| *** saneax is now known as saneax-_-|AFK | 00:00 | |
| *** saneax-_-|AFK is now known as saneax | 00:16 | |
| SpamapS | jeblair: so.. the thing is, none of these changes are "dependent on one another", so all 3 merge builds do not contain the others. | 00:37 |
|---|---|---|
| SpamapS | jeblair: we could test that history has all 3 in it once everything is released I suppose. | 00:38 |
| * SpamapS isn't sure why he used the scare quotes there :-P | 00:38 | |
| *** saneax is now known as saneax-_-|AFK | 00:46 | |
| *** jamielennox is now known as jamielennox|away | 00:50 | |
| *** Cibo_ has joined #zuul | 01:00 | |
| *** jamielennox|away is now known as jamielennox | 01:21 | |
| timrchavez_ | jeblair: Wish this was pure Python :) https://github.com/yaronn/blessed-contrib. Could add https://github.com/lovehandle/blessed-graphite and have a decent replacement for grafana :-) | 01:57 |
| timrchavez_ | Maybe it's time to start chartty? | 01:57 |
| *** Shuo has quit IRC | 02:05 | |
| mordred | chartty++ | 03:40 |
| *** saneax-_-|AFK is now known as saneax | 05:01 | |
| *** Cibo_ has quit IRC | 05:08 | |
| *** bhavik1 has joined #zuul | 06:36 | |
| *** bhavik1 has quit IRC | 06:47 | |
| *** hashar has joined #zuul | 08:36 | |
| *** saneax is now known as saneax-_-|AFK | 10:55 | |
| *** saneax-_-|AFK is now known as saneax | 11:58 | |
| *** bhavik1 has joined #zuul | 12:45 | |
| *** bhavik1 has quit IRC | 13:17 | |
| *** yolanda has quit IRC | 13:18 | |
| *** yolanda has joined #zuul | 13:18 | |
| *** timrchavez_ is now known as timrc_ | 13:55 | |
| *** saneax is now known as saneax-_-|AFK | 14:16 | |
| *** gundalow is now known as docbot | 15:42 | |
| *** docbot is now known as searchy | 15:43 | |
| *** searchy is now known as answerible | 15:43 | |
| *** answerible is now known as answerable | 15:43 | |
| *** answerable is now known as cowsay | 15:44 | |
| *** cowsay is now known as OptimusBot | 15:44 | |
| *** OptimusBot is now known as gundalow | 15:44 | |
| *** gundalow is now known as answerbot | 15:49 | |
| *** answerbot is now known as gundalow | 15:50 | |
| *** rbergeron has quit IRC | 16:14 | |
| *** dag_ has joined #zuul | 16:20 | |
| *** dag_ is now known as rbergeron | 16:21 | |
| *** rbergeron has quit IRC | 16:21 | |
| *** rbergeron has joined #zuul | 16:21 | |
| *** Cibo_ has joined #zuul | 16:38 | |
| *** Cibo_ has quit IRC | 16:43 | |
| jeblair | SpamapS: yeah, it's (incidentally) testing sequencing. they should share a change queue, and therefore, in the repo state produced by the launcher-merger, builds for C should contain A and B. if they don't, then it's probably because we need to update the test configuration to manually specify that they share a change queue. | 16:48 |
| jeblair | SpamapS: (we automatically created shared change queues based on job names in v2, but need to manually specify them in v3) | 16:48 |
| jeblair | timrc, timrc_: chartty :) | 16:48 |
| *** hashar has quit IRC | 16:51 | |
| openstackgerrit | Paul Belanger proposed openstack-infra/nodepool: DNM: prepare for fedora-25 https://review.openstack.org/419557 | 16:52 |
| *** rmoe has quit IRC | 17:08 | |
| *** rmoe has joined #zuul | 17:09 | |
| SpamapS | jeblair: hm. I feel like every time we talk I get schooled on something I don't understand about the data model. | 17:25 |
| SpamapS | jeblair: I thought gate automatically implied a shared change queue? | 17:26 |
| jeblair | SpamapS: this is where the terms 'pipeline' and 'queue' take on real importance -- the gate pipeline ('gate' isn't the special thing here, but the fact that we tell it to use the 'dependent' pipeline manager is) observes and automatically creates (based on the order equeued) dependencies between changes, but only within a shared change queue (that's model.ChangeQueue). | 17:29 |
| jeblair | SpamapS: in zuulv2, any projects which share a job get combined into a shared change queue. (eg. "gate-tempest-dsvm" running on both nova and tempest cause both projects to share a change queue in the gate pipeline) | 17:30 |
