Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

ci: enable new source reclocking strategy #30031

Draft
wants to merge 2 commits into
base: main
Choose a base branch
from

Conversation

teskje
Copy link
Contributor

@teskje teskje commented Oct 16, 2024

This PR enables the new source reclocking scheme ("reclock to latest upper", RLU) in CI, in preparation for switching it on in production too.

Motivation

  • This PR adds a known-desirable feature.

Part of https://github.com/MaterializeInc/database-issues/issues/7020

Tips for reviewer

Checklist

  • This PR has adequate test coverage / QA involvement has been duly considered. (trigger-ci for additional test/nightly runs)
  • This PR has an associated up-to-date design doc, is a design doc (template), or is sufficiently small to not require a design.
  • If this PR evolves an existing $T ⇔ Proto$T mapping (possibly in a backwards-incompatible way), then it is tagged with a T-proto label.
  • If this PR will require changes to cloud orchestration or tests, there is a companion cloud PR to account for those changes that is tagged with the release-blocker label (example).
  • If this PR includes major user-facing behavior changes, I have pinged the relevant PM to schedule a changelog post.

@teskje teskje force-pushed the ci-use-latest-reclocking branch 3 times, most recently from a1a680b to 2a4f969 Compare October 28, 2024 09:54
@teskje teskje force-pushed the ci-use-latest-reclocking branch 4 times, most recently from 7696bf2 to b73a85e Compare November 8, 2024 13:24
@teskje teskje force-pushed the ci-use-latest-reclocking branch 2 times, most recently from 7894eb5 to 83d21ab Compare November 11, 2024 12:58
@teskje teskje force-pushed the ci-use-latest-reclocking branch 7 times, most recently from 53517f8 to ee439b0 Compare November 14, 2024 13:15
@benesch benesch force-pushed the ci-use-latest-reclocking branch from ee439b0 to 9225142 Compare November 14, 2024 17:01
@benesch
Copy link
Member

benesch commented Nov 14, 2024

I twiddled the instantiation of the metadata Kafka client. A bit of a hail mary, but maybe it improves something.

@benesch
Copy link
Member

benesch commented Nov 14, 2024

I was having trouble understanding which commit was causing which test failures, so I split this out into two separate PRs and triggered nightly on both:

@teskje teskje force-pushed the ci-use-latest-reclocking branch 2 times, most recently from 804ae60 to ab45a65 Compare December 4, 2024 10:30
@teskje teskje force-pushed the ci-use-latest-reclocking branch 4 times, most recently from 180363f to 30a8b09 Compare December 20, 2024 11:41
@teskje teskje force-pushed the ci-use-latest-reclocking branch 2 times, most recently from c319c57 to c7835c5 Compare January 8, 2025 14:21
@teskje teskje force-pushed the ci-use-latest-reclocking branch from c7835c5 to 4dde088 Compare January 9, 2025 09:41
A bunch of tests set the `kafka_default_metadata_fetch_interval` or the
source `TOPIC METADATA REFRESH INTERVAL` to low values. This was done
because the default for `kafka_default_metadata_fetch_interval` was 30s
and some tests were slowed down by that. The default has now been
changed to 1s, so manually lowering the fetch interval is not necessary
anymore.

Removing the manual setting of the Kafka metadata interval in tests gets
rid of some noise. But more importantly, it fixes the issue of some
tests setting the interval to a very low value, like '10ms', which now
that the metadata fetch interval influences the source tick frequency
could negatively impact performance and make tests slower or even time
out.
@teskje teskje force-pushed the ci-use-latest-reclocking branch 2 times, most recently from 14b40f4 to 1293981 Compare January 13, 2025 17:05
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants