| *** timrc has quit IRC | 00:32 | |
| *** jlk has quit IRC | 00:35 | |
| *** jamielennox is now known as jamielennox|away | 00:41 | |
| *** jamielennox|away is now known as jamielennox | 00:58 | |
| *** rattboi has quit IRC | 01:12 | |
| *** jesusaur has quit IRC | 01:12 | |
| *** tobiash_ has quit IRC | 01:12 | |
| *** herlo has quit IRC | 01:12 | |
| *** patrickeast has quit IRC | 01:12 | |
| *** mgagne has quit IRC | 01:12 | |
| *** saneax-_-|AFK has quit IRC | 01:12 | |
| *** mordred has quit IRC | 01:12 | |
| *** jhesketh has quit IRC | 01:12 | |
| *** olaph has quit IRC | 01:12 | |
| *** herlo has joined #zuul | 01:12 | |
| *** herlo has quit IRC | 01:12 | |
| *** herlo has joined #zuul | 01:12 | |
| *** olaph has joined #zuul | 01:13 | |
| *** mordred has joined #zuul | 01:13 | |
| *** jhesketh has joined #zuul | 01:13 | |
| *** rattboi has joined #zuul | 01:13 | |
| *** Guest96631 has joined #zuul | 01:13 | |
| *** saneax-_-|AFK has joined #zuul | 01:13 | |
| *** tobiash has joined #zuul | 01:13 | |
| *** jesusaur has joined #zuul | 01:14 | |
| *** patrickeast has joined #zuul | 01:17 | |
| *** timrc has joined #zuul | 01:19 | |
| *** timrc_ has quit IRC | 01:23 | |
| *** morgan has quit IRC | 01:23 | |
| *** auggy has quit IRC | 01:23 | |
| *** mattclay has quit IRC | 01:23 | |
| *** timrchavez_ has joined #zuul | 01:24 | |
| *** jlk has joined #zuul | 01:26 | |
| *** jlk has quit IRC | 01:26 | |
| *** jlk has joined #zuul | 01:26 | |
| *** auggy has joined #zuul | 01:28 | |
| *** mattclay has joined #zuul | 01:29 | |
| *** morgan has joined #zuul | 01:36 | |
| *** adam_g has quit IRC | 02:44 | |
| *** adam_g has joined #zuul | 02:45 | |
| *** bhavik1 has joined #zuul | 05:01 | |
| *** bhavik1 has quit IRC | 05:03 | |
| *** saneax-_-|AFK is now known as saneax | 05:06 | |
| *** Cibo_ has quit IRC | 06:14 | |
| *** bhavik1 has joined #zuul | 07:04 | |
| *** hashar has joined #zuul | 07:43 | |
| *** bhavik1 has quit IRC | 08:03 | |
| *** openstackgerrit has quit IRC | 08:18 | |
| *** mptacekx has joined #zuul | 08:25 | |
| mptacekx | Hi, I have one general zuul question. Can someone please reply on following ? Should projects section of layout.yaml contains all openstack projects (3,5k+) or just the projects we would like to listen to and potential dependencies of some commits should be merged by zuul-merger automatically? | 08:30 |
|---|---|---|
| *** mattclay has quit IRC | 09:37 | |
| *** mattclay has joined #zuul | 09:38 | |
| *** saneax is now known as saneax-_-|AFK | 10:18 | |
| *** saneax-_-|AFK is now known as saneax | 11:12 | |
| mptacekx | Hi, can someone please give me some estimation when zuul v3 is ready to be taken by 3rd party CI's ? | 11:52 |
| *** wznoinsk has joined #zuul | 12:00 | |
| *** hashar is now known as hasharLunch | 12:12 | |
| *** saneax is now known as saneax-_-|AFK | 12:41 | |
| *** saneax-_-|AFK is now known as saneax | 12:48 | |
| *** saneax is now known as saneax-_-|AFK | 13:11 | |
| *** saneax-_-|AFK is now known as saneax | 13:15 | |
| *** jamielennox is now known as jamielennox|away | 13:39 | |
| *** saneax is now known as saneax-_-|AFK | 14:00 | |
| *** mptacekx has quit IRC | 15:14 | |
| *** hasharLunch is now known as hashar | 15:48 | |
| *** hashar is now known as hasharAway | 17:32 | |
| jeblair | mordred: what's the status of 410801and 410802? | 17:41 |
| jeblair | mordred: see comment on 297950 | 17:42 |
| * SpamapS knocks off the zuul dust | 17:42 | |
| *** toabctl has quit IRC | 17:42 | |
| SpamapS | jeblair: the automated board looks positively achievable btw.. seems we've either been chipping away nicely, or untagging things. ;) | 17:42 |
| jeblair | SpamapS: \o/ | 17:43 |
| jeblair | clarkb: istr some discussion about 410939; are you okay with that? | 17:44 |
| jeblair | i'm going to self-approve 412571 | 17:45 |
| clarkb | re 410939 I do still think we should avoid change for changes sake. Yes, we are making a lot of changes but the fewer that are user visible the easier ti will be for them to transition to the new stuff | 17:46 |
| clarkb | but I am claerly outnumbered on that so... | 17:46 |
| jeblair | clarkb: i agree with your principle there. however, i think this is a good change to make: we *almost never* want to stop both of those daemons at the same time, so i feel bad telling users that's an okay configuration when it might back them into a corner. | 17:49 |
