| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [opendev/zone-zuul-ci.org] 877727: Add registry.zuul-ci.org https://review.opendev.org/c/opendev/zone-zuul-ci.org/+/877727 | 01:25 | |
| -@gerrit:opendev.org- Zuul merged on behalf of James E. Blair https://matrix.to/#/@jim:acmegating.com: [opendev/zone-zuul-ci.org] 877727: Add registry.zuul-ci.org https://review.opendev.org/c/opendev/zone-zuul-ci.org/+/877727 | 02:04 | |
| -@gerrit:opendev.org- Zuul merged on behalf of James E. Blair https://matrix.to/#/@jim:acmegating.com: [zuul/nodepool] 877707: Fix exception in cache update https://review.opendev.org/c/zuul/nodepool/+/877707 | 07:26 | |
| -@gerrit:opendev.org- Tobias Urdin proposed: [zuul/zuul] 877599: web: fix build and buildoutput pages for dark mode https://review.opendev.org/c/zuul/zuul/+/877599 | 07:37 | |
| -@gerrit:opendev.org- Tobias Urdin proposed: [zuul/zuul] 877740: web: dont toggle dark mode in resetState https://review.opendev.org/c/zuul/zuul/+/877740 | 07:40 | |
| -@gerrit:opendev.org- Tobias Urdin proposed: [zuul/zuul] 877744: web: dont set css when prefers dark https://review.opendev.org/c/zuul/zuul/+/877744 | 08:10 | |
| -@gerrit:opendev.org- Tobias Urdin proposed: [zuul/zuul] 877599: web: fix build and buildoutput pages for dark mode https://review.opendev.org/c/zuul/zuul/+/877599 | 09:22 | |
| -@gerrit:opendev.org- Tobias Urdin proposed: [zuul/zuul] 877740: web: dont toggle dark mode in resetState https://review.opendev.org/c/zuul/zuul/+/877740 | 09:22 | |
| -@gerrit:opendev.org- Tobias Urdin proposed: [zuul/zuul] 877744: web: dont set css when prefers dark https://review.opendev.org/c/zuul/zuul/+/877744 | 09:22 | |
| @scrungusss:matrix.org | could anyone point me at some docs for how to add a log source and file, like how zuul collects devstack logs and displays them in the job summary? | 10:56 |
|---|---|---|
| -@gerrit:opendev.org- Zuul merged on behalf of Antoine Musso: [zuul/zuul-jobs] 877700: Remove ignored success-url job attribute https://review.opendev.org/c/zuul/zuul-jobs/+/877700 | 12:39 | |
| @tobias-urdin:matrix.org | the series at https://review.opendev.org/q/topic:web-dark-mode is ready for review, you can check preview from the last patch in the series https://review.opendev.org/c/zuul/zuul/+/877744 – let me know what you think! :) | 12:56 |
| @jim:acmegating.com | Tobias Urdin: thanks, i plan to take a look next week :) | 14:06 |
| @jim:acmegating.com | scrungus: https://zuul-ci.org/docs/zuul-jobs/log-roles.html#role-fetch-output is a role that helps with that (but the docs aren't complete, you might want to look at the code too). | 14:08 |
| @fungicide:matrix.org | scrungus: you can see it configured in the devstack-base job here: https://opendev.org/openstack/devstack/src/branch/master/.zuul.yaml#L411-L452 | 14:11 |
| @scrungusss:matrix.org | thanks, i've seen this. I was wondering if it is possible to pipe output from a command? i.e. instead of something like `/var/log/mylog : logs` you could do something like `$( make_my_logs) : logs` ? | 14:15 |
| @fungicide:matrix.org | scrungus: for copying command output, i'd just create a new ansible task to run that command and write it to a file, then add the file to the list of log files for collecting | 14:17 |
| @fungicide:matrix.org | i think it already does similar things to get service logs out of journalctl | 14:18 |
| @scrungusss:matrix.org | i'm guessing that this playbook would be specified in the `post-run` step? If i am inheriting from a job that already has a defined `run` and `post-run` which i don't want to override, is there a way to add another task? | 14:20 |
| @fungicide:matrix.org | a job can add its own post-run playbooks, they will be inserted prior to the post-run playbooks for any inherited jobs | 14:21 |
| @fungicide:matrix.org | basically if job C inherits from B inherits from A then zuul runs A-pre,B-pre,C-pre,run,C-post,B-post,A-post in that order | 14:21 |
| @scrungusss:matrix.org | ah perfect, thanks a lot! | 14:22 |
| @fungicide:matrix.org | like onion layers | 14:22 |
| @jim:acmegating.com | https://zuul-ci.org/docs/zuul/latest/config/job.html#job talks about the onion -- last paragraph | 14:23 |
| @fungicide:matrix.org | now if only it also had pickle relish too | 14:23 |
| @fungicide:matrix.org | and a bit of mustard | 14:24 |
| @gchauvel:matrix.org | Tobias Urdin: Builds -> a job -> Console | 15:20 |
| @gchauvel:matrix.org | * Tobias Urdin: Builds -> a job -> Console , mouse over | 15:21 |
| -@gerrit:opendev.org- Tobias Urdin proposed: [zuul/zuul] 877822: web: fix console tab on build page for dark mode https://review.opendev.org/c/zuul/zuul/+/877822 | 16:17 | |
| @tobias-urdin:matrix.org | guillaumec: nice catch! fixed with ^ | 16:17 |
| @g_gobi:matrix.org | Do we have any docs on which merge option we need to use in gitlab for Zuul? | 16:49 |
| @g_gobi:matrix.org | If I use fast-forward merge, it failing with `MERGE_FAILURE`. | 16:51 |
| @g_gobi:matrix.org | * If I use fast-forward merge in gitlab repo setting, it failing with `MERGE_FAILURE`. | 16:54 |
| @clarkb:matrix.org | tdlaw: Is zuul failing to do its local merges? I assume so and that MERGE_FAILURE is being reported by zuul. In general you should set your merge type to match what is in the code review system. I'm not sure about gitlab specifically but I would look at your merger logs to see why it failed | 16:54 |
