*** odyssey4me has quit IRC | 00:09 | |
*** odyssey4me has joined #zuul | 00:09 | |
*** xinliang has quit IRC | 00:11 | |
*** mrwtaqjb has joined #zuul | 00:12 | |
*** mrwtaqjb has quit IRC | 00:12 | |
*** xinliang has joined #zuul | 00:24 | |
*** xinliang has quit IRC | 00:24 | |
*** xinliang has joined #zuul | 00:24 | |
*** JasonCL has joined #zuul | 00:33 | |
*** JasonCL has quit IRC | 00:38 | |
*** JasonCL has joined #zuul | 01:11 | |
*** JasonCL has quit IRC | 01:17 | |
*** JasonCL has joined #zuul | 01:29 | |
openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul master: Add allowed-triggers and allowed-reporters tenant settings https://review.openstack.org/554082 | 01:30 |
---|---|---|
*** JasonCL has quit IRC | 01:34 | |
tristanC | SpamapS: tobiash: here you go: https://review.openstack.org/#/q/topic:nodepool-web | 01:39 |
SpamapS | sweeeeeeet | 01:40 |
tristanC | it actually deployed it here: https://softwarefactory-project.io/zuul/local/labels.html | 01:40 |
*** Wei_Liu has joined #zuul | 02:07 | |
*** swest1 has joined #zuul | 02:17 | |
*** swest has quit IRC | 02:18 | |
*** JasonCL has joined #zuul | 02:59 | |
*** JasonCL has quit IRC | 03:04 | |
jhesketh | tristanC: those improvements to the webui are awesome btw :-) | 03:27 |
dmsimard | tristanC: I'm not sure I would expose the IP and the port in the web UI by default. Can you make that an opt-in toggle maybe ? | 03:31 |
dmsimard | Is the node name/id available to display? | 03:33 |
tristanC | jhesketh: glad you like them, I hope we'll be able to demonstrate admin action through the webui soon, like hold a build, retrigger a job or even delete nodes | 04:12 |
tristanC | dmsimard: yes sure, i think we need to announce admin capabilities on the /info endpoint, and then the ui could be adapted | 04:13 |
*** JasonCL has joined #zuul | 04:35 | |
*** JasonCL has quit IRC | 04:41 | |
jhesketh | tristanC: neat :-) | 04:41 |
openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul master: web: add zookeeper driver and /{tenant}/labels controller https://review.openstack.org/553979 | 05:06 |
openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul master: web: add nodepool labels list page https://review.openstack.org/553980 | 05:06 |
openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul master: web: add /{tenant}/nodes controller https://review.openstack.org/553998 | 05:06 |
openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul master: web: add nodepool nodes status page https://review.openstack.org/553999 | 05:06 |
*** openstackgerrit has quit IRC | 05:49 | |
*** bhavik1 has joined #zuul | 06:10 | |
*** flepied has quit IRC | 06:32 | |
*** bhavik1 has quit IRC | 06:35 | |
*** yolanda_ is now known as yolanda | 07:21 | |
*** swest1 has quit IRC | 07:37 | |
*** AJaeger has quit IRC | 07:53 | |
*** aluria` is now known as aluria | 08:04 | |
*** aluria has quit IRC | 08:05 | |
*** aluria has joined #zuul | 08:05 | |
*** flepied has joined #zuul | 08:08 | |
*** jpena|off is now known as jpena | 08:59 | |
*** hashar has joined #zuul | 09:32 | |
*** JasonCL has joined #zuul | 09:33 | |
*** hashar has quit IRC | 09:39 | |
*** JasonCL has quit IRC | 09:43 | |
*** jamielennox has quit IRC | 09:43 | |
*** jamielennox has joined #zuul | 09:48 | |
*** hashar has joined #zuul | 10:03 | |
*** Wei_Liu has quit IRC | 11:01 | |
*** Wei_Liu has joined #zuul | 11:02 | |
*** JasonCL has joined #zuul | 11:41 | |
*** logan- has quit IRC | 11:52 | |
*** logan- has joined #zuul | 11:52 | |
*** dmellado has quit IRC | 12:24 | |
*** dmellado has joined #zuul | 12:27 | |
*** odyssey4me has quit IRC | 12:42 | |
*** odyssey4me has joined #zuul | 12:42 | |
*** rlandy has joined #zuul | 12:43 | |
*** dkranz has joined #zuul | 12:48 | |
*** jpena is now known as jpena|lunch | 12:56 | |
*** AJaeger has joined #zuul | 13:14 | |
*** sshnaidm is now known as sshnaidm|afk | 13:31 | |
*** myoung|afk is now known as myoung|rover | 13:58 | |
*** jpena|lunch is now known as jpena | 14:06 | |
*** flepied has quit IRC | 14:23 | |
*** flepied_ has joined #zuul | 14:23 | |
*** flepied_ is now known as flepied | 14:24 | |
dmsimard | btw I came across this last weekend, just sharing -- it's a recording of what every Ansible callback does. I thought it was pretty cool, there's a lot of nice callbacks I didn't even know existed: https://rndmh3ro.github.io/ | 14:31 |
