16:30:51 <djmitche> #startmeeting weekly
16:30:51 <bb-supy> Meeting started Tue Jul 26 16:30:51 2016 UTC and is due to finish in 60 minutes.  The chair is djmitche. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:30:51 <bb-supy> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
16:30:51 <bb-supy> The meeting name has been set to 'weekly'
16:30:57 <djmitche> #topic Introduction
16:32:01 <djmitche> titanpad is being slow...
16:32:22 <djmitche> #link https://titanpad.com/ep/pad/view/buildbot-agenda/U5dPsJ0lYs
16:32:34 <djmitche> who's around? tardyp?
16:32:44 <djmitche> did you have a good break?
16:33:04 <tardyp> hey!
16:33:10 <tardyp> yes good
16:33:23 <djmitche> cool :)
16:33:45 <djmitche> #topic Week in Review
16:33:53 <djmitche> looks like this was a light week?
16:33:59 <djmitche> #info Some work on Hyper.sh integration
16:34:11 <djmitche> https://github.com/buildbot/buildbot/pull/2318
16:34:36 <djmitche> #info Amar updated the Trac configuration to hopefully reject fewer logins for non-spammers
16:34:44 <djmitche> anything else?
16:35:18 <tardyp> there is another PR that wont show in the report
16:36:09 <tardyp> https://github.com/buildbot/buildbot/pull/2315
16:36:24 <tardyp> improvements to the trigger steps
16:37:30 <djmitche> #info landed some improvements to trigger steps around unimportant schedulers
16:37:34 <djmitche> awesome
16:37:53 <djmitche> #info Release Updates
16:38:03 <djmitche> I think we're still waiting for feedback on rc1, right?
16:38:09 <tardyp> I created a bug fix for rc1
16:38:24 <tardyp> which I discovered myself, and is about buildrequest cancellation
16:38:42 <tardyp> I think this is a good reason for an rc2
16:38:56 <tardyp> I will create that once it has landed
16:39:30 <djmitche> ok
16:39:49 <djmitche> #info tardyp discovered a bug regarding buildrequest cancellation in rc1; an rc2 will be out once that fix is landed
16:40:12 <djmitche> #topic Hyper.sh integration
16:40:14 <djmitche> and updates on this?
16:40:29 <djmitche> looks like it's just you and me.. we should have found a nice restaurant to have the meeting in ;)
16:40:35 <tardyp> I've been playing this beg of week with latent, and I can see lots of issues regarding to  worker lifecycle
16:40:46 <tardyp> I'm gathering a bunch of fixes
16:41:27 <tardyp> there are issues like deadlock when workers wont attach for some reason (e.g ngrok not setup)
16:42:07 <tardyp> also issues when reconfiguring the master and dropping some workers, then the build will finish too soon
16:42:48 <tardyp> and also the fact that when master is stopped, the builds are finished as a side effect of worker detachment which is not very good
16:43:18 <tardyp> so I am adding quick gracefullShutdown to the master stop procedure
16:43:49 <tardyp> I'll try to split all those into few PRs
16:43:50 <djmitche> sounds good
16:44:01 <djmitche> sounds closely related to the bounty bug
16:44:48 <tardyp> I do think so as well. The thing is we need to have solid latent support if we want to switch metabuildbot to hyper for testing PRs
16:45:03 <tardyp> which is the major purpose of all this hyper work
16:45:24 <tardyp> travis is really too slow for our workload
16:45:32 <djmitche> yeah
16:45:56 <djmitche> botherders are specifically allowed to get bounties, if you'd like
16:46:36 <tardyp> yep
16:47:12 <djmitche> ok
16:47:16 <tardyp> if you have a chance to review #2325 and #2324 that will help
16:47:19 <djmitche> #topic Infrastructure improvements
16:47:19 <djmitche> ok
16:48:17 <djmitche> #info a little bit of conversation about infrastructure this week, and things seem to be back up and running
16:48:33 <djmitche> #info looking at ways to test deployments with ansible -- perhaps via EC2 or some other virtualization tech
16:48:42 <djmitche> #topic OpenStack v3 support
16:48:49 <djmitche> cmouse: did you get a chance to file a bug?
16:49:48 <skelly> I don't see anything
16:50:00 <djmitche> me neither
16:50:07 <djmitche> ok, any other business?
16:50:11 <djmitche> #info no updates this week
16:51:56 <djmitche> #endmeeting