| @clarkb:matrix.org | it is entirely possible the failure is valid and you need to rebase | 16:54 |
| @clarkb:matrix.org | in general though I think fast-forward is somewhat at odds with zuul since zuul can build large speculative git states. If you need to ensure everything is fast forwardable then you end up having human construct those trees and dependencies and you lose this benefit of zuul | 16:56 |
| @g_gobi:matrix.org | merger logs means? Where I need to look that? | 16:57 |
| @g_gobi:matrix.org | Makes sense. Problem is I previously used 4.8.2 now only I update it to 8.1.0. After upgrade I'm seeing this issue. | 16:58 |
| @clarkb:matrix.org | tdlaw: the zuul-merger and zuul-executor processes (the zuul-executor also runs an internal merger process) | 16:59 |
| @g_gobi:matrix.org | I don't see zuul-merger service. I'm using docker based zuul environment. | 17:01 |
| @clarkb:matrix.org | then it is likely going to be in your executor logs | 17:02 |
| @g_gobi:matrix.org | Then I need to enable debug logs right? | 17:03 |
| @clarkb:matrix.org | maybe? I would check the logs first to see if they provide sufficient info | 17:04 |
| @g_gobi:matrix.org | I can able to see job completed and zk nodes released but I don't see merge info. | 17:44 |
| @clarkb:matrix.org | if you find the event id for the triggering event you can grep that value and should get a very complete picture. But jobs shouldn't run at all if the zuul merge is what failed. | 17:46 |
| @g_gobi:matrix.org | yeah. I checked the same. It ended with `Job execution took: 13027.151 seconds` | 17:52 |
| @clarkb:matrix.org | ok the merge failure may have been when it tried to merge the chnage in gitlab then? Can you check the final report for the buildset? | 17:53 |
| @clarkb:matrix.org | it should try to merge the MR and then leave a comment iirc and that would be the next place to look at if jobs are running since that implies the pre build merge was fine | 17:53 |
| @g_gobi:matrix.org | I checked both `Build succeeded` and `This change was not merged by the code review system.` both were not present in executor-logs. | 18:02 |
| @clarkb:matrix.org | yup so the issue was merging on the gitlab side | 18:03 |
| @clarkb:matrix.org | you will probably need to check your gitlab logs or otherwise determine why it failed. Likely because the change is no logner in a fast forwardable state? | 18:04 |
| @fungicide:matrix.org | and consider an alternative merge policy in gitlab, i don't know what the options are there or how you configure them but creating merge commits, cherry-picking, rebasing, squashing, etc when the commit(s) is/are non-fastforwardable | 18:06 |
| @g_gobi:matrix.org | Yeah merge commits seems to be working. | 18:07 |
| @fungicide:matrix.org | like clark said, restricting merge actions to only work for fast-forwardable changes loses most of zuul's useful functionality | 18:07 |
| @fungicide:matrix.org | merge commits are what we do in opendev's gerrit, and have done for a decade (i think... when did openstack give up on the cherry-pick method there?), works quite well for us in highly active projects | 18:08 |
| @g_gobi:matrix.org | If I set fastforward merge I can't able to see the merge option in UI too. If there is any commit added to the target-branch. Then I need to check gitlab-release notes. There was upgrade happened few weeks back. | 18:08 |
| @g_gobi:matrix.org | The problems is | 18:11 |
| I did zuul upgrade from 4.8.2 -> 8.1.0 3 days back | ||
| Gitlab upgrade happened 2 weeks back | ||
| So, I don't know in which area change happened. | ||
| @fungicide:matrix.org | the main drawback to merge commits is that they are additional commits in the history. some people don't like them because they think they hurt bisectability, but that's not really the case when you're using them for each change unit, more of a problem when you're infrequently merging tons of history at once between different branches. up side to merge commits is they precisely preserve your original commits, which makes developers' local repo copies more sensible as the same commit ids are in the final merge result | 18:11 |
| @fungicide:matrix.org | cherry-pick, rebase and squash annoyingly rewrite commits so they no longer match what's on the dev's local tree | 18:12 |
| @fungicide:matrix.org | which makes things like reverts more challenging too | 18:13 |
| @fungicide:matrix.org | fast-forward does have the benefit of still preserving the commits exactly, but with the major down side of forcing everyone to amend their commits and/or coordinate to sequence them in the exact order they plan to approve them | 18:14 |
| @gchauvel:matrix.org | Tobias Urdin: 👍️ | 18:18 |
| @gchauvel:matrix.org | Job "click for details" | 18:19 |
| @g_gobi:matrix.org | fungi: facts💯 | 18:22 |
| -@gerrit:opendev.org- Clark Boylan proposed: [zuul/zuul-jobs] 877834: Add ensure-quay-repo role https://review.opendev.org/c/zuul/zuul-jobs/+/877834 | 19:34 | |
| -@gerrit:opendev.org- Clark Boylan proposed: [zuul/zuul-jobs] 877834: Add ensure-quay-repo role https://review.opendev.org/c/zuul/zuul-jobs/+/877834 | 20:44 | |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!