dmsimard | fungi: am I supposed to make that particular story public btw ? | 14:56 |
dmsimard | corvus: ^ | 14:56 |
*** hashar is now known as hasharAway | 15:14 | |
Shrews | ugh. whenever i start over in playing around with devstack, i always fall into some fresh devstack-brokenness-hell that hinders my progress | 15:23 |
* Shrews (like Garfield) hates Mondays | 15:24 | |
*** openstackgerrit has joined #zuul | 15:33 | |
openstackgerrit | Fabien Boucher proposed openstack-infra/zuul master: Add a loading_errors info to {tenant}/info endpoint https://review.openstack.org/553873 | 15:33 |
clarkb | Shrews: I tend to start it from scratch on a fresh VM which seems to make it happier than trying to resurrect old ones | 15:34 |
Shrews | clarkb: that's what i did | 15:35 |
Shrews | clarkb: just going to use the stable/queens branch | 15:35 |
Shrews | master is b0rked b/c of https://bugs.launchpad.net/nova/+bug/1755756 | 15:35 |
openstack | Launchpad bug 1755756 in OpenStack Compute (nova) "ImageUnacceptable: Image is unacceptable: 'qemu-img info' parsing failed" [Undecided,New] | 15:35 |
*** flepied has quit IRC | 15:37 | |
Shrews | this weekend, i finally got a machine with enough umph to run my own zuul, so trying to setup all the things | 15:38 |
corvus | dmsimard: you mean 2001683? if you're satisfied there isn't a security issue, then yes, please make it public. | 15:40 |
dmsimard | corvus: ack, I'll try and reproduce on a master deployment before opening it up. | 15:41 |
corvus | dmsimard: sounds good, thx | 15:41 |
openstackgerrit | Clark Boylan proposed openstack-infra/zuul-jobs master: Refactor "emit-ara-html" into a new role: "ara-report" https://review.openstack.org/513874 | 15:45 |
clarkb | Shrews: reporter on that bug says it affects pike too? in any case could not open /var/lib is probably a system level issue clashing with nova's expectations? | 15:50 |
Shrews | clarkb: doesn't matter. queens is working :) | 15:50 |
openstackgerrit | James E. Blair proposed openstack-infra/zuul master: Add irrelevant-branches negative matcher https://review.openstack.org/552809 | 16:00 |
corvus | SpamapS: assume you're busy so i fixed up the things i saw wrong with that ^ | 16:00 |
corvus | i was thinking once again whether we could do something other than 'irrelevant'... maybe just 'not-branches' and 'not-files' would be okay. | 16:01 |
*** dtruong has joined #zuul | 16:02 | |
*** flepied has joined #zuul | 16:16 | |
openstackgerrit | James E. Blair proposed openstack-infra/zuul master: Update README and add GPL license https://review.openstack.org/554280 | 16:25 |
openstackgerrit | Matthieu Huin proposed openstack-infra/zuul master: zuul web: add admin endpoint, enqueue & autohold commands https://review.openstack.org/539004 | 16:43 |
openstackgerrit | Matthieu Huin proposed openstack-infra/zuul master: zuul autohold: allow operator to specify nodes TTL https://review.openstack.org/543403 | 16:43 |
*** jpena is now known as jpena|brb | 16:43 | |
*** myoung|rover is now known as myoung|food | 16:48 | |
*** weshay is now known as weshay_brb | 17:05 | |
*** dtruong has quit IRC | 17:05 | |
hughsaunders | Oh, no electrofelix today :( | 17:19 |
hughsaunders | Hey mordred, I'm working on a nodepool plugin for Jenkins (https://github.com/rcbops/nodepool-plugin) Its not in a working state but isn't that pretty. However it might be a good time to move it into the openstack-infra namespace, get some test and publication jobs going.. whos the best person to talk to about that? | 17:20 |
hughsaunders | *it IS in a working state | 17:21 |
pabelanger | I've noticed when I got to zuul.o.o, for a brief moment, I see some unrendered HTML (missing CSS) flash in the top left of my browser apply the CSS. Any ideas where to look to fix that? | 17:26 |
pabelanger | before applying* | 17:26 |
pabelanger | using chrome | 17:27 |
clarkb | pabelanger: I believe mordred's change series does correct that | 17:28 |
pabelanger | ah, cool! | 17:28 |
clarkb | pabelanger: the problem was it loaded the templates before applying them or something but in proposed change there is a change that does it all together | 17:28 |
openstackgerrit | Monty Taylor proposed openstack-infra/zuul-jobs master: Uninstall and reinstall siblings one at a time https://review.openstack.org/554297 | 17:28 |
