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

tests: update partition limits #9257

Merged
merged 2 commits into from
Mar 6, 2023

Conversation

jcsp
Copy link
Contributor

@jcsp jcsp commented Mar 3, 2023

Change 3e3975b caused some scale tests to fail https://ci-artifacts.dev.vectorized.cloud/vtools/0186a681-02da-49ac-8a24-e33caa78970f/vbuild/ducktape/results/2023-03-03--001/report.html

These limits need updating so that the tests operate within the supported range. ManyPartitionsTest was only out by a few, partition balancer test was further out.

Backports Required

  • none - not a bug fix
  • none - this is a backport
  • none - issue does not exist in previous branches
  • none - papercut/not impactful enough to backport
  • v23.1.x
  • v22.3.x
  • v22.2.x

Release Notes

  • none

jcsp added 2 commits March 3, 2023 19:03
In 3e3975b the
default topic_partitions_per_shard changed to 1000: the
test was previously working because the configured limit
was so much higher than the test's limit.

The internal_partition_slack value was probably
set before we changed the default number of consumer
offsets partitions to 16.
This was exceeding the 1000 partitions per core limit.  It
was getting away with it because it is a relatively low
traffic test compared with ManyPartitionsTest.

Revise its partition count to fit within the supported max.
@jcsp jcsp added kind/bug Something isn't working area/tests labels Mar 3, 2023
@jcsp
Copy link
Contributor Author

jcsp commented Mar 3, 2023

/cdt tests/rptest/scale_tests/many_partitions_test.py tests/rptest/scale_tests/partition_balancer_scale_test.py

Copy link
Contributor

@andrwng andrwng left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The cdt runs seemed to have trouble kicking off -- I retried the regular build, and maybe we can retry the cdt run once that passes (assuming the issue is with ordering). Opened https://github.com/redpanda-data/vtools/issues/1561 for it

Comment on lines +187 to +188
partitions_count = 1000 * self.redpanda.get_node_cpu_count() * len(
self.redpanda.nodes) - 32
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Shouldn't we be dividing by the replication factor?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Sorry, itchy merge finger -- I think was looking at a github view that had updated test statuses but not shown comments.

You are right! Followup PR here #9274

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
area/tests kind/bug Something isn't working
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants