feat(ray): Add actor resilience with zombie detection and lease recovery#1408
Draft
jioffe502 wants to merge 1 commit intoNVIDIA:mainfrom
Draft
feat(ray): Add actor resilience with zombie detection and lease recovery#1408jioffe502 wants to merge 1 commit intoNVIDIA:mainfrom
jioffe502 wants to merge 1 commit intoNVIDIA:mainfrom
Conversation
Adds defense-in-depth for actor failures: - Zombie detection: stat collector detects dead processing threads and quarantines actors after 3 consecutive unhealthy cycles - Health probes: actors expose processing_thread_alive in get_stats() and a lightweight health_probe() method - Readiness verification: scale-up retries actors that fail readiness checks instead of admitting broken actors to the scheduling pool - Lease-based job recovery: source claims leases on dequeued jobs; sink atomically pushes results and acks leases; expired leases are periodically swept and requeued - Min-replica enforcement: _reconcile_min_replicas runs even when dynamic scaling is disabled to replace quarantined actors Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> Signed-off-by: Jacob Ioffe <jioffe@nvidia.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Description
Defense-in-depth so the pipeline self-heals when any actor dies (OOM, native crash, segfault) instead of silently black-holing jobs and hanging:
discovered from debugging ray zombies in regression that #1379 fixes.
Checklist