| *** openstackgerrit has joined #zuul | 17:49 | |
| openstackgerrit | Merged openstack-infra/nodepool: Fix some doc typos https://review.openstack.org/412571 | 17:49 |
| mordred | jeblair: I believe they are still good - I totally thought they'd landed already | 17:49 |
| jeblair | (i feel the same way about zuul and geard) | 17:49 |
| clarkb | jeblair: if that is the argument I am not sure it was ever expressed. Just that it is legacy code and we want to chagne it | 17:49 |
| jeblair | mordred: oh, hrm, were they duplicated on a branch by any chance? (though if so, i'd expect them to be in merge conflict now) | 17:50 |
| clarkb | jeblair: and it will require everyone to update their daemon invocations (either to remove the invalid --no-builder flag, or to add a new daemon invocation for the builder separately) | 17:50 |
| openstackgerrit | David Shrewsbury proposed openstack-infra/nodepool: Create per-provider ZK work threads https://review.openstack.org/417994 | 17:50 |
| jeblair | clarkb: you know my sysadmin preferences -- i hate change :) | 17:50 |
| Shrews | jeblair: If you get a chance, would like to see if you think 994 is on the right path for what you mentioned last week | 17:51 |
| *** Guest96631 is now known as mgagne | 17:51 | |
| *** mgagne has quit IRC | 17:51 | |
| *** mgagne has joined #zuul | 17:51 | |
| clarkb | jeblair: pabelanger maybe we can write down what the intent is there so that if we get angry users they can see what the justifcation was instead of just changing what was perceived to be bad legacy code? | 17:51 |
| jeblair | clarkb: maybe we should deprecate --no-builder to make it just print a warning for a bit? | 17:52 |
| clarkb | jeblair: that is probably a good idea too (on the master side ya?) | 17:52 |
| jeblair | clarkb: yep | 17:52 |
| mordred | jeblair: I submitted them to both branches | 17:53 |
| mordred | jeblair: they have landed on neither | 17:53 |
| jeblair | mordred: ah, i see the master ones are better reviewed | 17:54 |
| pabelanger | jeblair: clarkb: +1 for deprecate --no-builder, removal at a later date | 17:54 |
| jeblair | pabelanger: that wasn't exactly what i was suggesting :) | 17:54 |
| jeblair | pabelanger: i say remove it, but leave the option working so it doesn't blow up. | 17:54 |
| clarkb | I think jeblair is saying put the warning in now on the master side and remove it on the v3 side now | 17:54 |
| clarkb | then when v3 merges into master later people will have had warning of the change | 17:55 |
| mordred | no-op command line args are cheap | 17:55 |
| clarkb | oh gotcha ya that works too | 17:55 |
| jeblair | yep. i don't think we need to go through a full deprecation cycle for this | 17:55 |
| pabelanger | okay, no-op is fine too | 17:55 |
| pabelanger | I'll update my patch | 17:55 |
| clarkb | but also would be good to add the justification to the commit message | 17:55 |
| jeblair | mordred: should we +3 those now then? | 17:55 |
| mordred | jeblair: sure - unless you have concern about the nodepool-dsvm job being broken at the moment | 17:56 |
| clarkb | "Nodepool is now a properly distributed system and we do not want users to couple the starting/stopping of individual daemons when this is not required. As a result we have explicitly uncoupled the daemons" | 17:56 |
| clarkb | or similar | 17:56 |
| pabelanger | clarkb: Oh, BTW: mind looking at https://review.openstack.org/#/c/414662/. Was the only way I could get glean to actually work, I needed to use a tarball | 17:56 |
| pabelanger | clarkb: ++ | 17:56 |
| clarkb | pabelanger: re glean, I had it working just fine with git locally | 17:57 |
| clarkb | pabelanger: let me boot my VM and check what flags I needed | 17:57 |
| jeblair | mordred: i do have concerns with those words, but i don't know what they mean :( are you saying if we land that, the job will be broken, or...? | 17:57 |
| jeblair | mordred: (cause it looks green to me?) | 17:57 |
| mordred | jeblair: the job is currently broken | 17:57 |
| mordred | it broke sometime after jan 4 | 17:57 |
| jeblair | mordred: oh, not related to this change | 17:57 |
| mordred | yah | 17:57 |
| pabelanger | clarkb: issues I had with that, I could never get the depend-on version of glean to actually checkout inside DIB. I couldn't get the right ref | 17:58 |
| jeblair | mordred: so put another way: "hey, maybe we should make that job work before we land anything to nodepool" ? | 17:58 |
