Repository navigation
2.0: concurrent-ensure benchmark for missing-service reconnect herd #38569
Description
Parent: #36285
Problem
Reconnect-herd mitigations need a focused missing-service reproducer before landing policy.
Existing coverage checks jitter bounds and concurrent reuse of an already-healthy endpoint.
It does not measure contender count or recovery latency when several clients call Service.ensure against a missing service.
Lifecycle logs also show startup windows where a short reconnect spread would not cover recovery.
Expected
A small benchmark/harness that reports, for N concurrent ensures against an absent registration:
- contender launch count
- recovery latency (first / all healthy)
No reconnect policy change required in the first cut — baseline current Service.ensure behavior so later PRs can keep only changes that reduce launches without delaying single-contender recovery.
Metadata
Metadata
Assignees
Labels
No labels
Activity
github-actions commentedon Jul 23, 2026
This issue doesn't fully meet our contributing guidelines.
What needs to be fixed:
[FEATURE]:) would be most appropriate.Please edit this issue to address the above within 2 hours, or it will be automatically closed.
This issue might be related to existing issues. Please check:
If you believe this was flagged incorrectly, please let a maintainer know.
github-actions commentedon Jul 23, 2026
This issue has been automatically closed because it was not updated to meet our contributing guidelines within the 2-hour window.
Feel free to open a new issue that follows our issue templates.