Repository navigation
Add jitter to resource lock requeue delay to prevent starvation - #274
Merged
Merged
Conversation
felix-kaestner
enabled auto-merge (rebase)
April 2, 2026 07:59
swagner-de
approved these changes
Apr 2, 2026
In 7fafc92 we adjusted the resource lock requeue dealy, however we forgot to also adjust the default settings of the locker duration and renew interval which we now adjust as well.
When multiple controllers targeting the same device fail to acquire the per-device resource lock, they all requeue after exactly 1 second. This creates a lock convoy: all controllers wake up simultaneously and compete for the lock in a deterministic order, which can cause some controllers to be perpetually starved. By adding ±10% jitter (900ms–1100ms) to the requeue delay, controllers wake up at slightly different times, breaking the synchronized thundering herd and giving each controller a fair chance to acquire the lock over time.
Merging this branch will decrease overall coverage
Coverage by fileChanged files (no unit tests)
Please note that the "Total", "Covered", and "Missed" counts above refer to code statements instead of lines of code. The value in brackets refers to the test coverage of that file in the old version of the code. |
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
RequeueAfterdelay when a controller fails to acquire the per-device resource lock, preventing lock convoy starvation across all 25+ controllersProblem
All controllers targeting the same device use a Lease-based resource lock to serialize
reconciliations. When the lock is already held, every controller requeues after exactly
time.Second. This creates a lock convoy: all blocked controllers wake up simultaneously,compete for the lock in deterministic work-queue order, and the same controllers can
repeatedly win — starving others indefinitely.
Solution
Wrap the requeue delay with the existing
Jitter()helper (±10%, range 900ms–1100ms).This breaks the synchronized thundering herd by spreading wake-up times, giving each
controller a statistically fair chance to acquire the lock over time.
This is the same pattern used in Kubernetes leader election, client-go retry backoffs,
and other distributed systems to prevent convoy/starvation effects.