| jeblair | SpamapS: in v3, since we're dropping unique job names per-project (in v2 we have gate-nova-python27, gate-zuul-python27 -- they are different, so no shared change queue; in v3, they will both run "python27") so we have to manually say "nova and tempest share a change queue" | 17:31 |
| jeblair | that's done with the "queue:" attribute of the project-pipeline configuration; git grep "queue: integrated" for examples | 17:32 |
| SpamapS | jeblair: ok, the bulb is getting les dim in my head. ;) | 17:34 |
| SpamapS | less | 17:34 |
| jeblair | fiat lux! | 17:34 |
| SpamapS | jeblair: thanks, I think that is, in fact, it. | 17:34 |
| SpamapS | les dim bulb, c'est bon. | 17:35 |
| *** pabelanger has quit IRC | 17:41 | |
| *** pabelanger has joined #zuul | 17:41 | |
| *** Shuo has joined #zuul | 17:51 | |
| SpamapS | jeblair: funny enough, a few other tests seem to have been re-worked to deal with the lack of ordering for org/project's gate | 17:54 |
| jeblair | SpamapS: hrm :( i think it might be good to correct this to match the old behavior (and if we want to change it, do that separately later, possibly while inventorying which tests cover things like shared change queue ordering) | 18:01 |
| SpamapS | jeblair: well, org/project2 is actually just org/project with the integrated gate | 18:03 |
| SpamapS | so changing the test to project2 works | 18:03 |
| *** Shuo_ has joined #zuul | 18:10 | |
| *** bhavik1 has joined #zuul | 18:22 | |
| *** bhavik1 has quit IRC | 18:26 | |
| SpamapS | hm.. there may be a race or some kind of system dependence in test_success_url | 18:31 |
| SpamapS | it just failed for me.. then passed on retry | 18:31 |
| *** openstackgerrit has quit IRC | 18:33 | |
| *** openstackgerrit has joined #zuul | 18:59 | |
| openstackgerrit | Clint 'SpamapS' Byrum proposed openstack-infra/zuul: Re-enable test_build_configuration_branch https://review.openstack.org/419609 | 18:59 |
| SpamapS | jeblair: ^ the final product. | 18:59 |
| *** pabelanger has quit IRC | 19:39 | |
| *** pabelanger has joined #zuul | 20:02 | |
| pabelanger | jeblair: mind reviewing 412160 again? allows nodepool-builder to only build diskimages, if configured for that | 21:10 |
| *** openstackgerrit has quit IRC | 21:18 | |
| *** Shuo has quit IRC | 22:18 | |
| *** Shuo_ is now known as Shuo | 22:18 | |
| SpamapS | adam_g: argh! I duplicated your work | 23:01 |
| *** openstackgerrit has joined #zuul | 23:02 | |
| openstackgerrit | Jesse Keating proposed openstack-infra/zuul: Add trigger capability on github pr review https://review.openstack.org/419684 | 23:02 |
| SpamapS | adam_g: https://review.openstack.org/409376 revamps test_build_configuration_branch but you never claimed it :-/ | 23:02 |
| * SpamapS abandons https://review.openstack.org/419609 | 23:03 | |
| adam_g | SpamapS: i thought i had a task for it in sb, one sec | 23:03 |
| adam_g | SpamapS: yea, #3443 | 23:05 |
| SpamapS | nope | 23:05 |
| SpamapS | Clint Byrum created the task "test_build_configuration_branch" on 2017-01-10 at 19:56:41 | 23:05 |
| SpamapS | adam_g: 3443 is only for the first one | 23:05 |
| SpamapS | err no | 23:05 |
| SpamapS | adam_g: sorry you confused me | 23:06 |
| SpamapS | adam_g: I just reassigned 3443 to you | 23:06 |
| SpamapS | adam_g: the one you have is 3424, which is just for _conflict | 23:06 |
| SpamapS | adam_g: anyway, it's fine to touch 3 in one task, but if the name of the test isn't in the story's task list.. this happens. :-P | 23:07 |
| adam_g | SpamapS: ohh i see. my baddy. | 23:07 |
| SpamapS | it's ok I probably learned something | 23:07 |
| SpamapS | adam_g: I made a 3445 for _multi_branch too | 23:07 |
| adam_g | SpamapS: if you have a simple fix to get that one going, tho, maybe we can merge that instead? digging into that group of tests kinda got me stuck in the mud with other things | 23:08 |
| SpamapS | adam_g: see 419609 ... it does pass here locally. Not sure why it didn't pass on the check test | 23:12 |
| *** saneax-_-|AFK is now known as saneax | 23:19 | |
| adam_g | SpamapS: sure gimme a few. need to kick off an os install on desktop after a hard drive blew up | 23:23 |
| adam_g | SpamapS: so on my version of the test i was looking at the repos in the each builds JobDir | 23:50 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!