| *** hamalq has quit IRC | 00:06 | |
| openstackgerrit | melanie witt proposed zuul/zuul master: Add button to go to top of build page https://review.opendev.org/c/zuul/zuul/+/790319 | 01:02 |
|---|---|---|
| openstackgerrit | melanie witt proposed zuul/zuul master: Add button to go to top of build page https://review.opendev.org/c/zuul/zuul/+/790319 | 01:07 |
| *** josefwells_ has joined #zuul | 01:12 | |
| *** josefwells has quit IRC | 01:16 | |
| *** josefwells has joined #zuul | 01:18 | |
| openstackgerrit | melanie witt proposed zuul/zuul master: Add button to go to top of build page https://review.opendev.org/c/zuul/zuul/+/790319 | 01:18 |
| *** josefwells_ has quit IRC | 01:19 | |
| openstackgerrit | Ian Wienand proposed zuul/nodepool master: Dockerfile: podman -- handle new configuration file https://review.opendev.org/c/zuul/nodepool/+/790531 | 01:36 |
| *** josefwells has quit IRC | 01:52 | |
| *** sam_wan has joined #zuul | 01:57 | |
| *** evrardjp has quit IRC | 02:33 | |
| *** evrardjp has joined #zuul | 02:33 | |
| *** bhavikdbavishi has joined #zuul | 02:57 | |
| *** bhavikdbavishi1 has joined #zuul | 03:00 | |
| *** bhavikdbavishi has quit IRC | 03:02 | |
| *** bhavikdbavishi1 is now known as bhavikdbavishi | 03:02 | |
| *** bhavikdbavishi has quit IRC | 03:28 | |
| *** bhavikdbavishi has joined #zuul | 03:33 | |
| *** rlandy has quit IRC | 03:57 | |
| *** saneax has joined #zuul | 04:27 | |
| *** jfoufas1 has joined #zuul | 04:42 | |
| *** ykarel has joined #zuul | 04:53 | |
| *** bhavikdbavishi has quit IRC | 05:25 | |
| *** bhavikdbavishi has joined #zuul | 05:26 | |
| *** fdegir has quit IRC | 05:26 | |
| *** fdegir has joined #zuul | 05:27 | |
| *** bhavikdbavishi has quit IRC | 05:48 | |
| openstackgerrit | Merged zuul/zuul master: getHoldRequest: pad zeros if not full length https://review.opendev.org/c/zuul/zuul/+/788668 | 06:06 |
| *** bhavikdbavishi has joined #zuul | 06:16 | |
| *** bhavikdbavishi1 has joined #zuul | 06:19 | |
| *** bhavikdbavishi has quit IRC | 06:21 | |
| *** bhavikdbavishi1 is now known as bhavikdbavishi | 06:21 | |
| *** vishalmanchanda has joined #zuul | 06:23 | |
| *** sam_wan has quit IRC | 06:32 | |
| *** bhavikdbavishi has quit IRC | 06:41 | |
| openstackgerrit | Simon Westphahl proposed zuul/zuul master: Implement API for caching raw config files in ZK https://review.opendev.org/c/zuul/zuul/+/789577 | 06:48 |
| *** jcapitao has joined #zuul | 06:49 | |
| openstackgerrit | Simon Westphahl proposed zuul/zuul master: Cache unparsed config files in Zookeeper https://review.opendev.org/c/zuul/zuul/+/771455 | 06:57 |
| openstackgerrit | Merged zuul/zuul master: Optimize layout re-calculation after re-enqueue https://review.opendev.org/c/zuul/zuul/+/788681 | 06:58 |
| openstackgerrit | Felix Edel proposed zuul/zuul master: Calculate build start and end time on executor server https://review.opendev.org/c/zuul/zuul/+/788915 | 07:01 |
| openstackgerrit | Felix Edel proposed zuul/zuul master: Lock/unlock nodes on executor server https://review.opendev.org/c/zuul/zuul/+/774610 | 07:01 |
| *** sshnaidm|afk has quit IRC | 07:03 | |
| openstackgerrit | Simon Westphahl proposed zuul/zuul master: Add a logical timestamp to management events https://review.opendev.org/c/zuul/zuul/+/789576 | 07:14 |
| openstackgerrit | Simon Westphahl proposed zuul/zuul master: Implement API for caching raw config files in ZK https://review.opendev.org/c/zuul/zuul/+/789577 | 07:14 |
| openstackgerrit | Simon Westphahl proposed zuul/zuul master: Cache unparsed config files in Zookeeper https://review.opendev.org/c/zuul/zuul/+/771455 | 07:14 |
| *** sshnaidm has joined #zuul | 07:20 | |
| *** hashar has joined #zuul | 07:22 | |
| openstackgerrit | Simon Westphahl proposed zuul/zuul master: Stream API for sharded Zookeeper data https://review.opendev.org/c/zuul/zuul/+/771454 | 07:27 |
| *** bhavikdbavishi has joined #zuul | 07:28 | |
| *** bhavikdbavishi1 has joined #zuul | 07:31 | |