| clarkb | pabelanger: its teh same ref as in the calling environment | 17:58 |
| mordred | jeblair: well - yes. sc86cal is currently logged in to a held node investigating | 17:58 |
| clarkb | pabelanger: however it is possible we need to make it a proper branch/tag due to some dib thing | 17:59 |
| mordred | jeblair: the problem is that in the nodepool-dsvm and the shade-functional-neutron devstacks, nova is not getting ips back from neutron for servers | 17:59 |
| pabelanger | clarkb: Ya, that was actually the fix. I had to create a local branch, then use that branch for DIB | 17:59 |
| pabelanger | clarkb: so, I can do that, or the tarball | 17:59 |
| mordred | jeblair: it is exceptionally unclear as to why it isn't - but it's not related to shade (the same break happens with python-openstackclient too) | 17:59 |
| jeblair | mordred: okay. yeah, i would be annoyed with myself if i +3d these and we ended up with 2 problems somehow, even though these should be fine. so i'll hold off for now. | 18:00 |
| mordred | ++ | 18:00 |
| pabelanger | clarkb: https://review.openstack.org/#/c/414662/7/devstack/plugin.sh worked I believe | 18:01 |
| pabelanger | clarkb: using branch ttt | 18:01 |
| clarkb | pabelanger: source-repository-simple-init:glean git /tmp/glean.git https://git.openstack.org/openstack-infra/glean is how the repo gets installed | 18:02 |
| clarkb | pabelanger: iirc source-repository element only preserves branches and tags so that would likely explain it | 18:02 |
| jeblair | Shrews: +2 with note | 18:03 |
| Shrews | jeblair: cool. i thought that was a very weird option anyway | 18:03 |
| clarkb | pabelanger: I think my only concern with the tarball is that its not really how I would expect someone locally to do things (so we might run across corner case issues), but if it works its not that different than using a branch | 18:04 |
| clarkb | pabelanger: and with a tarball the benefit is that you don't add any new state to the repository (which itself could be side effecty) | 18:05 |
| jeblair | Shrews: yeah, totally insane. but we were backed up against the edge of a snake pit with a bunch of angry wet cats chasing us. or at least that's how i remember it. | 18:05 |
| pabelanger | clarkb: I can switch patch to git, are you find with a hardcode branch? or do you want something more dynamic? | 18:05 |
| clarkb | pabelanger: I think we can just use tarball until we have problems with it | 18:05 |
| pabelanger | clarkb: okay | 18:05 |
| pabelanger | clarkb: well, we already have an issue with it. There is a bug in diskimage-builder, so we actually need a new release for it to work | 18:06 |
| pabelanger | so... | 18:06 |
| clarkb | oh in that case lets use git :) because my theoretical tar isn't common and may be prone to issues is maybe less theoretical now | 18:07 |
| clarkb | pabelanger: I would make a branch called nodepool-devstack-plugin or similar and set that as the ref | 18:08 |
| pabelanger | ++ | 18:09 |
| openstackgerrit | David Shrewsbury proposed openstack-infra/nodepool: Remove --no-launches nodepoold option https://review.openstack.org/418001 | 18:09 |
| Shrews | pabelanger: re 410939, you want to submit for features/zuulv3 branch too? or want me to go ahead and add it to my stack? | 18:22 |
| Shrews | i'm in that code now, so i can do it if you want | 18:22 |
| pabelanger | Shrews: sure, I haven't started yet | 18:22 |
| Shrews | cool. it simplifies my testing a bit by not having to type that extra option :) | 18:26 |
| openstackgerrit | David Shrewsbury proposed openstack-infra/nodepool: Remove --no-builder nodepoold option https://review.openstack.org/418013 | 18:28 |
| *** harlowja_ has joined #zuul | 18:31 | |
| *** harlowja has quit IRC | 18:31 | |
| Shrews | I think the -s option is next on the chopping block | 18:32 |
| jeblair | Shrews: I hope so, though i'm worried we may end up needing something like that for authenticated zk access? | 18:34 |
| Shrews | jeblair: hrm, good point | 18:35 |
| openstackgerrit | Paul Belanger proposed openstack-infra/nodepool: Remove the ability for nodepoold to launch a builder https://review.openstack.org/410939 | 18:39 |
| openstackgerrit | Merged openstack-infra/zuul: Fix test_rerun_on_abort https://review.openstack.org/411426 | 18:40 |
