-@gerrit:opendev.org- Ian Wienand proposed: [zuul/nodepool] 872083: Revert "Update git" https://review.opendev.org/c/zuul/nodepool/+/872083 | 01:31 | |
-@gerrit:opendev.org- Ian Wienand proposed: [zuul/zuul] 872092: Revert "Update git" https://review.opendev.org/c/zuul/zuul/+/872092 | 04:28 | |
-@gerrit:opendev.org- Tobias Henkel proposed: [zuul/zuul-jobs] 872104: Fix creating bucket in s3 log upload https://review.opendev.org/c/zuul/zuul-jobs/+/872104 | 09:53 | |
-@gerrit:opendev.org- Marvin Becker proposed wip: [zuul/nodepool] 871060: Add volumes to openshift pod workers https://review.opendev.org/c/zuul/nodepool/+/871060 | 12:42 | |
-@gerrit:opendev.org- Marvin Becker proposed wip: [zuul/nodepool] 871060: Add volumes to openshift pod workers https://review.opendev.org/c/zuul/nodepool/+/871060 | 12:47 | |
-@gerrit:opendev.org- Marvin Becker proposed wip: [zuul/nodepool] 871060: Add volumes to openshift pod workers https://review.opendev.org/c/zuul/nodepool/+/871060 | 12:48 | |
-@gerrit:opendev.org- Matthieu Huin https://matrix.to/#/@mhuin:matrix.org proposed: [zuul/zuul] 794854: Test zuul-client console-stream https://review.opendev.org/c/zuul/zuul/+/794854 | 15:16 | |
@newbie23:matrix.org | hi guys, | 16:29 |
---|---|---|
is there a way to attach the jobs to a pipeline so that they are executed in a specific order one at the time? | ||
I tried using dependencies (or soft dependencies), but that is different. | ||
If a job fail then all the following jobs will not be executed. | ||
I also found abou this https://zuul-ci.org/docs/zuul/latest/config/pipeline.html#value-pipeline.manager.serial but not sure this is what I need. | ||
@clarkb:matrix.org | newbie23: you can use a semaphore then it should execute the jobs listed in order. Note I'm not sure this is expected behavior and it could possibly change also if a job needs to be retried it is retried after the others have run so it will be out of order in that case | 17:15 |
@clarkb:matrix.org | But I think this is the closest way to achieve what you describe today as the exact functionality doesn't exist | 17:15 |
@newbie23:matrix.org | Thanks a lot Clark! I will experiment with the semaphore then. | 18:21 |
@fungicide:matrix.org | newbie23: an alternative would be to combine those individual jobs into one job, depending on what your requirements are. zuul's job definition language is designed to allow modularity and composability so that you can have one job which might by necessity have been implemented as separate jobs on some other CI/CD system | 18:45 |
-@gerrit:opendev.org- Clark Boylan proposed: [zuul/zuul] 872205: Revert "Update git" https://review.opendev.org/c/zuul/zuul/+/872205 | 19:06 | |
-@gerrit:opendev.org- Clark Boylan proposed: [zuul/nodepool] 872207: Revert "Update git" https://review.opendev.org/c/zuul/nodepool/+/872207 | 19:09 | |
-@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/zuul] 872026: Pin sqlalchemy to <2.0 and revert "Update git" https://review.opendev.org/c/zuul/zuul/+/872026 | 19:46 | |
-@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed on behalf of Clark Boylan: | 19:52 | |
- [zuul/zuul] 872034: Address SqlAlchemy Removed in 2.0 Warnings https://review.opendev.org/c/zuul/zuul/+/872034 | ||
- [zuul/zuul] 872044: Switch to sqlalchemy 2.0 https://review.opendev.org/c/zuul/zuul/+/872044 | ||
@clarkb:matrix.org | ianw: I abandoned https://review.opendev.org/c/zuul/zuul/+/872092 as it had to be squashed into 872026 | 20:31 |
@iwienand:matrix.org | ok | 20:40 |
-@gerrit:opendev.org- Zuul merged on behalf of Ian Wienand: [zuul/nodepool] 872083: Revert "Update git" https://review.opendev.org/c/zuul/nodepool/+/872083 | 20:42 | |
@clarkb:matrix.org | following up on why sqlalchemy 2.0 is not working after fixing the warnings I am able to run the failing tests locally one by one | 20:50 |
@clarkb:matrix.org | I am also able to run the same test multiple times so it very likely isn't hurting the mysql test install | 20:52 |
@iwienand:matrix.org | > <@gerrit:opendev.org> Zuul merged on behalf of Ian Wienand: [zuul/nodepool] 872083: Revert "Update git" https://review.opendev.org/c/zuul/nodepool/+/872083 | 20:52 |
hrm, it says the promote failed | ||
@iwienand:matrix.org | https://zuul.opendev.org/t/zuul/build/359612a3a3a24b0e88b001d9a60e66a1/console | 20:52 |
@jim:acmegating.com | i'll re-enqueue | 20:53 |
@iwienand:matrix.org | https://zuul.opendev.org/t/zuul/build/359612a3a3a24b0e88b001d9a60e66a1/console doesn't have any info | 20:54 |
@jim:acmegating.com | yeah, it's notoriously hard to debug due to the no_logs since it's all just docker api calls | 20:55 |
@jim:acmegating.com | failed with the same error, so someone will need to run those steps manually to figure out what's wrong | 20:55 |
@clarkb:matrix.org | I wonder if docker hub doesn't like the attestation info like skopeo didn't? | 20:58 |
@clarkb:matrix.org | though I guess we pushed to docker hub just fine in the gate | 20:58 |
-@gerrit:opendev.org- Zuul merged on behalf of James E. Blair https://matrix.to/#/@jim:acmegating.com: [zuul/nodepool] 871803: Speed up openstack server listing https://review.opendev.org/c/zuul/nodepool/+/871803 | 21:03 | |
@iwienand:matrix.org | > <@jim:acmegating.com> failed with the same error, so someone will need to run those steps manually to figure out what's wrong | 21:05 |
i can take alook | ||
@clarkb:matrix.org | I grabbed the list of 30 tests that run before the first failure under python 3.8 and told stestr to run those locally with no concurrency so they all share the same test runner and I can't reproduce the failure doing that either (was thinking maybe a test interaction) | 21:07 |
@clarkb:matrix.org | I created an autohold for the currently running py311 job since I'm completely failing at reproducing locally | 21:10 |
@iwienand:matrix.org | so there is a change_872083_latest tag, which is a start (https://hub.docker.com/layers/zuul/nodepool-builder/change_872083_latest/images/sha256-0a90f09892a8804ab05909d0e77b0bb7587d532e55763b1fdf0527c530d93dda?context=explore) | 21:11 |
@jim:acmegating.com | that should be the thing it's fetching in that step | 21:14 |
@iwienand:matrix.org | so, what is was trying to do was get the manifest from https://registry.hub.docker.com/v2/zuul/manifests/zuul/nodepool-builder/change_872083_latest. i'll just have to manually get a token to try it by hand | 21:15 |
@iwienand:matrix.org | ```token=$(curl --location --request POST 'https://hub.docker.com/v2/users/login/' \ | 21:31 |
--header 'Content-Type: application/json' --header 'Accept: application/json' \ | ||
--data-raw '{ "username": "$ZUUL_USERNAME", "password": "$ZUUL_PASSWORD" | ||
}' | jq -r .token) | ||
curl -H "Authorization: Bearer $token" -H "Accept: application/vnd.docker.distribution.manifest.list.v2+json, application/vnd.docker.distribution.manifest.v2+json" https://registry.hub.docker.com/v2/zuul/nodepool-builder/manifests/change_872083_latest | ||
``` | ||
@iwienand:matrix.org | ``` | 21:31 |
token=$(curl --location --request POST 'https://hub.docker.com/v2/users/login/' \ | ||
--header 'Content-Type: application/json' --header 'Accept: application/json' \ | ||
--data-raw '{ "username": "$ZUUL_USERNAME", "password": "$ZUUL_PASSWORD" | ||
}' | jq -r .token) | ||
curl -H "Authorization: Bearer $token" -H "Accept: application/vnd.docker.distribution.manifest.list.v2+json, application/vnd.docker.distribution.manifest.v2+json" https://registry.hub.docker.com/v2/zuul/nodepool-builder/manifests/change_872083_latest | ||
``` | ||
@iwienand:matrix.org | that gets me a token, but then a 403 ... anyone see any typos?! | 21:31 |
-@gerrit:opendev.org- Ade Lee proposed: [zuul/zuul-jobs] 866881: Add ubuntu to enable-fips role https://review.opendev.org/c/zuul/zuul-jobs/+/866881 | 21:35 | |
@iwienand:matrix.org | i think i've stuffed up the quoting. have to do school run but bib | 21:36 |
-@gerrit:opendev.org- Zuul merged on behalf of James E. Blair https://matrix.to/#/@jim:acmegating.com: [zuul/zuul] 872026: Pin sqlalchemy to <2.0 and revert "Update git" https://review.opendev.org/c/zuul/zuul/+/872026 | 21:38 | |