| *** bhavikdbavishi has quit IRC | 07:33 | |
| *** bhavikdbavishi1 is now known as bhavikdbavishi | 07:33 | |
| *** rpittau|afk is now known as rpittau | 07:38 | |
| *** tosky has joined #zuul | 07:39 | |
| *** mgoddard has quit IRC | 07:45 | |
| *** nils has joined #zuul | 07:46 | |
| *** bhavikdbavishi has quit IRC | 07:48 | |
| *** mgoddard has joined #zuul | 07:50 | |
| *** jpena|off is now known as jpena | 07:51 | |
| mhu | clarkb: my reasoning was that a user would go through the following process: 1. Build for job J on change A fails 2. I don't have enough info for a fix and need to do a post-mortem 3. I want to hold the nodes next time a build of J fails on change A | 08:08 |
| mhu | In that spirit adding the autohold request button on the failed build's page since all the required arguments for the request can be easily pre-filled | 08:10 |
| mhu | makes sense | 08:10 |
| mhu | We could also add a "new Autohold" button to the autoholds page in https://review.opendev.org/c/zuul/zuul/+/768199 - that'd pop a form where you can tweak your autohold query | 08:13 |
| mhu | I'll answer your comment on gerrit as well | 08:14 |
| *** sam_wan has joined #zuul | 08:23 | |
| mhu | clarkb, shower thoughts made me realize the user who wants the autohold won't necessarily be the one with the power to create one. That'd mean the user would ask the operator "hey could you hold future failing builds similar to this one?" | 08:38 |
| *** ykarel_ has joined #zuul | 09:19 | |
| *** ykarel has quit IRC | 09:21 | |
| *** ykarel_ is now known as ykarel | 09:28 | |
| *** sam_wan has quit IRC | 09:30 | |
| openstackgerrit | Benjamin Schanzel proposed zuul/zuul master: Add tenant name on NodeRequests for Nodepool https://review.opendev.org/c/zuul/zuul/+/788680 | 09:35 |
| *** sam_wan has joined #zuul | 09:47 | |
| *** ykarel is now known as ykarel|lunch | 09:54 | |
| *** ykarel_ has joined #zuul | 09:58 | |
| *** ykarel|lunch has quit IRC | 10:00 | |
| *** sam_wan6 has joined #zuul | 10:13 | |
| *** sam_wan has quit IRC | 10:14 | |
| *** sam_wan6 has quit IRC | 10:22 | |
| *** ykarel_ is now known as ykarel | 10:56 | |
| *** jcapitao is now known as jcapitao_lunch | 10:57 | |
| *** jpena is now known as jpena|lunch | 11:15 | |
| *** Shrews has joined #zuul | 11:21 | |
| *** rlandy has joined #zuul | 11:40 | |
| *** jcapitao_lunch is now known as jcapitao | 11:46 | |
| *** jpena|lunch is now known as jpena | 11:56 | |
| openstackgerrit | Simon Westphahl proposed zuul/nodepool master: Assign waiting static nodes by request priority https://review.opendev.org/c/zuul/nodepool/+/790674 | 12:04 |
| *** dpawlik has quit IRC | 12:12 | |
| *** ykarel_ has joined #zuul | 12:18 | |
| openstackgerrit | Benjamin Schanzel proposed zuul/zuul master: Add tenant name on NodeRequests for Nodepool https://review.opendev.org/c/zuul/zuul/+/788680 | 12:19 |
| *** dpawlik3 has joined #zuul | 12:20 | |
| *** ykarel has quit IRC | 12:20 | |
| *** dpawlik3 is now known as dpawlik | 12:26 | |
| *** josefwells has joined #zuul | 12:27 | |
| openstackgerrit | Simon Westphahl proposed zuul/zuul master: Remove use of item's layout in Nodepool API https://review.opendev.org/c/zuul/zuul/+/790682 | 12:27 |
| swest | zuul-maint: ^ I found a place where the layout was still used outside the pipeline manager that I missed in 788681. | 12:32 |
| msuszko | ianw: Thanks for help with VHD images for Azure last Wednsday. I've made some progress since then. There is MS article to align RAW to full 1MiB, and then use `qemu-img convert -f raw -O vpc` instead of vhd-util. | 13:22 |
| msuszko | and to use fixed and force_size options: https://docs.microsoft.com/en-us/azure/virtual-machines/linux/create-upload-generic | 13:27 |
| *** ykarel_ has quit IRC | 13:42 | |
| *** ykarel_ has joined #zuul | 13:44 | |
| *** jcapitao has quit IRC | 13:58 | |
| *** jcapitao has joined #zuul | 13:59 | |
| *** iurygregory has quit IRC | 14:22 | |