| harlowja_ | mordred any knowledge on https://www.ansible.com/open-tower yet :-P | 18:53 |
| harlowja_ | is that happenening yet, lol | 18:53 |
| mordred | rbergeron: ^^ I think that question was really for you ;) | 18:54 |
| jlk | "soon" | 18:54 |
| mordred | harlowja_: so - _yes_ it is definitely happening. but no I don't have any info on timeframe | 18:55 |
| harlowja_ | :-P | 18:55 |
| harlowja_ | be real nice | 18:55 |
| harlowja_ | i think | 18:55 |
| harlowja_ | especially with stuff like kolla-ansible | 18:56 |
| mordred | yah | 19:00 |
| mordred | I'm looking forward to it | 19:00 |
| rbergeron | i know nothing about kolla-ansible! | 19:03 |
| rbergeron | oh, you'er asking about open-tower. | 19:03 |
| rbergeron | WE ARE WORKING ON IT and i yell about it on a weekly basis at the bare minimum. | 19:03 |
| harlowja_ | rbergeron ya | 19:03 |
| harlowja_ | is that like maybe this month? | 19:03 |
| harlowja_ | :-P | 19:03 |
| rbergeron | harlowja_: i cannot confirm or deny such things but i would not recommend you hold your breath on such things until january 31st. :( | 19:05 |
| rbergeron | since we like you being alive. | 19:05 |
| harlowja_ | lol | 19:05 |
| * mordred enjoys live harlowja_s more than dead harlowja_s | 19:05 | |
| clarkb | however this may be good science for testing the hypothesis that if harlowja_ holds hsi breaks for 2 weeks we all get what we want | 19:06 |
| rbergeron | harlowja_: all that said, if you have specific use cases or "we're not waiting so we're going to xyz instead" stories you would like to share with me via email, I am all for additional ammunition to cut through the bureaucracy (there really isn't much of this) or the politics (there is a bit more here, but not much, but enough to move things along slowly) | 19:06 |
| harlowja_ | rbergeron i can propose something via email, sure | 19:07 |
| rbergeron | but the important thing to me is to do it right, and make sure it's resourced correctly (since handling inevitable flood of contributors and governance and documentation takes effort, and it's Muy UnBueno for happiness without) | 19:07 |
| harlowja_ | right right | 19:08 |
| harlowja_ | makes sense | 19:08 |
| rbergeron | blah blah, y'all know this stuff :) | 19:08 |
| harlowja_ | ya | 19:08 |
| rbergeron | robyn@redhat if you want to drop me a line, i am happy to add it to my boiling pot of | 19:08 |
| rbergeron | of .. I guess my boiling pot of Feeling of Hurry Up Already | 19:08 |
| rbergeron | :) | 19:09 |
| harlowja_ | or i'll suffocate; i can't hold my breath much longer | 19:09 |
| harlowja_ | lol | 19:09 |
| rbergeron | (and i was joking about kolla-ansible, also ;D) | 19:09 |
| harlowja_ | :-P | 19:13 |
| openstackgerrit | David Shrewsbury proposed openstack-infra/nodepool: Handle provider removal from config https://review.openstack.org/418112 | 21:00 |
| *** jamielennox|away is now known as jamielennox | 21:00 | |
| *** timrc has quit IRC | 21:01 | |
| Shrews | pabelanger: i thought https://review.openstack.org/410939 was supposed to go against master, and features/zuulv3 branch would totally remove it (as in https://review.openstack.org/418013). Did I misread scrollback earlier? | 21:05 |
| *** jlk has quit IRC | 21:06 | |
| pabelanger | Shrews: Ah, yes. I need to move that to master | 21:08 |
| *** hasharAway is now known as hashar | 21:26 | |
| *** timrc has joined #zuul | 21:49 | |
| *** jlk has joined #zuul | 21:51 | |
| *** jlk has quit IRC | 21:51 | |
| *** jlk has joined #zuul | 21:51 | |
| clarkb | yup thats it, note you can't have that change number on master iirc because new change will be made due to branch move | 21:54 |
| jhesketh | Morning | 21:55 |
| jeblair | it's zuul meeting time in #openstack-meeting-alt | 22:01 |
| openstackgerrit | Paul Belanger proposed openstack-infra/nodepool: Source glean installs in simple-init https://review.openstack.org/414662 | 22:02 |
| openstackgerrit | Paul Belanger proposed openstack-infra/nodepool: Remove the ability for nodepoold to launch a builder https://review.openstack.org/418137 | 22:18 |
| openstackgerrit | Paul Belanger proposed openstack-infra/nodepool: Remove the ability for nodepoold to launch a builder https://review.openstack.org/418137 | 22:19 |
| *** saneax-_-|AFK is now known as saneax | 23:25 | |
| *** harlowja_ has quit IRC | 23:40 | |
| *** harlowja has joined #zuul | 23:41 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!