14:00:54 #startmeeting fedora-qadevel 14:00:54 Meeting started Mon May 22 14:00:54 2017 UTC. The chair is tflink. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:00:54 Useful Commands: #action #agreed #halp #info #idea #link #topic. 14:00:54 The meeting name has been set to 'fedora-qadevel' 14:00:54 #topic roll call 14:01:27 * mkrizek is here 14:01:57 welcome, welcome 14:04:30 but if it's just going to be the two of us, it seems a bit pointless :) 14:05:08 * kparal is here 14:05:11 * roshi is here 14:06:04 #chair mkrizek roshi kparal 14:06:04 Current chairs: kparal mkrizek roshi tflink 14:06:52 ok, let's get started. I expect this to be a short meeting 14:07:18 #topic Announcements and Information 14:07:25 #info libtaskotron docs have a new theme, and some parts were updated (not deployed yet in production) -- kparal 14:07:43 #info D1195 updated to run "ansible -c local" on a minion :( - mkrizek 14:07:58 mkrizek: :( ? 14:08:21 I am sad about the inner runner :D 14:08:47 I posted a comment about this into T939 14:09:03 * tflink just saw that comment 14:09:47 it's a conversation we can bring up with the ci folks but there's a certain level of sense to why 14:10:21 * kparal is interested in that 14:10:30 yeah, I understand the reason, just as I said, was hoping to get rid of the inner runner :) 14:12:38 * tflink understands, was also hoping to be rid of that 14:13:08 on the bright side, the inner runner will end up being someone elses' problem :) 14:14:22 kparal: is that something you'd like to get into now or in ticket T939? 14:14:40 we can discuss in the ticket. I was just curious what I'm missing 14:15:27 I don't think it's much, reading your comment 14:15:32 * tflink will get to that after the meeting 14:15:54 ok, moving on 14:16:00 #topic Tasking 14:16:10 Is anyone in need of tasks or clarification around tasks? 14:16:52 * tflink suspects not but wanted to ask anyways 14:17:59 not with Go/No-Go this week :( 14:18:50 the little things :-P 14:18:55 moving on 14:19:28 #topic Test Invocation Status 14:19:47 This is as much for keeping stuff written down as anything 14:20:10 #info D1195 is still in progress, will support the new ansible-based standard interface when merged 14:20:52 #info porting existing tasks to new ansible format has yet to start, will hopefully not take long 14:21:08 anything I'm forgetting? 14:21:53 rewriting directives to ansible modules 14:22:10 are the new directives in a branch somewhere? 14:22:18 separate repo 14:22:22 please link 14:23:37 #info process of porting libtaskotron directives to ansible modules is ongoing, some discussion required 14:23:40 #link https://pagure.io/ansible-fedoraqa-modules 14:23:44 thanks for the reminder 14:24:11 does it make sense to have it as part of the taskotron project, or not? 14:24:20 I just notice it's not in the namespace 14:24:31 kparal: eventually, yes. I didn't want to get in the way of the other refactoring 14:24:47 figured it would find a more permanent home once the dust had settled a bit more 14:24:59 ok 14:25:30 the questions that I have around the directives is mostly "how many of the directives really need to be ported?" 14:25:58 the ones I came up with are koji and bodhi as the required ones 14:26:06 resultsdb 14:26:31 mash and yumrepoinfo are being used by depcheck 14:26:32 yeah, that one as well 14:26:48 kparal: sure but do they really need to be directives if they're used by only one check? 14:26:54 directives/modules 14:27:09 that's a good question 14:27:42 other than the three listed above, I'm wondering if they can be replaced by either existing ansible modules or just cli calls 14:27:52 how do we deal with yumrepoinfo then? 14:28:28 put it as a library method in libtaskotron? 14:28:44 can be 14:28:50 method/module whatever 14:28:57 cli tool 14:29:07 is the first thing that comes to mind 14:29:27 or both 14:29:42 * tflink is wondering if we want to expand the number of cli tools that come out of libtaskotron 14:29:49 instead of having directives 14:30:05 yumrepoinfo 14:30:32 taskotron-xunit, taskotron-resultsyaml etc. 14:30:51 as cli entry points, not additional packages 14:30:56 I just replied to roshi how I dislike cli tools instead of api :) 14:31:05 CLI is an API 14:31:15 roshi: I covered that point as well 14:31:19 that's like I dislike plants but like fruit 14:31:21 :p 14:31:31 kparal: I'm open to other suggestions. I'm not thinking of a better idea ATM 14:31:43 sure, we'll think about it 14:31:57 expecting people to only use python is a bit optimistic, I think 14:32:08 fair point 14:32:17 whether we like it or not, CLI/bash is about the most standard interface we can have 14:33:11 * roshi reads 14:33:28 * tflink needs to get to that discussion today :-/ 14:34:04 any other questions/comments on this? 14:34:09 nop 14:34:10 e 14:34:22 not here 14:34:52 this next one should be short 14:34:59 #topic Testcloud and ppc 14:35:06 * roshi will respond in ticket :) 14:35:22 is anyone working on the koji errors around ppc64 and ppc64le for testcloud? 14:35:31 not me 14:35:35 * roshi didn't even know 14:35:48 I noticed them, but wasn't sure what the errors were 14:35:56 roshi: if only someone got around to being a packager ... 14:35:57 libguestfs seems to be building on ppc 14:36:14 lol 14:36:15 yeah... 14:36:21 my read is that there's a problem with libguestfs but hadn't gotten any farther 14:36:21 in my spare cycles... 14:36:37 it could be a transient issue, though if that is indeed building on the ppc arches 14:36:41 tflink: add roshi to qa-tools-sig and he'll receive the emails 14:37:04 * tflink will do that after the meeting 14:37:07 yay, moar emails! 14:37:16 roshi: you're welcome ;) 14:37:19 lol 14:37:30 roshi: just what you wanted on a monday, right? 14:37:38 we just wanted to make your day a bit brighter 14:37:52 with friends like these, who needs spammers? :p 14:37:54 and by brighter, I mean turn the firehose up a little more 14:38:04 see if I can crest 200k emails in a year 14:38:09 cron helps a lot with that 14:39:00 something in the back of my mind says that this was a problem in previous releases 14:39:08 but I don't remember enough details 14:39:19 anyhow, will start a thread on qa-devel 14:39:50 anything else on this? 14:40:03 * roshi has nothing, obviously :p 14:40:10 if not, moving on to open floor. this has already taken longer than I had hoped :) 14:40:57 #topic Open Floor 14:41:05 Anything else that folks wanted to bring up? 14:41:22 nothing here 14:41:48 * tflink assumes not 14:41:52 thanks for coming, everyone 14:41:57 * tflink will send out minutes shortly 14:42:07 and for last week, which I seem to have forgotten about :-/ 14:42:13 #endmeeting