pabelanger | clarkb: nice | 17:28 |
clarkb | hughsaunders: if you haven't read it yet the infra-manual has a project creators section that will walk you through adding projects to gerrit. I think with the proposed zuul split out of openstack proper we might consider not adding any new zuul related proejcts to openstack proper as well (they'd still be hosted by our gerrit just not in openstack governance) | 17:31 |
mordred | pabelanger, clarkb: yes - you can verify the fix here: http://logs.openstack.org/27/551327/7/check/build-javascript-content/de50d9b/npm/html/status.html | 17:32 |
hughsaunders | clarkb: Thanks.. so openstack-infra maybe but not openstack? | 17:32 |
hughsaunders | clarkb: I guess I was also wanting to gauge interest as openstack is moving away from Jenkins so wondered if Jenkins plugins are still interesting. | 17:32 |
hughsaunders | I know electrofelix is interested, but he isn't around :/ | 17:33 |
clarkb | hughsaunders: I think you'll find that many of us are not very interested in making jenkins things work at this point :) we'll happily host the project for you under "stackforge" if you want access to the tooling and CI | 17:33 |
pabelanger | mordred: ++ yup, no longer see the issue | 17:33 |
clarkb | which is related to electrofelix's other question. I don't think the hosting of related projects changes for now. They still go in stackforge unless they want to be part of openstack governance. | 17:34 |
clarkb | I should respond to that email thread once our zuul install is happy with new projects | 17:35 |
odyssey4me | clarkb stackforge as a namespace still is used for things? | 17:35 |
clarkb | odyssey4me: not as a namespace, but it is still technically what things are called that we host that aren't part of openstack governance iirc | 17:35 |
clarkb | odyssey4me: ara for example is not an openstack project but we host it as a "stackforge" project I think | 17:35 |
clarkb | fungi: ^ as our resident tc member can probably clarify anything I got wrong there | 17:35 |
dmsimard | yeah ara is kinda like a stackforge project | 17:35 |
odyssey4me | clarkb ah ok, alright - so where do we import the jenkins plugin to? | 17:36 |
mordred | yah | 17:36 |
dmsimard | I mean, unless it gets formally picked up by openstack foundation, openstack or zuul :p | 17:36 |
clarkb | odyssey4me: openstack/your-project-here is the current location | 17:36 |
clarkb | odyssey4me: the project creation docs should touch on that too I think | 17:36 |
clarkb | basically hosting on openstack's project hosting doesn't make an openstack project. Joining the openstack governance does | 17:36 |
odyssey4me | ok, we thought that *maybe* this might need to go into openstack-infra, rather than openstack, given that's where zuul lives | 17:36 |
fungi | clarkb: odyssey4me: we stopped officially using the term "stackforge" a couple years back and just refer to "official OpenStack deliverable repositories" vs "unofficial repositories" | 17:37 |
*** jpena|brb is now known as jpena | 17:37 | |
fungi | "stackforge" is still sort of used as shorthand for unofficial repos | 17:38 |
fungi | odyssey4me: nah, git namespaces are entirely irrelevant at this point | 17:38 |
fungi | we'd like to work out a way to get rid of them and flatten our git hosting | 17:39 |
odyssey4me | okie dokey, 'openstack' it is hughsaunders :) | 17:39 |
fungi | they're mostly an artifact of setting them up to map to different github orgs for repository mirroring | 17:39 |
hughsaunders | k | 17:39 |
hughsaunders | are they any jenkins plugins in stackforge that I can copy publishing pipelines from? | 17:40 |
fungi | basically "openstack" is the catch-all git namespace we host things out of (official or unofficial) | 17:40 |
clarkb | hughsaunders: some of our old plugins may have jobs but I'm unsure if they are tested post zuulv3 transition | 17:40 |
clarkb | with zuulv2 we did have maven publishing, but I'd be surprised if any of it is exercised post zuulv3 transition | 17:41 |
clarkb | we can move that discussion over to #openstack-infra too as its likely more service based than zuul specific | 17:41 |
clarkb | unless maybe we want to bake that stuff into zuul-jobs for reuse? | 17:42 |
odyssey4me | clarkb perhaps best to start small ;) | 17:42 |
