Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

tweak cooldown thresholds and evaluation logic #1973

Merged
merged 2 commits into from
Jul 11, 2019
Merged

tweak cooldown thresholds and evaluation logic #1973

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
d0fadbf
tweak cooldown thresholds and evaluation logic
ssalinas Jul 10, 2019
96e42ec
fix failure window condition
ssalinas Jul 10, 2019
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions ...ularityService/src/main/java/com/hubspot/singularity/config/SingularityConfiguration.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -120,15 +120,15 @@ public class SingularityConfiguration extends Configuration {

private int fastFailureCooldownCount = 3;

private long fastFailureCooldownMs = 60000;
private long fastFailureCooldownMs = 30000;

private long fastCooldownExpiresMinutesWithoutFailure = 5;

private int slowFailureCooldownCount = 5;

private long slowFailureCooldownMs = 600000;

private long slowCooldownExpiresMinutesWithoutFailure = 8;
private long slowCooldownExpiresMinutesWithoutFailure = 5;

private long cooldownMinScheduleSeconds = 120;

Expand Down
5 changes: 3 additions & 2 deletions SingularityService/src/main/java/com/hubspot/singularity/scheduler/SingularityCooldown.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -67,8 +67,9 @@ private boolean hasFailureLoop(SingularityDeployStatistics deployStatistics, Opt
.count();
java.util.Optional<Long> mostRecentFailure = failureTimestamps.stream().max(Comparator.comparingLong(Long::valueOf));

return failureCount >= cooldownCount
&& (!mostRecentFailure.isPresent() || mostRecentFailure.get() > System.currentTimeMillis() - TimeUnit.MINUTES.toMillis(expiresAfterMins));
boolean mostRecentFailureOutsideWindow = !mostRecentFailure.isPresent() || mostRecentFailure.get() < System.currentTimeMillis() - TimeUnit.MINUTES.toMillis(expiresAfterMins);

return failureCount >= cooldownCount && !mostRecentFailureOutsideWindow;
}

boolean hasCooldownExpired(SingularityDeployStatistics deployStatistics, Optional<Long> recentFailureTimestamp) {
Expand Down
9 changes: 4 additions & 5 deletions SingularityService/src/main/java/com/hubspot/singularity/scheduler/SingularityScheduler.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -591,7 +591,10 @@ private Optional<PendingType> handleCompletedTaskWithStatistics(Optional<Singula
}

if (!status.hasReason() || !status.getReason().equals(Reason.REASON_INVALID_OFFERS)) {
if (!state.isSuccess() && taskHistoryUpdateCreateResult == SingularityCreateResult.CREATED && cooldown.shouldEnterCooldown(request, requestState, deployStatistics, timestamp)) {
if (state != ExtendedTaskState.TASK_KILLED
&& !state.isSuccess()
&& taskHistoryUpdateCreateResult == SingularityCreateResult.CREATED
&& cooldown.shouldEnterCooldown(request, requestState, deployStatistics, timestamp)) {
LOG.info("Request {} is entering cooldown due to task {}", request.getId(), taskId);
requestState = RequestState.SYSTEM_COOLDOWN;
requestManager.cooldown(request, System.currentTimeMillis());
Expand Down Expand Up @@ -715,10 +718,6 @@ private void updateDeployStatistics(SingularityDeployStatistics deployStatistics
} else {
bldr.setAverageSchedulingDelayMillis(Optional.of(startedAt - dueTime));
}

final SingularityDeployStatistics newStatistics = bldr.build();

deployManager.saveDeployStatistics(newStatistics);
}

bldr.setNumTasks(bldr.getNumTasks() + 1);
Expand Down