| *** iurygregory has joined #zuul | 14:22 | |
| *** ykarel_ is now known as ykarel | 14:25 | |
| josefwells | Should I expect github checks API to be totally worky? I'm seeing jobs start, finish on zuul, but never notify back to github that the check completed (success or fail). If it fails, I see the job re-run in zuul repeatedly, but again never report back. Using the example pipeline_check from here: https://zuul-ci.org/docs/zuul/howtos/github_setup.html | 14:26 |
| josefwells | Also, I pulled my code and new images, so I am up to date as of Saturday. | 14:26 |
| *** hashar is now known as hasharAway | 14:29 | |
| fungi | josefwells: i recommend checking the scheduler logs, if it's having trouble (e.g. permissions) writing to the github api then it should log that | 14:31 |
| openstackgerrit | Benjamin Schanzel proposed zuul/zuul master: Log Ansible Error Lines Completely https://review.opendev.org/c/zuul/zuul/+/790708 | 14:37 |
| *** ykarel has quit IRC | 14:42 | |
| *** jfoufas1 has quit IRC | 15:03 | |
| josefwells | fungi, I see zuul.GithubConnection Updating existing check run... then I see zuul.GithubRequest: hit the check-runs api url and get result:200 | 15:24 |
| fungi | josefwells: that certainly does look like a success code, i agree | 15:24 |
| josefwells | fungi, I do see: X-Github-Event: check_suite (I wonder if this is the abort button being pushed) | 15:24 |
| josefwells | Unhandled X-Github-Event: check_suite | 15:25 |
| josefwells | maybe my enterprise github is too old? | 15:25 |
| josefwells | is there a version requirement? | 15:25 |
| fungi | that might be benign... do you see any tracebacks? | 15:25 |
| josefwells | no traceback there, still crawling the log, I have debug on the scheduler and executor as well, so lots to dig through | 15:25 |
| fungi | yeah, debug is not brief | 15:28 |
| *** jcapitao has quit IRC | 15:58 | |
| *** ykarel has joined #zuul | 16:00 | |
| *** hasharAway is now known as hashar | 16:02 | |
| *** bhavikdbavishi has joined #zuul | 16:25 | |
| *** ykarel has quit IRC | 16:29 | |
| *** bhavikdbavishi has quit IRC | 16:36 | |
| *** rpittau is now known as rpittau|afk | 16:36 | |
| *** jpena is now known as jpena|off | 17:07 | |
| *** hashar is now known as hasharAway | 17:27 | |
| *** nils has quit IRC | 17:33 | |
| openstackgerrit | Clark Boylan proposed zuul/zuul master: Check if statsd is set before using it https://review.opendev.org/c/zuul/zuul/+/790798 | 17:41 |
| josefwells | fungi, my github enterprise is 2.22.. but I scrubbed pretty hard and didn't see anything that looked like an error | 18:19 |
| fungi | josefwells: and you've looked over https://zuul-ci.org/docs/zuul/reference/drivers/github.html to make sure you've set the options you need? | 18:22 |
| fungi | you're using it as an application not a webhook config right? | 18:22 |
| josefwells | yes | 18:22 |
| josefwells | It was working when I was not using the checks API | 18:23 |
| josefwells | when zuul left comments on the PR, etc. Let me check the github driver again, maybe something changed and I missed it in my pull this weekend | 18:23 |
| fungi | and you've granted checks read and write to the zuul app? | 18:24 |
| josefwells | yeah. so now I commented "recheck" in the PR and the job is just running over an over | 18:25 |
| fungi | does the build log indicate why the job is restarting? | 18:25 |
| fungi | if you don't have the console log enabled for your builds, then you may also find it in the executor's debug log | 18:26 |
| fungi | it's possible the reason the build is restarting is unrelated to the switch to the checks api | 18:27 |
| josefwells | let me see what I can see.. interesting I was just looking at scheduler logs before.. so I might have misses something else | 18:29 |
| *** vishalmanchanda has quit IRC | 18:34 | |
| *** hamalq has joined #zuul | 18:50 | |
| josefwells | I moved the pipeline back to github status and opened a new PR, worked fine. reported the status back. "recheck" comment runs once, no more infinite runs | 18:52 |
| josefwells | This was the only change: | 18:53 |
| josefwells | @ zuul.d/pipeline_check.yaml:18 @ | 18:53 |
| josefwells | - event: pull_request | 18:53 |
| josefwells | action: comment | 18:53 |
| josefwells | comment: (?i)^\s*recheck\s*$ | 18:53 |
| josefwells | - - event: check_run | 18:53 |
| josefwells | start: | 18:53 |
| josefwells | github: | 18:53 |
| josefwells | - check: 'in_progress' | 18:53 |
| josefwells | - comment: false | 18:53 |
| josefwells | + status: 'pending' | 18:53 |
| josefwells | + comment: true | 18:53 |
| josefwells | success: | 18:53 |
| josefwells | github: | 18:53 |
| josefwells | - check: 'success' | 18:53 |
| josefwells | - comment: false | 18:53 |
| josefwells | + status: 'success' | 18:53 |
| josefwells | + comment: true | 18:53 |
| josefwells | failure: | 18:53 |
| josefwells | github: | 18:54 |
| josefwells | - check: 'failure' | 18:54 |
| josefwells | - comment: false | 18:54 |
| josefwells | + status: 'failure' | 18:54 |
| josefwells | + comment: true | 18:54 |
| josefwells | ... | 18:54 |
| fungi | josefwells: in the future please paste to something like http://paste.openstack.org/ and just give the url | 18:54 |
| josefwells | Oh, sorry, it was a bit much | 18:57 |
| fungi | josefwells: is that an inverse diff? | 18:57 |
| josefwells | git show, + diff-so-fancy | 18:58 |
| fungi | it looks like it's switching to from the checks api to the status api rather than the other way around | 18:59 |
| josefwells | right, sorry, I switched back from checks to the status api and it fixed the re-run issue | 19:00 |
| josefwells | and status gets updated correctly, whereas checks never seemed to get a pass/fail signal | 19:00 |
| *** saneax has quit IRC | 19:01 | |
| fungi | did you find any explanation in build console output or the executor log to explain why the build was getting retried? the executor should normally log the reason for retrying | 19:02 |
| josefwells | for the re-running issue, I only saw that once I commented "recheck" in the pr. Then I saw it get "matched" against the pipeline rules repeatedly | 19:03 |
| josefwells | I'm guessing because it never gets marked as success/fail | 19:03 |
| avass | josefwells: maybe reporting somwhow triggers the pipeline? | 19:04 |
| avass | josefwells: or is it specifically executor retries and not the job running repeatedly? | 19:04 |
| fungi | josefwells: you mentioned something about being concerned your version of github enterprise might be too old, i vaguely recall it got the checks api features somewhat later than public github did but i don't remember any exact timeframe/release... tobiash might recall but it's probably getting late in the day over there | 19:06 |
| josefwells | the scheduler seems to be requesting the runs, so it doesn't seem like executor | 19:06 |
| josefwells | fungi, github public is on 3.1+? we are lagging a bit, 2.22, which is the release before 3.0 | 19:07 |
| avass | josefwells, fungi: this parts looks suspicious in the triggers, I don't have it in my config: "- event: check_run" | 19:07 |
| fungi | avass: that's supposedly to support the button github shows on failed results | 19:07 |
| avass | josefwells: can you try to reapply the config without that line in the triggers? | 19:07 |
| fungi | maybe that's a "new" feature, yeah | 19:08 |
| avass | fungi: oh huh, I haven't seen those | 19:08 |
| fungi | 'When using the checks API to report results, failed runs will have a "re-run" button which emits this event.' | 19:09 |
| avass | I think it could be triggering on check_run completed? https://docs.github.com/en/developers/webhooks-and-events/webhook-events-and-payloads#check_run | 19:09 |
| fungi | inline comment from the reference pipeline example | 19:09 |
| avass | maybe there's a missing "requirements" section in the examples? | 19:10 |
| tobiash | josefwells: the check_run trigger is too wide | 19:11 |
| tobiash | that will trigger on any check run event including the report of zuul itself | 19:11 |
| tobiash | hence the infinite loop | 19:11 |
| fungi | oh neat, so the example we have in the docs is waaaay wrong ;) | 19:12 |
| josefwells | yeah ;) | 19:12 |
| josefwells | hey, I'm helping! Yaya | 19:12 |
| openstackgerrit | Albin Vass proposed zuul/zuul master: Add 'action: rerequested' to example github pipelines https://review.opendev.org/c/zuul/zuul/+/790807 | 19:12 |
| avass | tobiash: ^ ? | 19:12 |
| *** yoctozepto has quit IRC | 19:13 | |
| josefwells | could be that the checks updates did indeed work (as the log suggested) but it immediately triggered a rerun or otherwise cleared the status | 19:13 |
| *** yoctozepto6 has joined #zuul | 19:14 | |
| fungi | that sounds like a reasonable theory | 19:14 |
| josefwells | I didn't see a re-run loop until I commented "recheck", when it was just PR open, I got 2x runs, then nothing | 19:14 |
| tobiash | avass: yes that's missing | 19:14 |
| josefwells | anyway, lemme try the above PR suggestion and see if that fixes it | 19:14 |
| avass | tobiash: good :) | 19:14 |
| tobiash | commented, we have another filter on our pipeline definition | 19:14 |
| tobiash | this is our full check pipeline definition: http://paste.openstack.org/show/805253/ | 19:15 |
| openstackgerrit | Albin Vass proposed zuul/zuul master: Add 'action: rerequested' to example github pipelines https://review.opendev.org/c/zuul/zuul/+/790807 | 19:15 |
| corvus | tobiash: https://zuul-ci.org/docs/zuul/discussion/github-checks-api.html mentions 'rerequested' but your def only has 'requested' (plus check) | 19:16 |
| avass | corvus: oh interesting | 19:17 |
| tobiash | corvus: hrm, I'm pretty sure that my re-run button works so that might be a bug in the docs as well | 19:17 |
| fungi | interestingly, i find several examples of us not tightening it in opendev, like https://opendev.org/pyca/project-config/src/branch/master/zuul.d/pipelines.yaml#L37 | 19:17 |
| corvus | omg | 19:21 |
| tobiash | corvus, avass: zuul normalizes the event to 'requested' so it must be 'requested' in the pipeline definition: https://opendev.org/zuul/zuul/src/branch/master/zuul/driver/github/githubconnection.py#L579 | 19:21 |
| corvus | yeah that was what i was about to say :( | 19:21 |
| corvus | we should really try to never do that in drivers | 19:22 |
| corvus | because it leads to 4 senior developers spending a bunch of time trying to figure out why the program doesn't match the docs | 19:22 |
| corvus | the general idea in drivers is input=output, so we don't have to write docs that duplicate upstream | 19:23 |
| tobiash | any idea how to fix that in a compatible way? | 19:24 |
| corvus | tobiash: we might be able to have the trigger def to the opposite of that line (so 'requested' -> 'rerequested') at the same time we remove that line. then tell everyone to change requested -> rerequested in pipeline defs before 5.0, then remove the compat | 19:25 |
| corvus | and hopefully all that happens before github expands their api to add a 'requested' action ;) | 19:27 |
| josefwells | haha, yeah, indeed rerequested doesn't work today, but putting any restrition on "check_run" fixes the reporting and the infinite loop, Thanks! | 19:28 |
| corvus | josefwells: in short, use http://paste.openstack.org/show/805253/ as a reference right now, and expect a deprecation to change the spelling to 'rerequested' in the near future. | 19:29 |
| josefwells | yeah, I can dig it | 19:29 |
| corvus | josefwells: thanks for the find :) | 19:29 |
| avass | we could also make action a regex and tell people to check ^(re)?requested$ :) | 19:29 |
| josefwells | wait till I get going ;) | 19:29 |
| corvus | avass: could, but that could break if github adds a 'requested' action | 19:30 |
| corvus | which, if we step back a second, is not at all far-fetched :) | 19:30 |