hughsaunders | odyssey4me: we will need maven publishing though | 17:43 |
odyssey4me | hughsaunders That should be reasonably straightforward to setup as an in-repo job to start with. | 17:43 |
hughsaunders | odyssey4me: yeah, doesn't matter where it is, just needs to be possible :) | 17:44 |
odyssey4me | It can be migrated into a built-in zuul job once it's stable. | 17:44 |
clarkb | hughsaunders: it should be possible. You sign up for an account on the nexus. Add credentials to your secrets, push from there. | 17:45 |
openstackgerrit | Merged openstack-infra/zuul master: Update README and add GPL license https://review.openstack.org/554280 | 17:46 |
*** weshay_brb is now known as weshay | 17:50 | |
openstackgerrit | Monty Taylor proposed openstack-infra/zuul master: web: add /{tenant}/jobs/{job_name} route https://review.openstack.org/550978 | 17:55 |
openstackgerrit | Monty Taylor proposed openstack-infra/zuul master: web: add /{tenant}/projects routes https://review.openstack.org/550979 | 17:55 |
openstackgerrit | Monty Taylor proposed openstack-infra/zuul master: web: add /{tenant}/pipelines route https://review.openstack.org/541521 | 17:55 |
openstackgerrit | Monty Taylor proposed openstack-infra/zuul master: Rename javascript package to @zuul-ci/dashboard https://review.openstack.org/551999 | 17:55 |
*** myoung|food is now known as myoung|rover | 17:58 | |
Shrews | What are we using to install bubblewrap on ubuntu? | 18:00 |
mordred | Shrews: ppa:openstack-ci-core/bubblewrap | 18:01 |
Shrews | mordred: thx | 18:01 |
dmsimard | mordred: I've been meaning to peek at your frontend stack patches out of curiosity .. is the build-javascript-content job supposed to work ? | 18:02 |
dmsimard | I got blank pages when I looked yesto | 18:02 |
dmsimard | yesterday* | 18:02 |
mordred | dmsimard: yes - there is a bug that needs to be fixed starting at the angularjs -> angular transition patch | 18:03 |
mordred | dmsimard: ifyou look at the build-javascript-content jobs on the stack ahead of that, they shoudl all work | 18:04 |
dmsimard | Ok, neat. Thanks. | 18:04 |
mordred | corvus: 536882 to 551999 are ready for review | 18:04 |
dmsimard | The job was green but the page was blank so.. | 18:04 |
dmsimard | ¯\_(ツ)_/¯ | 18:04 |
*** myoung|rover is now known as myoung|taxi | 18:06 | |
*** myoung|taxi is now known as myoung|biab | 18:06 | |
mordred | dmsimard: yah. generally speaking that page should ALWAYS work and if it doesn't it's a big problem. however, in some cases (that one is a good example) we don't have any tests yet that will turn red | 18:08 |
dmsimard | That's where stuff like selenium comes in I guess ? | 18:08 |
mordred | obviously adding tests that turn red is a thing we should do | 18:08 |
mordred | dmsimard: yah - it might be a selenium test - it might be possible to add more targetted unit/functional tests to the js code - it's possible we want both - or it's possible we can add a test to the python zuul-web unittests that would catch the break in the angular patch ... or, perhaps all of the above :) | 18:09 |
corvus | mordred: do you want to merge some/all of that stack before the v3 release? | 18:09 |
openstackgerrit | Merged openstack-infra/zuul-jobs master: Refactor "emit-ara-html" into a new role: "ara-report" https://review.openstack.org/513874 | 18:10 |
mordred | corvus: yes, I'd like to get at least through 551327 before 3.0 | 18:14 |
corvus | mordred: that's in a stack that is dependent on an outdated change | 18:15 |
mordred | corvus: yes - we need to land the API endpoints, and there is an update needed to one of those patches, so we're only good up through 551999 | 18:17 |
corvus | mordred: there's a lot of stuff that we were deferring post v3 there. i'm surprised by this. | 18:17 |
corvus | mordred: is there something that's broken that absolutely must be merged? or is there something that can't wait until after v3 because it will prevent us from fixing it later? | 18:18 |
openstackgerrit | David Shrewsbury proposed openstack-infra/zuul master: Use unix sockets instead of TCP sockets https://review.openstack.org/542469 | 18:19 |
mordred | corvus: a little of both - it's more the second thing I'm concerned about, although the page flashing on load issue is pretty ugly and is the thing I'd consider in the first category | 18:20 |
corvus | mordred: we're not advertising this as a public api right, so why can't we fix it later? | 18:20 |
mordred | corvus: the stestr patch is in there because I hit an "I can't figure out what in the test suite is hanging/timing out" moment and could only solve it by locally updating things to use stestr - that can be removed from the stack easily though if you prefer | 18:22 |
mordred | corvus: page flashing is a pure-javascript issue - although we could release without fixing it of course | 18:22 |
corvus | mordred: from my pov, there are problems with the js/web stuff, but we've crossed the threshold we set for the v3 release, which was "make it installable". i'd like to just focus on the remaining v3 blockers (tests for console streaming, and security issues). | 18:25 |
corvus | mordred: https://review.openstack.org/547576 landed a bunch of "TODO: Fill in specifics in the next patch" to the docs. where is that patch? | 18:27 |
openstackgerrit | David Shrewsbury proposed openstack-infra/zuul master: Start log receiver in executor https://review.openstack.org/546304 | 18:27 |
mordred | corvus: ok. there is still one more issue I want to solve to verify that we don't need to do anything that would require a simultaneous update of python/javascript to get it rolled out | 18:28 |
corvus | mordred: okay, i agree that's the sort of thing we should try to avoid doing post-release :) | 18:29 |
mordred | corvus: that patch is on the todo list right after making sure we're good to go with being able to roll out changes | 18:29 |
mordred | corvus: yah- that's my #1 concern | 18:30 |
mordred | corvus: breaking people post 3.0 release would give me the sads :) | 18:30 |
*** rlandy is now known as rlandy|brb | 18:31 | |
openstackgerrit | James E. Blair proposed openstack-infra/zuul master: Remove empty sections from docs https://review.openstack.org/554315 | 18:32 |
corvus | mordred: ^ i think that patch would be a good idea, but i'd like to try to keep the docs as polished as possible, so i just removed the empty sections for now | 18:33 |
mordred | corvus: WFM | 18:33 |
corvus | (i just went to look something up, and they looked broken: https://docs.openstack.org/infra/zuul/admin/installation.html#static-offload ) | 18:34 |
mordred | corvus: well, the didn't just look broken... I'd contend they were broken :) | 18:34 |
corvus | mordred: oh good, i'm not going more insane :) | 18:34 |
mordred | :) | 18:34 |
*** openstackgerrit has quit IRC | 18:48 | |
*** openstackgerrit has joined #zuul | 18:58 | |
openstackgerrit | Andreas Jaeger proposed openstack-infra/zuul-jobs master: Uninstall and reinstall siblings one at a time https://review.openstack.org/554297 | 18:58 |
*** myoung|biab is now known as myoung|rover | 19:08 | |
*** rlandy|brb is now known as rlandy | 19:12 | |
*** jpena is now known as jpena|off | 19:15 | |
Shrews | corvus: would it be safe to move writeLoggingConfig() from _execute() to before startLogReceiver() in your log streaming change (line 623 here https://review.openstack.org/#/c/546304/3/zuul/executor/server.py)? | 19:23 |
Shrews | i'm not seeing any immediate reason why not, and the log receiver needs that file to exist before we start it | 19:23 |
Shrews | ok, that seems to make it break farther along, so seems ok | 19:27 |
corvus | Shrews: yeah, i can't think of a reason not to | 19:33 |
*** hasharAway has quit IRC | 19:46 | |
*** openstackgerrit has quit IRC | 19:48 | |
*** openstackgerrit has joined #zuul | 20:06 | |
openstackgerrit | David Shrewsbury proposed openstack-infra/zuul master: Start log receiver in executor https://review.openstack.org/546304 | 20:06 |
openstackgerrit | David Shrewsbury proposed openstack-infra/zuul master: Start log receiver in executor https://review.openstack.org/546304 | 20:08 |
clarkb | if I read that right it is essentially a proxy because we can only have a single connection back out of the ansible execution? | 20:09 |
clarkb | then executor can proxy as many connections from finger or wherever? | 20:09 |
clarkb | oh wait no its doing python logging so it can output to wherever? | 20:13 |
clarkb | Shrews: corvus left some comments but its mostly minor things that would be nice to address if new patchsets are required | 20:19 |
*** flepied has quit IRC | 20:19 | |
*** hashar has joined #zuul | 20:21 | |
Shrews | clarkb: i would not review yet | 20:22 |
