fix(fleet): bound per-tick heartbeat and memory sampling load - #6443
Merged
Merged
Conversation
Four Fleet manager lifecycle tests failed on macOS CI under full-suite
load ("first Fleet attempt never started", "fake worker never
started", managers never converging) and passed alone from the same CI
binary in under 1.2 s. In the concurrent-managers failure the standby
executor held the worker, yet the child's first shell builtin never ran
within 15 s. Timeouts were already raised from 5 s to 15 s without
effect, so this does not raise them again.
Product side: every scheduler tick appended a durable heartbeat per
leased task (sync_data, which is F_FULLFSYNC on macOS) and spawned `ps`
to sample worker memory. At the tests' 10 ms tick that is ~100
full-drive flushes and ~100 process spawns a second per test, against a
300 s stale window. Heartbeats are now skipped when the worker already
has one stamped this second (timestamps are whole-second), and memory
is sampled at most once a second per worker, keeping the last value.
Test side: the fleet::manager module runs in a max-threads = 1 nextest
group, the same remedy as the exec-persistent-service tests (#5355).
Not established: that this is the whole cause. Late spawn vs. late
child start was not distinguished; acceptance is a green loaded hosted
macOS run.
Evidence: 431 passed, 0 failed (12,828 skipped) on the fleet:: selection;
`nextest show-config test-groups` lists the manager tests in the new
group. TUI all-target/all-feature Clippy with CI flags and fmt passed.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
This was referenced Sep 23, 2026
Hmbown
pushed a commit
to gaord/CodeWhale
that referenced
this pull request
Sep 24, 2026
Version drift failed on Hmbown#6431, Hmbown#6443 and Hmbown#6456 with `rm: cannot remove '/tmp/tmp.*/checkout': Directory not empty` right after "require-release-tag-checkout tests passed": every assertion held, and only the EXIT trap failed. The fixture commits and pushes, which can spawn a detached `git maintenance run --auto` still writing into .git while the trap deletes it. Disable auto maintenance/gc for every git process in the fixture via GIT_CONFIG_COUNT, and retry the removal once. Verification: the fixture passes locally; shellcheck clean. The race was never reproduced locally, so the root cause is inferred from the log; CI reruns are the proof. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.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
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.
Refs #6424
Four Fleet manager lifecycle tests failed on macOS CI under full-suite
load ("first Fleet attempt never started", "fake worker never
started", managers never converging) and passed alone from the same CI
binary in under 1.2 s. In the concurrent-managers failure the standby
executor held the worker, yet the child's first shell builtin never ran
within 15 s. Timeouts were already raised from 5 s to 15 s without
effect, so this does not raise them again.
Product side: every scheduler tick appended a durable heartbeat per
leased task (sync_data, which is F_FULLFSYNC on macOS) and spawned
psto sample worker memory. At the tests' 10 ms tick that is ~100
full-drive flushes and ~100 process spawns a second per test, against a
300 s stale window. Heartbeats are now skipped when the worker already
has one stamped this second (timestamps are whole-second), and memory
is sampled at most once a second per worker, keeping the last value.
Test side: the fleet::manager module runs in a max-threads = 1 nextest
group, the same remedy as the exec-persistent-service tests (#5355).
Not established: that this is the whole cause. Late spawn vs. late
child start was not distinguished; acceptance is a green loaded hosted
macOS run.
Evidence: 431 passed, 0 failed (12,828 skipped) on the fleet:: selection;
nextest show-config test-groupslists the manager tests in the newgroup. TUI all-target/all-feature Clippy with CI flags and fmt passed.
🤖 Generated with Claude Code