@jim:acmegating.com | ianw: zuul just promoted successfully | 21:39 |
@jim:acmegating.com | i will re-enqueue nodepool yet again | 21:39 |
@jim:acmegating.com | ianw: Clark nodepool failed again; that does start to point the finger at the multi-arch stuff in nodepool | 21:41 |
-@gerrit:opendev.org- Clark Boylan proposed: [zuul/zuul] 872226: DNM use mariadb instead of mysql https://review.opendev.org/c/zuul/zuul/+/872226 | 21:42 | |
@clarkb:matrix.org | corvus: ya I think it grabs the manifest to build the new multiarch image manifest | 21:42 |
@clarkb:matrix.org | the use mariadb instead of mysql change is a bit of a hunch since I noticed I have mysql 5.7 locally and can't reproduce the issues with sqla 2.0 locally but in CI we test with mysql 8.0. Can't easily downgrade that to 5.7 but can run mariadb and check | 21:43 |
@clarkb:matrix.org | wow I think maybe the mariadb swap fixes it. This is likely a sqla bug if it can't talk to mysql 8.0 anymore. But need full results back to make that call. I'll reach out to zzzeek if that pans out though | 22:02 |
@maheshbarai1234:matrix.org | > <@fungicide:matrix.org> there's no intentional file limit for the github driver in zuul, but i suppose it's possible there's some sort of bug or something changed with the github api which only affects pull requests over a certain size... keep in mind that zuul 3.19.1 is several years old (july 2020), so you may want to run a more current version of zuul in order to get the fixes for bugs we've identified in that time and changes we've had to make as external system apis have evolved. for example, here's the changes just in the drivers/github tree: https://opendev.org/zuul/zuul/commits/branch/master/zuul/driver/github | 22:04 |
Thank you for the analysis from your side! | ||
@iwienand:matrix.org | ok i think i figured out the right tokens!! | 22:31 |
@iwienand:matrix.org | ```{"errors":[{"code":"MANIFEST_UNKNOWN","message":"OCI index found, but accept header does not support OCI indexes"}]}``` | 22:31 |
@iwienand:matrix.org | ```-H 'Accept: application/vnd.docker.distribution.manifest.list.v2+json, application/vnd.docker.distribution.manifest.v2+json, application/vnd.oci.image.index.v1+json, application/vnd.oci.image.manifest.v1+json'``` does work | 22:34 |
-@gerrit:opendev.org- Ian Wienand proposed: [zuul/zuul-jobs] 872230: promote-docker-image: also accept OCI manifest formats https://review.opendev.org/c/zuul/zuul-jobs/+/872230 | 22:40 | |
@jim:acmegating.com | ianw: should we land that, re-enqueue a nodepool promote job, and revert if it fails? | 22:46 |
@iwienand:matrix.org | corvus: i think so ... maybe not revert straight away but debug the next problem (if there is one! :) | 22:47 |
@jim:acmegating.com | yeah; could always re-enqueue a zuul promote to make sure we didn't regress | 22:47 |
@iwienand:matrix.org | it seems like just sending back the same manifest with a different tag should be ok ... we don't seem to parse it or anything | 22:47 |
@jim:acmegating.com | i think the main concern is that that (obviously since it caused a 403) does change the behavior on the remote side, so it does have the potential to cause a regression | 22:48 |
@jim:acmegating.com | i +3d it | 22:49 |
@iwienand:matrix.org | ++ i'm around to keep debugging :) | 22:50 |
@iwienand:matrix.org | i'd really like to get a new builder image with a new dib! should fix rocky builds | 22:50 |
-@gerrit:opendev.org- Zuul merged on behalf of Ian Wienand: [zuul/zuul-jobs] 872230: promote-docker-image: also accept OCI manifest formats https://review.opendev.org/c/zuul/zuul-jobs/+/872230 | 23:03 | |
@iwienand:matrix.org | ^ trying now | 23:08 |
@iwienand:matrix.org | seems like it worked | 23:10 |
@iwienand:matrix.org | https://zuul.opendev.org/t/zuul/build/7e0801cbc98a45b69dd57c96fa1bfd2f | 23:10 |
@jim:acmegating.com | ianw: woot, thanks! | 23:11 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!