[
https://issues.apache.org/jira/browse/BEAM-14226?focusedWorklogId=753729&page=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-753729
]
ASF GitHub Bot logged work on BEAM-14226:
-----------------------------------------
Author: ASF GitHub Bot
Created on: 06/Apr/22 22:11
Start Date: 06/Apr/22 22:11
Worklog Time Spent: 10m
Work Description: ibzib commented on PR #17239:
URL: https://github.com/apache/beam/pull/17239#issuecomment-1090864191
> Perhaps the seed job just doesn't create itself any more? That would
probably be smart.
Whoa, judging from the logs the seed job does generate itself. That
definitely doesn't seem ideal.
> GeneratedJob{name='beam_SeedJob'}
> GeneratedJob{name='beam_SeedJob_Standalone'}
> Ah, and specifically, the seed job shared "library" code with all the
other jobs, so it could be broken by breakage in the shared groovy files.
At least it looks like it doesn't share code with other jobs anymore. The
only thing it imports is Committers.groovy, which is just a placeholder.
https://github.com/apache/beam/blob/f90379a67f9cbc89425cd161452cdf6172354e93/.test-infra/jenkins/job_00_seed.groovy#L21
> I expect it is out of date as well.
It's only trivially different.
```diff
$ diff .test-infra/jenkins/job_seed_standalone.groovy
.test-infra/jenkins/job_00_seed.groovy
23c23
< job('beam_SeedJob_Standalone') {
—
> job('beam_SeedJob') {
77c77
< cron('0 */5 * * *')
—
> cron('H */6 * * *')
85c85
< triggerPhrase('Run Standalone Seed Job')
—
> triggerPhrase('Run Seed Job')
90c90
< context("Jenkins: Standalone Seed Job")
—
> context("Jenkins: Seed Job")
110c110
< python3.8 -m venv ve3 && source ./ve3/bin/activate &&
—
> python3.8 -m venv ve3 && source ve3/bin/activate &&
```
Issue Time Tracking
-------------------
Worklog Id: (was: 753729)
Time Spent: 50m (was: 40m)
> Remove standalone seed job?
> ---------------------------
>
> Key: BEAM-14226
> URL: https://issues.apache.org/jira/browse/BEAM-14226
> Project: Beam
> Issue Type: Improvement
> Components: testing
> Reporter: Kyle Weaver
> Priority: P3
> Time Spent: 50m
> Remaining Estimate: 0h
>
> I'm not sure what the historical reason for having two separate seed jobs is,
> but it looks like today the two are almost identical, so we can probably
> delete one of them.
> $ diff .test-infra/jenkins/job_seed_standalone.groovy
> .test-infra/jenkins/job_00_seed.groovy
> 23c23
> < job('beam_SeedJob_Standalone') {
> ---
> > job('beam_SeedJob') {
> 77c77
> < cron('0 */5 * * *')
> ---
> > cron('H */6 * * *')
> 85c85
> < triggerPhrase('Run Standalone Seed Job')
> ---
> > triggerPhrase('Run Seed Job')
> 90c90
> < context("Jenkins: Standalone Seed Job")
> ---
> > context("Jenkins: Seed Job")
> 110c110
> < python3.8 -m venv ve3 && source ./ve3/bin/activate &&
> ---
> > python3.8 -m venv ve3 && source ve3/bin/activate &&
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
