| -@gerrit:opendev.org- Zuul merged on behalf of James E. Blair https://matrix.to/#/@jim:acmegating.com: [zuul/nodepool] 871007: Update git https://review.opendev.org/c/zuul/nodepool/+/871007 | 01:16 | |
| -@gerrit:opendev.org- Zuul merged on behalf of James E. Blair https://matrix.to/#/@jim:acmegating.com: [zuul/zuul] 871008: Update git https://review.opendev.org/c/zuul/zuul/+/871008 | 01:25 | |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/zuul] 871012: Use importlib for versioning https://review.opendev.org/c/zuul/zuul/+/871012 | 02:03 | |
| @jim:acmegating.com | zuul-maint: how does this look for a zuul release? commit 944b9852c9f565ec433a808385d4ed0ad3f0abcc (HEAD -> master, tag: 8.1.0, origin/master, gerrit/master, refs/changes/08/871008/1) | 02:17 |
|---|---|---|
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/nodepool] 871013: Use zuul-promote-nox-docs https://review.opendev.org/c/zuul/nodepool/+/871013 | 02:21 | |
| @jim:acmegating.com | zuul-maint: ^ i'm going to self-approve that nodepool docs fix for expediency but would appreciate review | 02:22 |
| @jim:acmegating.com | neat. release notes on the web site don't have bullets any more. | 02:53 |
| @jim:acmegating.com | * release notes on the web site don't have bullets any more, so it's hard to tell when one ends and the next starts. | 02:53 |
| @jim:acmegating.com | maybe a change to reno? or the sphinx theme? | 02:53 |
| @clarkb:matrix.org | both seem possible. I'm not quite sure how reno injects its contents | 02:54 |
| -@gerrit:opendev.org- Zuul merged on behalf of James E. Blair https://matrix.to/#/@jim:acmegating.com: [zuul/nodepool] 871013: Use zuul-promote-nox-docs https://review.opendev.org/c/zuul/nodepool/+/871013 | 03:44 | |
| @jim:acmegating.com | zuul-maint: how does this look for a nodepool release? commit bfdf6a476301ed9dd43ab94477ed39a9c4fc3117 (HEAD -> master, tag: 8.1.0, origin/master, refs/changes/13/871013/1) | 03:53 |
| -@gerrit:opendev.org- Artem Goncharov proposed: [zuul/zuul] 859939: Implement Gitea driver https://review.opendev.org/c/zuul/zuul/+/859939 | 07:12 | |
| -@gerrit:opendev.org- Artem Goncharov proposed: [zuul/zuul] 859940: Implement tests of the gitea driver https://review.opendev.org/c/zuul/zuul/+/859940 | 07:13 | |
| -@gerrit:opendev.org- Zuul merged on behalf of Christian von Schultz: [zuul/nodepool] 867333: Fix AWS quota limits for vCPUs https://review.opendev.org/c/zuul/nodepool/+/867333 | 07:30 | |
| -@gerrit:opendev.org- Marvin Becker proposed wip: [zuul/nodepool] 867971: Add additional pod specs to openshift driver https://review.opendev.org/c/zuul/nodepool/+/867971 | 09:45 | |
| -@gerrit:opendev.org- Marvin Becker proposed wip: [zuul/nodepool] 871060: Add volumes to openshift pod workers https://review.opendev.org/c/zuul/nodepool/+/871060 | 09:49 | |
| -@gerrit:opendev.org- Alex Hornung proposed: [zuul/zuul] 871100: gerrit driver: fix bug around unicode branch names https://review.opendev.org/c/zuul/zuul/+/871100 | 14:21 | |
| -@gerrit:opendev.org- Per Wiklund proposed: [zuul/nodepool] 871102: Introduce driver for openshift virtualmachines https://review.opendev.org/c/zuul/nodepool/+/871102 | 14:49 | |
| @elpell:matrix.org | > <@morucci:matrix.org> Per Wiklund: Hi that might be a cool feature. How about adding a new type "vmi" here https://zuul-ci.org/docs/nodepool/latest/openshift.html#attr-providers.[openshift].labels.type ? From my understanding this is just another k8s resource type that Nodepool might be able to request. | 14:53 |
| Well yes it should be just a new resource, however you need to add a new api client for managing these objects. And connecting to it differs as it in our setup requires a loadbalancer service for exposing an ssh endpoint, which then the driver would have to handle. | ||
| @elpell:matrix.org | > <@morucci:matrix.org> Per Wiklund: do you already have an open review on Gerrit for that new driver ? | 14:53 |
| https://review.opendev.org/c/zuul/nodepool/+/871102 please review :) | ||
| -@gerrit:opendev.org- Simon Westphahl proposed: | 15:04 | |
| - [zuul/zuul] 871106: Require latest layout for processing mgmt events https://review.opendev.org/c/zuul/zuul/+/871106 | ||
| - [zuul/zuul] 871107: Periodically cleanup leaked pipeline state https://review.opendev.org/c/zuul/zuul/+/871107 | ||
| - [zuul/zuul] 871108: Cleanup deleted pipelines and and event queues https://review.opendev.org/c/zuul/zuul/+/871108 | ||
| @jpew:matrix.org | Does `job.requires` not make the job wait if the `job.provides` is in the same change? | 15:48 |