| openstackgerrit | Albin Vass proposed zuul/zuul master: Add 'action: rerequested' to example github pipelines https://review.opendev.org/c/zuul/zuul/+/790807 | 19:32 |
| avass | err | 19:32 |
| openstackgerrit | Albin Vass proposed zuul/zuul master: Add 'action: requested' to example github pipelines https://review.opendev.org/c/zuul/zuul/+/790807 | 19:33 |
| avass | fungi: no the question is why opendev doesn't retrigger itself in github | 19:33 |
| avass | now* | 19:33 |
| avass | fungi: missing checks permissions? | 19:36 |
| avass | ah or zuul isn't subscribed to check run events | 19:37 |
| openstackgerrit | Tobias Henkel proposed zuul/zuul master: Remove translation of check run action in github driver https://review.opendev.org/c/zuul/zuul/+/790812 | 19:44 |
| tobiash | corvus: how does that look? ^ | 19:45 |
| tobiash | feel free to improve the deprecation message, you're better than me on that ;) | 19:45 |
| josefwells | I don't get a re-run button in the github checks page unless the check failed. Sounds like a github issue, not zuul, but FYI and maybe work the a mention in the docs (and may be tied to my github enterprise version) | 19:47 |
| tobiash | josefwells: that's a github limitation | 19:47 |
| josefwells | tobiash, figures, thanks for the confirm | 19:47 |
| openstackgerrit | Tobias Henkel proposed zuul/zuul master: Remove translation of check run action in github driver https://review.opendev.org/c/zuul/zuul/+/790812 | 19:50 |
| avass | tobiash: wanna squash 790807 with 790812?.. oh and put rerequested instead of requested :) | 20:06 |
| tobiash | avass: can do | 20:08 |
| *** hamalq has quit IRC | 20:09 | |
| *** dcastellani has quit IRC | 20:09 | |
| *** erbarr has quit IRC | 20:11 | |
| *** mordred has quit IRC | 20:11 | |
| fungi | avass: it's possible we have missing permissions, opendev isn't gating any repos on github, we just perform advisory reporting and auxiliary architecture-specific builds for some of our projects' dependencies hosted in github | 20:11 |
| openstackgerrit | Tobias Henkel proposed zuul/zuul master: Remove translation of check run action in github driver https://review.opendev.org/c/zuul/zuul/+/790812 | 20:11 |
| josefwells | Ok, another odd issue.. if I "abort", the check doesn't get set. which leads to a bit of a stalemate. I couldn't trigger a rerun with "recheck" and needed to push an additional commit | 20:11 |
| avass | fungi: I think it's more likely that opendevs zuul just isn't subscribed to those events, otherwise zuul shouldn't be able to report any checks | 20:12 |
| fungi | josefwells: you could make your pipeline also trigger on specific comments | 20:12 |
| josefwells | https://zuul-ci.org/docs/zuul/reference/drivers/github.html refers to the github check api status of "cancelled" but zuul doesn't have a pipeline for that | 20:12 |
| tobiash | josefwells: check out the dequeue reportin in http://paste.openstack.org/show/805253/ | 20:12 |
| josefwells | tobiash: thx | 20:13 |
| tobiash | I guess you meant is zuul aborts the check run isn't aborted in github? | 20:13 |
| tobiash | or the other way round? | 20:13 |
| *** hamalq has joined #zuul | 20:13 | |
| *** dcastellani has joined #zuul | 20:13 | |
| *** erbarr has joined #zuul | 20:13 | |
| *** mordred has joined #zuul | 20:13 | |
| corvus | tobiash: 812 lgtm | 20:14 |
| corvus | tobiash: https://review.opendev.org/790798 is a quick fix for a bug seen in the wild | 20:14 |
| openstackgerrit | Tobias Henkel proposed zuul/zuul master: Add dequeue reporting to the github example pipelines https://review.opendev.org/c/zuul/zuul/+/790816 | 20:16 |
| corvus | tristanC: do you have a minute to look at https://review.opendev.org/790508 -- related to tenant config which i know is near and dear to you :) | 20:16 |
| *** mordred has quit IRC | 20:16 | |
| *** hasharAway has quit IRC | 20:27 | |
| josefwells | tobiash: I meant if you abort from the magic github button, and you don't have the dequeue set up, the checkapi never gets the update and still thinks it is "in_progress" so it won't run again, etc. The pastebin you posted worked perfectly for me: http://paste.openstack.org/show/805253/. I *almost* had it myself, but I was trying "{ cancelled: { check: "cancelled", comment: false }}" instead of "{ dequeue: { check: "cancelled", | 20:33 |
| josefwells | comment: false }}" I see the docs here, so I understand better now: https://zuul-ci.org/docs/zuul/reference/pipeline_def.html | 20:33 |
| fungi | josefwells: so https://review.opendev.org/790816 would have helped? | 20:34 |
| josefwells | fungi: yes, very much so | 20:36 |
| josefwells | fungi: and thank you for indirectly confirming that I am making the right changes to my gate pipeline as well ;) | 20:36 |
| fungi | cool, thanks again for helping find the holes in our examples! | 20:37 |
| *** mordred has joined #zuul | 20:41 | |
| josefwells | fungi: I noticed what I think is duplication in the reference gate pipeline: https://review.opendev.org/plugins/gitiles/zuul/zuul/+/refs/changes/16/790816/1/doc/source/examples/pipelines/github-reference-pipelines.yaml lines 77-79 vs. 83-85 | 20:42 |
| openstackgerrit | Albin Vass proposed zuul/zuul master: remove duplicate trigger from github example pipelines https://review.opendev.org/c/zuul/zuul/+/790823 | 20:44 |
| avass | josefwells: ^ | 20:44 |
| avass | :) | 20:45 |
| josefwells | avass: thanks! | 20:45 |
| avass | oh that diff in gerrit is slightly confusing | 20:45 |
| josefwells | This work-from-home thing is going to make actually contributing to zuul easier.. but I'm still learning.. and I have to run it by my companies lawyers (hint, largest law-firm in the US) | 20:47 |
| fungi | josefwells: fingers crossed you get the okay! we love help | 20:49 |
| *** mordred has quit IRC | 20:57 | |
| openstackgerrit | Merged zuul/zuul master: Check if statsd is set before using it https://review.opendev.org/c/zuul/zuul/+/790798 | 21:09 |
| *** Shrews has quit IRC | 21:09 | |
| fungi | anybody recall what the situation is with the zuul-nodepool-functional job? i know it's non-voting, but it hasn't succeeded for nearly 3 months now. is anyone aware of work in progress to fix it? | 21:11 |
| fungi | er, nodepool-zuul-functional | 21:12 |
| corvus | fungi: not that i recall | 21:15 |
| fungi | looks like it fails waiting for zk to come ready | 21:15 |
| fungi | i wonder if it was never updated for mandatory tls | 21:21 |
| *** mordred has joined #zuul | 21:26 | |
| fungi | looks like it's still using 2181/tcp | 21:30 |
| fungi | https://opendev.org/zuul/zuul/src/branch/master/tests/nodepool/test_nodepool_integration.py#L34 | 21:34 |
| fungi | but the base test case updated to use 2281 around the time that job stopped working, with https://review.opendev.org/777489 | 21:35 |
| fungi | i'll see if the fix is as simple as i think | 21:35 |
| openstackgerrit | Jeremy Stanley proposed zuul/zuul master: Use ZK config for nodepool integration test https://review.opendev.org/c/zuul/zuul/+/790826 | 21:42 |
| fungi | no idea if that's the way to do it, mostly cribbed from another test change around that timeframe | 21:42 |
| ianw | if someone has a second to look over https://review.opendev.org/q/topic:%22nodepool-func-podman-mount that would be good. it's a few changes to the nodepool podman install/config to make containerfile dib jobs work | 21:51 |
| ianw | with it, i have a fedora 34 containerfile based job booting -> https://zuul.opendev.org/t/openstack/build/8d7b05f13ba4400dbc5b8edfde552a72 | 21:51 |
| openstackgerrit | Merged zuul/zuul master: Remove translation of check run action in github driver https://review.opendev.org/c/zuul/zuul/+/790812 | 22:53 |
| openstackgerrit | Merged zuul/zuul master: Add dequeue reporting to the github example pipelines https://review.opendev.org/c/zuul/zuul/+/790816 | 22:54 |
| *** tosky has quit IRC | 23:24 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!