Skip to content

Count all lost tasks as inactive #2263

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

Merged
rosalind210 merged 1 commit into from
Jan 25, 2022
Merged

Count all lost tasks as inactive #2263

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
21 changes: 1 addition & 20 deletions ...Service/src/main/java/com/hubspot/singularity/scheduler/SingularityDeployCheckHelper.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -444,25 +444,6 @@ public static Set<SingularityTaskId> getNewInactiveDeployTasks(
pendingDeploy.getDeployProgress().getFailedDeployTasks()
);

return newInactiveDeployTasks
.stream()
.filter(
t -> {
// All TASK_LOSTs that are not resource limit related should be able to be retried
for (SingularityTaskHistoryUpdate historyUpdate : taskManager.getTaskHistoryUpdates(
t
)) {
String historyUpdateReason = historyUpdate.getStatusReason().orElse("");
if (
historyUpdate.getTaskState() == ExtendedTaskState.TASK_LOST &&
!historyUpdateReason.startsWith("REASON_CONTAINER_LIM")
) {
return false;
}
}
return true;
}
)
.collect(Collectors.toSet());
return newInactiveDeployTasks;
}
}
34 changes: 34 additions & 0 deletions ...larityService/src/test/java/com/hubspot/singularity/scheduler/SingularityDeploysTest.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -31,6 +31,7 @@
import java.util.concurrent.TimeUnit;
import javax.ws.rs.WebApplicationException;
import org.apache.mesos.v1.Protos.TaskState;
import org.apache.mesos.v1.Protos.TaskStatus.Reason;
import org.junit.jupiter.api.Assertions;
import org.junit.jupiter.api.MethodOrderer;
import org.junit.jupiter.api.Order;
Expand Down Expand Up @@ -1622,4 +1623,37 @@ public void testDeployWithImmediateRunSchedulesAfterRunningImmediately() {
Assertions.assertEquals(0, taskManager.getNumActiveTasks());
Assertions.assertEquals(1, taskManager.getNumScheduledTasks());
}

@Test
public void testDeployWithLostTasksFails() {
initRequest();
deploy(
firstDeployId,
Optional.<Boolean>empty(),
Optional.of(1),
Optional.of(false),
false
);
deployChecker.checkDeploys();
scheduler.drainPendingQueue();

Assertions.assertEquals(1, taskManager.getPendingTaskIds().size());

resourceOffers();
Assertions.assertEquals(
1,
taskManager.getActiveTaskIdsForDeploy(requestId, firstDeployId).size()
);

SingularityTaskId firstTaskId = taskManager
.getActiveTaskIdsForDeploy(requestId, firstDeployId)
.get(0);
statusUpdate(taskManager.getTask(firstTaskId).get(), TaskState.TASK_LOST);

deployChecker.checkDeploys();
Assertions.assertEquals(
DeployState.FAILED,
deployManager.getDeployResult(requestId, firstDeployId).get().getDeployState()
);
}
}