[
https://issues.apache.org/jira/browse/BEAM-6011?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16681558#comment-16681558
]
Etienne Chauchot commented on BEAM-6011:
----------------------------------------
I agree that phrase triggering is needed on PRs for nexmark. Storing to BQ the
results is also needed because it will allow us to track
performance/functionnal regression before merging a PR allowing us to avoid
rollbacks in case of problems. Of course, saving to other dataset is needed
also to avoid pollution of the graphs that relate perfs on master. Avoiding
pollution was the original argument to avoid phrase triggering in the first
place, but write elsewhere is better.
Besides, I think it is a lot of work to create the dashboards for the new
dataset (I made more than 100 graphs at the time) and it is not needed, having
a jenkins script that shows the output of the nexmark run triggered (by
selecting in perfs and output size tables) is enough IMHO.
+1 to do the same with IOs
> Enable Phrase triggering in Nexmark jobs
> ----------------------------------------
>
> Key: BEAM-6011
> URL: https://issues.apache.org/jira/browse/BEAM-6011
> Project: Beam
> Issue Type: Bug
> Components: testing
> Reporter: Lukasz Gajowy
> Assignee: Lukasz Gajowy
> Priority: Critical
>
> We need to enable Phrase Triggering (running Jenkins jobs from PR) for
> Nexmark jobs so that we could check if pull requests are not breaking
> anything before merging them.
> *Note*: Currently Nexmark jobs run post commit on master and publish their
> results to BigQuery database. In order not to pollute the results collected
> for master we should save the results for Pr-triggered jobs in some other
> tables/datasets or even not save them at all (turn off publishing to BQ).
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
