Skip to content

Conversation

@Shivs11
Copy link
Member

@Shivs11 Shivs11 commented Dec 23, 2025

What changed?

  • WISOTT
  • To be a bit more descriptive: This PR attributes a portion of the unversioned backlog when a current/ramping version is set. However, I had forgotten to properly consider the case when someone ramps to unversioned.

Why?

  • Correctness.

How did you test it?

  • built
  • run locally and tested manually
  • covered by existing tests
  • added new unit test(s)
  • added new functional test(s)

Potential risks

  • None

Note

Ensures task queue stats correctly handle ramping to the unversioned target.

  • Adjusts Describe attribution in task_queue_partition_manager.go to use isRamping from CalculateTaskQueueVersioningInfo and avoid subtracting ramp share from unversioned when ramping target is unversioned; current share still absorbed appropriately
  • Adds functional tests validating current absorbs (100 - ramp%) of unversioned backlog when ramping to unversioned (single/multi-partition) and keeps legacy stats reporting total backlog
  • Adds unit test in worker_versioning_test.go for mixed new-format data where current is versioned and ramping is unversioned
  • Minor test naming/label adjustments

Written by Cursor Bugbot for commit 6cdc60e. This will update automatically on new commits. Configure here.

@Shivs11 Shivs11 marked this pull request as ready for review December 23, 2025 20:53
@Shivs11 Shivs11 requested review from a team as code owners December 23, 2025 20:53
@Shivs11 Shivs11 merged commit 6189c1e into ss/current-ramping-unversioned-tasks Dec 29, 2025
95 of 100 checks passed
@Shivs11 Shivs11 deleted the ss/unversioned-special-case-stats branch December 29, 2025 21:32
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.

3 participants