Shrews | trying to fix the tests | 20:22 |
clarkb | oh I did just to understand what was going on | 20:22 |
*** flepied has joined #zuul | 20:24 | |
openstackgerrit | Jeremy Stanley proposed openstack-infra/zuul master: Add instructions for reporting vulnerabilities https://review.openstack.org/554352 | 20:44 |
fungi | corvus: i may not be around for the meeting in a couple hours, but that's ^ a start at vulnerability reporting instructions. i'm still working on the vulnerability management process document to go with it | 20:44 |
*** dtruong has joined #zuul | 20:59 | |
Shrews | wait, meeting is in a couple of hours? | 21:01 |
pabelanger | yah another 58mins | 21:02 |
pabelanger | I won't be able to attend this week | 21:03 |
*** hashar has quit IRC | 21:08 | |
Shrews | corvus: in your 546304 review, did you get familiar enough with the ansible python api to figure out what "task._host.get_name()" should actually be? (task does not have a _host attribute) | 21:25 |
jhesketh | Morning | 21:56 |
corvus | it's zuul meeting time in #openstack-meeting-alt | 22:02 |
*** eb0t- has joined #zuul | 22:38 | |
jlk | Dropped a comment on the github3.py bug: https://github.com/sigmavirus24/github3.py/issues/794#issuecomment-374413339 | 23:01 |
corvus | SpamapS: https://review.openstack.org/552809 looks like it still has a test failure; i haven't had a chance to dig into that today | 23:01 |
ianw | corvus: i was going to ask if we had free time in mtg, but https://review.openstack.org/#/c/553999/ with the nodepool web stuff in zuul. that's the direction we're going? i wasn't sure on the coupling there | 23:02 |
ianw | (sorry, asked in infra but here is better) | 23:02 |
corvus | ianw: i think the thing we discussed at the ptg was: 1) make a real nodepool-web service similar to (but independent from) zuul-web. 2) have zuul-web use the api from #1 to incorporate information into a single dashboard. maybe even use the same javascript code. | 23:04 |
corvus | ianw, tristanC: it doesn't look like that quite matches that discussion | 23:05 |
ianw | yeah, i wasn't sure how much zuul-web was going to turn into horizon :) | 23:07 |
corvus | indeed. i'd rather not implement that twice, so i think the ptg plan is a good compromise -- it'll look the same for the user, but we don't have to have two web apis. | 23:08 |
corvus | moreover, there's a *bunch* of stuff about nodepool that zuul doesn't need to know right now. i'd like to keep it that way. | 23:09 |
SpamapS | corvus: yeah it's just half-done. I've got some non-zuul stuff the next 2 days so will likely get to it mid-week. | 23:10 |
ianw | right, it was too a deep a coupling that i thought might not be so great longer term; my thought was that it would want to be behind a enable/disable flag at least | 23:11 |
corvus | SpamapS: cool, thanks. while you're doing the non-zuul stuff -- i was thinking of opening pandora's box-of-names again and using "not-files" and "not-branches" instead of irrelevant... maybe you can cipher on whether that's a good idea :) | 23:11 |
SpamapS | corvus: i'll let it bounce in my head until then. :) | 23:12 |
SpamapS | pretty echo-y in there today :-P | 23:12 |
*** myoung|rover is now known as myoung|bbl | 23:13 | |
corvus | ianw: in fact, i think the 'info' endpoint for zuul is turning into a place for things like that. so presumably, if there was no nodepool or the admin turned it off, then that page wouldn't show up in zuul. | 23:13 |
corvus | ianw: but in general, zuul is pretty tied to nodepool; there's no other option right now other than "run jobs with no nodes". so some degree of coupling is okay i think. for me, it's mostly about making sure it ends up in the right place (most of the implementation should be closer to nodepool than zuul) | 23:14 |
ianw | cool; well i think what nodepool outputs is looking pretty api-ish; especially after we did the stuff like made it use content detection for json etc. | 23:16 |
fungi | is tests.unit.test_github_crd.TestGithubCrossRepoDeps.test_crd_dependent known to sometimes yield nondeterministic results? | 23:51 |
openstackgerrit | Monty Taylor proposed openstack-infra/zuul-jobs master: Uninstall and reinstall siblings one at a time https://review.openstack.org/554297 | 23:59 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!