| @jpew:matrix.org | I have a job that requires something, and it's running right away instead of waiting for the job that provides it | 15:48 |
| @jim:acmegating.com | jpew: nope, that's handled by job.dependencies. provides/requires is for relationships between different queue items. | 15:52 |
| @jim:acmegating.com | the "for other changes" bit in the docs is key. https://zuul-ci.org/docs/zuul/latest/config/job.html#attr-job.requires is the main explanation | 15:54 |
| @jim:acmegating.com | some systems may need to use both requires and dependencies | 15:55 |
| @jpew:matrix.org | It would be nice if dependencies could be inherited from the parent instead of overridden | 15:56 |
| @fungicide:matrix.org | there are a few places in zuul configuration where we've had to make a tough call between merging and overriding with inheritance. the problem with merging is that there's no convenient way to subtract what's coming from a parent, while the problem with overriding is that you sometimes end up having to duplicate a lot of things from the parent in order to "just add something" | 16:51 |
| @fungicide:matrix.org | the proposed fileset spec had an interesting grammar for that sort of thing, but boy does it get complex/fragile | 16:52 |
| @jpew:matrix.org | fungi: Ya, I can see how that gets complicated | 16:53 |
| @jpew:matrix.org | For dependencies, somthing like | 16:54 |
| ``` | ||
| dependencies: | ||
| - name: foo | ||
| soft: false | ||
| inherited: true | ||
| ``` | ||
| might be ideal | ||
| @jpew:matrix.org | * For dependencies, somthing like | 16:55 |
| ``` | ||
| dependencies: | ||
| - name: foo | ||
| soft: false | ||
| inherited: true | ||
| ``` | ||
| might be ideal | ||
| @g_gobi:matrix.org | Hi, | 17:55 |
| I want to pause the job in the middle of execution. That means I want to pause the execution after completion of some modules in the job. | ||
| Is it possible? | ||
| @clarkb:matrix.org | > <@g_gobi:matrix.org> Hi, | 17:57 |
| > I want to pause the job in the middle of execution. That means I want to pause the execution after completion of some modules in the job. | ||
| > Is it possible? | ||
| Should be. Look at the zuul_return modules pause funcyionality | ||
| @jim:acmegating.com | https://zuul-ci.org/docs/zuul/latest/job-content.html#pausing-the-job | 17:59 |
| @g_gobi:matrix.org | No Zuul_return pause the job until the dependent job gets completed. My usecase is like pause the job in the middle of its execution. Like after running some modules in the play. | 18:01 |
| @g_gobi:matrix.org | Like I have 2 jobs. | 18:05 |
| Job1 and job2. | ||
| Job1 creates files and do some stuffs using those files. | ||
| I just want to pause the job after file creation. I want to manually check whether the files were created. | ||
| I know we can do that using ansible itself. I just want to know whether this is possible. | ||
| @fungicide:matrix.org | other than having an ansible task check and report those things for you (which is my recommendation), or an ansible task to run a very long blocking process (sleep 3600), i don't know of a solution to your exact problem | 18:13 |
| @fungicide:matrix.org | the zuul executor pretty much completely surrenders task execution to ansible for each play, so any sort of mid-play action would need to be an instruction to ansible itself | 18:15 |
| @fungicide:matrix.org | the executor could in theory pause between individual plays, i think, but i don't know of any mechanism we've implemented to do that nor how you would expect to communicate that to zuul if it were to be a proposed feature | 18:17 |
| @g_gobi:matrix.org | One of my usecase is like I'm running my testing using zuul. I want to check the infrastructure after some of test-suite. Like if any of the test-suite failed pause the execution and debug it. | 18:37 |
| @g_gobi:matrix.org | I thought admin can able to pause the job in UI. | 18:40 |
| @g_gobi:matrix.org | Can we have an option like gitlab manual job? | 18:47 |
| @g_gobi:matrix.org | wait for user approval to begin the job | 18:48 |
| @clarkb:matrix.org | tdlaw: you can do that with post review pipelines | 18:49 |
| @g_gobi:matrix.org | You mean we can manually approve each job in the pipeline? | 18:51 |
| @clarkb:matrix.org | Reviewers can leave votes that trigger jobs in pipelines when they are ready for those jobs to run | 18:56 |
| @clarkb:matrix.org | Gating is a version of this which is reviewers indicating jobs should run and if they succeed the change can merge. But you can do similar without the merge step and just trigger jobs when you are satisifed a change has met certain criteria for those jobs. | 18:57 |
