15:01:02 <tflink> #startmeeting fedora-qadevel 15:01:02 <zodbot> Meeting started Mon Mar 7 15:01:02 2016 UTC. The chair is tflink. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:01:02 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic. 15:01:02 <zodbot> The meeting name has been set to 'fedora-qadevel' 15:01:02 <tflink> #meetingname fedora-qadevel 15:01:03 <tflink> #topic Roll Call 15:01:03 <zodbot> The meeting name has been set to 'fedora-qadevel' 15:01:14 * mkrizek 15:01:15 * kparal is here 15:01:33 * garretraziel is here 15:02:34 * jskladan is here 15:03:11 <tflink> welcome, welcome 15:03:19 <tflink> shall we get started? 15:03:29 <tflink> #topic Announcements and Information 15:03:51 <tflink> #info new planning project that will have our short term tasks going forward - tflink 15:03:51 <tflink> #info The roadmap has been reworked, easier to read now - tflink 15:03:51 <tflink> #info Implement result-naming whitelist checking in libtaskotron - mkrizek 15:03:51 <tflink> #link https://phab.qadevel.cloud.fedoraproject.org/T715 15:03:52 <tflink> #info Determine Top Level Namespaces - mkrizek 15:03:54 <tflink> #link https://phab.qadevel.cloud.fedoraproject.org/T717 15:03:57 <tflink> #info we will need to wait with F24 yumrepoinfo.conf changes until Alpha freeze, hopefully things will be better next time - kparal 15:03:59 <tflink> #link https://phab.qadevel.cloud.fedoraproject.org/T729 15:04:01 <tflink> #link https://github.com/fedora-infra/bodhi/issues/794 15:04:03 <tflink> #info working task-dockerautotest - suite of tests for Docker to be run with libtaskotron - jsedlak 15:04:05 <tflink> #link https://phab.qadevel.cloud.fedoraproject.org/D767 15:04:10 <tflink> any additional items, questions or comments? 15:04:56 <mkrizek> not here 15:05:05 * tflink assumes not 15:05:07 <tflink> moving on 15:05:16 <kparal> nice roadmap document 15:05:19 <tflink> #topic Dist-Git style tasks 15:05:19 <tflink> #link https://phab.qadevel.cloud.fedoraproject.org/T704 15:05:32 <tflink> kparal: thanks, now it needs to stay up to date :) 15:05:46 <kparal> tflink: you mean it doesn't auto-update itself? 15:05:58 <tflink> kparal: don't I wish 15:06:43 <tflink> anyhow, more thoughts on this particular topic? there were a few comments in the last week but I didn't see any objections 15:07:21 <kparal> I still plan to review D759, it's related 15:07:23 <tflink> assuming I'm not forgetting a blocking bit of this task again 15:08:15 <kparal> so, as a follow up of the discussion in T704 - I really thought we're trying to allow a formula-less tasks 15:08:30 <kparal> for the specific use case of pkg-specific checks 15:08:38 <tflink> can you elaborate a bit more on how that might work? 15:08:58 <tflink> i remember talking about being able to skip the scheduling files by relying on defaults 15:09:14 <tflink> but I don't recall a discussion around the forumla files themselves 15:09:30 <kparal> I haven't really thought it out well, I just assumed we want to attempt it based on our previous discussions, but it's very possible that I just understood it wrong 15:09:56 <kparal> I think it could be possible, but I'm not sure if it's worth the trouble to have a special case where formulas are not needed 15:10:01 <jskladan> tflink: iirc, we said that we'll allow a know-named executable, that will just return either exit code, or some standard output 15:10:12 <tflink> would the idea be that an executable script of some sort in the dir and if no formula file exists, it would be run against koji_build, have that build downloaded and the script executed, reporting the results to resultsdb? 15:10:13 <jskladan> and we'll devise all the needed information from the"metadata" 15:10:23 <kparal> something like that 15:10:25 <jskladan> tflink: that 15:11:20 <tflink> if we can figure out something reasonable, that sounds like it could be useful 15:12:16 <tflink> but I don't want to be a requirement for dist-git tasks until we have a better idea of how it would work and how much effort it would require 15:13:03 <kparal> let's go forward with formula-based approach for now, I think 15:13:08 <kparal> less complications 15:13:40 <kparal> we can always introduce some automatic mode later 15:13:54 <jskladan> tflink: my understanding was, that we will want to have that before the "mass roll-out", so the test-devs have the least complicated first step 15:14:09 <jskladan> but what kamil said sounds reasonable 15:14:17 <jskladan> we can add the icing on the cake later 15:14:20 <tflink> jskladan: I'd like to, sure but I don't want to take too long 15:14:21 <tflink> exactly 15:15:39 <tflink> #action tflink to write up ticket for formula-less dist-git tasks 15:15:55 <tflink> anything else on this topic? any reason not to start moving forward on it? 15:16:38 <kparal> nope 15:18:17 * tflink takes silence from others as a no, too 15:18:22 <tflink> ok, moving on 15:18:27 <tflink> #topic Tasking 15:18:37 <tflink> is anyone in need of a task? 15:19:11 <jskladan> I'm good 15:19:22 * kparal still has reviews waiting 15:19:41 <mkrizek> I'm good too 15:20:41 <tflink> that's pretty much all that I had which brings us to ... 15:20:45 <tflink> #topic Open Floor 15:21:09 * tflink is going to be doing more ticket mongering this week in an attempt to make the planning board cleaner 15:21:17 <tflink> useful suggestions would be welcome 15:23:36 <tflink> anything else? 15:23:46 <tflink> if not, will close out the meeting in a minute or so 15:24:18 <jskladan> nothing here 15:24:21 <jskladan> as far as I can tell 15:25:41 <tflink> alrighty, thanks for coming everyone 15:25:46 * tflink will send out minutes shortly 15:25:51 <tflink> #endmeeting