsdb/resultsdb_frontend 14:29:46 tflink: if you want to stopped being subscribed on every single change in Phab and just watch something selectively (to avoid notification overload), there are some pointers at https://fedoraproject.org/wiki/QA/Phabricator 14:30:20 kparal: or actually spend the time to fix my email filtering :) 14:30:30 whatever works for you :) 14:31:28 ok, I just wanted to make sure that I wasn't missing anything 14:31:40 it appears I was but that should be fixed now 14:31:49 which brings us to ... 14:31:53 #topic open floor 14:32:09 anything else to cover? 14:32:27 I don't have anything 14:32:38 nothing here 14:32:41 k 14:32:46 * tflink will send out minutes shortly 14:32:49 thanks for coming 14:32:53 thanks 14:32:53 #endmeeting