| @g_gobi:matrix.org | Is this possible in an already running pipeline? | 19:02 |
| @g_gobi:matrix.org | Like pipeline started. Job1 running. | 19:02 |
| Job 2 waiting for the vote/review | ||
| @clarkb:matrix.org | no, would require separate pipelines with different triggers and requirements | 19:03 |
| @g_gobi:matrix.org | Okay | 19:03 |
| @g_gobi:matrix.org | What are all the features available for the admin user. | 19:04 |
| Like deqeue the job. | ||
| @g_gobi:matrix.org | In zuul console | 19:04 |
| @g_gobi:matrix.org | When I tried quick installation that options were not available | 19:04 |
| @g_gobi:matrix.org | * What are all the features available for the admin user in zuul console. I heard they are able to deqeue the job. | 19:07 |
| @clarkb:matrix.org | I'm not completely sure myself. I don't use those features (though maybe I should) | 19:07 |
| @g_gobi:matrix.org | Oh okay. But there are some options there right? | 19:08 |
| @clarkb:matrix.org | I don't think there are options for triggering jobs there. | 19:08 |
| @clarkb:matrix.org | In general the pipelines define the rules for triggering jobs so triggering options would be documented by the pipelines | 19:09 |
| @g_gobi:matrix.org | No. I am asking for a dequeue job option | 19:10 |
| @clarkb:matrix.org | I'm pretty sure you can dequeue the buildset. Not sure about a specific job. | 19:10 |
| @g_gobi:matrix.org | Oh okay. Thanks a lot for those details. | 19:11 |
| @g_gobi:matrix.org | So we can't pause the job during execution. | 19:11 |
| @clarkb:matrix.org | you can using zuul_return and ansible tasks. But there isn't an external mechanism for that as far as I know | 19:12 |
| @clarkb:matrix.org | I'm not sure how to implement that safely either. We would have to kill ansible somehow and that can have bad side effects and you wouldn't necessarily be able to restart the ansible from where it left off after you kill it | 19:13 |
| @clarkb:matrix.org | If you did some sort of process pause at the linux level you'd be risking broken networking on resumption? In general I'm not sure how that would be possible | 19:13 |
| @g_gobi:matrix.org | Do you know how to enable those deqeue buildset options in the zuul console. | 19:13 |
| I'm using the quick installation script | ||
| @clarkb:matrix.org | you need to set up authentication | 19:14 |
| @clarkb:matrix.org | the docs should cover how to do that. THere are a couple of options | 19:14 |
| @g_gobi:matrix.org | Is that included in a quick installation? I was able to see the keycloak folder there | 19:15 |
| @clarkb:matrix.org | Looks like there is an add on tutorial with separate steps after you run quickstart | 19:16 |
| @clarkb:matrix.org | It isn't included by the base quickstaet but there is a separate docker-compose file and a tutorial document to get keycloak up and running with zuul if you follow up the quickstart with those steps | 19:16 |
| @g_gobi:matrix.org | Could you please share that tutorial document? | 19:17 |
| @clarkb:matrix.org | https://zuul-ci.org/docs/zuul/latest/tutorials/keycloak.html | 19:18 |
| @g_gobi:matrix.org | Thanks I'll try those steps.👍🎉 | 19:21 |
| @jpew:matrix.org | Should this work in a tenant config to prevent a specific branch from being always-dynamic: | 21:41 |
| ``` | ||
| include-branches: | ||
| - "^features/special$" | ||
| always-dynamic-branches: | ||
| - "^features/.*$" | ||
| ``` | ||
| @clarkb:matrix.org | I suspect not? You may need to use a negative lookahead on line 4 to exclude the one you want in the include branches | 21:45 |
| @clarkb:matrix.org | But we don't use those features so I'm not super familiar with them | 21:45 |
| @jpew:matrix.org | I think `TenantProjectConfig.includeBranches()` is actually just plain broken.... or at least doesn't match the documentation | 22:04 |
| @jpew:matrix.org | It definetly is not written so that `include-branches` has priority over `exclude-branches`, but the wording the docs is confusing | 22:05 |
| @jpew:matrix.org | * I think `TenantProjectConfig.includesBranches()` is actually just plain broken.... or at least doesn't match the documentation | 22:05 |
| -@gerrit:opendev.org- Joshua Watt proposed: [zuul/zuul] 871197: Fix include-branches being lower priority than exclude-branches https://review.opendev.org/c/zuul/zuul/+/871197 | 22:24 | |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/nodepool] 871199: Add API timing debug statements to openstack driver https://review.opendev.org/c/zuul/nodepool/+/871199 | 22:57 | |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!