[ENG-3964]: Set worker-thunderbolt to process pool#333
Open
fpopa wants to merge 1 commit into
Open
Conversation
Brings worker-thunderbolt in line with the other long-LLM workers (worker-compute/highmem/storage[-high]/monitors), which all run the process pool to avoid GIL contention starving the heartbeat thread under heavy concurrent agent dispatches. Saas has been running this override since 2026-05-20 without regression. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
dmigo
approved these changes
May 26, 2026
gtoonstra
reviewed
May 26, 2026
gtoonstra
approved these changes
May 26, 2026
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
3 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
worker-thunderbolt.temporal.workerPoolTypefromthreadtoprocessincharts/datafold/values.yaml, matching the other long-LLM workers (worker-compute,worker-highmem,worker-storage[-high],worker-monitors).charts/datafold/Chart.yamlfrom0.10.89to0.10.90.Why
ENG-3935 surfaced heartbeat timeouts on
worker-thunderboltunder heavy concurrent agent dispatches — GIL contention inthreadpool mode starves the heartbeat daemon thread. Sibling long-LLM workers all useprocesspool, where the SDK'sSharedHeartbeatSenderroutes heartbeats via a cross-process queue and is decoupled from any one process's GIL.worker-thunderboltwas inadvertently left onthread.A live saas override (
spec.components.worker-thunderbolt.rawValues.temporal.workerPoolType: "process") has been applied since 2026-05-20 with no regressions over the bake-in window.Once this is merged and the operator picks it up, the per-cluster
rawValuesoverride on saas (and ecolab, if applied) can be dropped.Ref: ENG-3964, ENG-3935, #331.
Test plan
rawValues.temporal.workerPoolTypeoverride on saas (and ecolab if applicable).🤖 Generated with Claude Code