@vaughnwalters what's requested from the catalyst team is confirmation we can support this workload and a catalyst API key to use for this, correct? Is there anything else?

In T388056#10607498, @jeena wrote:

@vaughnwalters what's requested from the catalyst team is confirmation we can support this workload and a catalyst API key to use for this, correct?

@jeena Yes, this is correct.

Is there anything else?

I believe that's all that is needed from the Catalyst team. @Jdforrester-WMF please correct me if you think of anything I am missing here.

@vaughnwalters @Jdforrester-WMF re: the Catalyst token, someone already created one for AW early this year:

{"id": 3, "description": "abstract-wikipedia-ci", "createdAt": "2025-01-07T21:22:06.337Z"}

Would you happen to know if someone in your team still has it? Otherwise we can of course issue a new token and revoke the existing one

Added T383065 as a blocker for this on a hunch that whatever we build there will be re-used for a nightly run (I think). At the end of T383065, we'll be able to run selenium tests for new patches to Catalyst, if we want to run daily tests then that will involve much of the same work. We'll talk more among Catalyst folks to see if that's correct.

Cunningham's law test: I think the value of daily tests is that selenium tests run with WikiLambda alongside its dependent extensions and therefore if there is a breaking change in a dependent extension WikiLambda developers are notified EVEN IF WikiLambda has not pushed a patch.

...is that right?

In T388056#10614579, @thcipriani wrote:

Cunningham's law test: I think the value of daily tests is that selenium tests run with WikiLambda alongside its dependent extensions and therefore if there is a breaking change in a dependent extension WikiLambda developers are notified EVEN IF WikiLambda has not pushed a patch.

...is that right?

Yes. We don't care whose fault it is as much as that things are broken in tests, and plausibly thus for users when the train next rolls.

In T388056#10609012, @jnuche wrote:

@vaughnwalters @Jdforrester-WMF re: the Catalyst token, someone already created one for AW early this year:

{"id": 3, "description": "abstract-wikipedia-ci", "createdAt": "2025-01-07T21:22:06.337Z"}

Would you happen to know if someone in your team still has it? Otherwise we can of course issue a new token and revoke the existing one

@jnuche I asked the team, and nobody said they have it. We should just issue a new token and revoke the old one.

@jnuche I asked the team, and nobody said they have it. We should just issue a new token and revoke the old one.

@vaughnwalters ack, np, let me (or anyone else in the Catalyst team) know once the new token is needed and we'll do that

AIUI this is now un-stalled?

Change #1171538 had a related patch set uploaded (by Jeena Huneidi; author: Jeena Huneidi):

[integration/config@master] Add daily trigger for wikilambda selenium tests

https://gerrit.wikimedia.org/r/1171538

Change #1171538 merged by jenkins-bot:

[integration/config@master] jjb: [wikilambda-catalyst-end-to-end] Also trigger daily

https://gerrit.wikimedia.org/r/1171538

@Jdforrester-WMF I am not sure how to tell if this has run daily or not. Is there a way to designate in a failure report (if there is one) that it came from the daily run, or to note its success? Right now if I get a failure notice I am not sure if it is from a daily run, wikilambda run, or orchestrator run.

Change #1176507 had a related patch set uploaded (by Hashar; author: Hashar):

[integration/config@master] jjb: create specific job for Wikilambda daily tests

https://gerrit.wikimedia.org/r/1176507

Change #1177387 had a related patch set uploaded (by Hashar; author: Hashar):

[integration/config@master] jjb: stop triggering Wikilambda end-to-end daily

https://gerrit.wikimedia.org/r/1177387

Change #1177388 had a related patch set uploaded (by Hashar; author: Hashar):

[integration/config@master] jjb: wikilambda-catalyst-end-to-end does not use Castor

https://gerrit.wikimedia.org/r/1177388

Change #1177393 had a related patch set uploaded (by Hashar; author: Hashar):

[integration/config@master] dockerfiles: [catalyst] allow setting environment name

https://gerrit.wikimedia.org/r/1177393

Change #1177394 had a related patch set uploaded (by Hashar; author: Hashar):

[integration/config@master] jjb: update wikilambda-catalyst-end-to-end image

https://gerrit.wikimedia.org/r/1177394

Change #1177387 merged by jenkins-bot:

[integration/config@master] jjb: [wikilambda-catalyst-end-to-end] Stop triggering daily

https://gerrit.wikimedia.org/r/1177387

Change #1177388 merged by jenkins-bot:

[integration/config@master] jjb: [wikilambda-catalyst-end-to-end] Don't set up with Castor, unused

https://gerrit.wikimedia.org/r/1177388

Change #1177393 merged by jenkins-bot:

[integration/config@master] dockerfiles: [catalyst] Allow setting environment name

https://gerrit.wikimedia.org/r/1177393

Mentioned in SAL (#wikimedia-releng) [2025-08-11T15:26:28Z] <James_F> Docker: [catalyst] Allow setting environment name, for T388056

Change #1177394 merged by jenkins-bot:

[integration/config@master] jjb: [wikilambda-catalyst-end-to-end] Update to new Catalyst image

https://gerrit.wikimedia.org/r/1177394

Change #1176507 merged by jenkins-bot:

[integration/config@master] jjb: Create specific job for Wikilambda daily tests

https://gerrit.wikimedia.org/r/1176507

OK, reactive and daily scheduled tasks are now split. @vaughnwalters, I should have triggered both in the past few minutes — did the failures come through, and were they distinguishable?

Yes thank you, I received both (one from WikiLambda, one from daily run) and they are distinguishable. Will this also allow me to distinguish between orchestrator and (when it lands) evaluator catalyst / selenium failures as well or is that a different scope?

wikilambda	daily run

In T388056#11075208, @vaughnwalters wrote:

Yes thank you, I received both (one from WikiLambda, one from daily run) and they are distinguishable. Will this also allow me to distinguish between orchestrator and (when it lands) evaluator catalyst / selenium failures as well or is that a different scope?

The orchestrator and evaluator tests run on gitlab so you won't see them on jenkins. If you are subscribed to the MR the test run belongs to then I think you would get an email for failures.

Change #1179783 had a related patch set uploaded (by Jeena Huneidi; author: Jeena Huneidi):

[integration/config@master] Update wikilambda daily selenium job

https://gerrit.wikimedia.org/r/1179783

Change #1179783 merged by jenkins-bot:

[integration/config@master] Update wikilambda daily selenium job

https://gerrit.wikimedia.org/r/1179783

Heya @thcipriani there are currently two wikilambda catalyst end to end daily jobs that are running, this one which passed last night, and this one which is still failing.

Can we just remove the failing one before calling this resolved? As for the successful one, awesome! I will keep an eye on it for the next few weeks to make sure it stays green. 🎉

In T388056#11129885, @vaughnwalters wrote:

Heya @thcipriani there are currently two wikilambda catalyst end to end daily jobs that are running, this one which passed last night, and this one which is still failing.

Can we just remove the failing one before calling this resolved? As for the successful one, awesome! I will keep an eye on it for the next few weeks to make sure it stays green. 🎉

Yep, @jeena got that one removed. Another successful build for the new one last night.