[jira] [Comment Edited] (YARN-10724) Overcounting of preemptions in CapacityScheduler (LeafQueue metrics)
[ https://issues.apache.org/jira/browse/YARN-10724?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17313363#comment-17313363 ] Anup Agarwal edited comment on YARN-10724 at 4/1/21, 6:20 PM: -- completedContainer getting called multiple times may or may not be an issue, but logging the same event multiple times might be. SchedulerApplicationAttempt maintains a liveContainers collection and uses it to deduplicate container completion (incl preemption) events; while leafQueue does no such thing, that's why the patch moved the preemption logging to AppAttempt rather than leafQueue, similar to FSAppAttempt. was (Author: 108anup): completedContainer getting called multiple times may or may not be an issue, but logging the same event multiple times might be. SchedulerApplicationAttempt maintains a liveContainers collection and uses it to deduplicate preemption events; while leafQueue does no such thing, that's why the patch moved the preemption logging to AppAttempt rather than leafQueue, similar to FSAppAttempt. > Overcounting of preemptions in CapacityScheduler (LeafQueue metrics) > > > Key: YARN-10724 > URL: https://issues.apache.org/jira/browse/YARN-10724 > Project: Hadoop YARN > Issue Type: Bug >Reporter: Anup Agarwal >Assignee: Anup Agarwal >Priority: Minor > Attachments: YARN-10724-trunk.001.patch, YARN-10724-trunk.002.patch > > > Currently CapacityScheduler over-counts preemption metrics inside > QueueMetrics. > > One cause of the over-counting: > When a container is already running, SchedulerNode does not remove the > container immediately from launchedContainer list and waits from the NM to > kill the container. > Both NODE_RESOURCE_UPDATE and NODE_UPDATE invoke > signalContainersIfOvercommited (AbstractYarnScheduler) which look for > containers to preempt based on the launchedContainers list. Both these calls > can create a ContainerPreemptEvent for the same container (as RM is waiting > for NM to kill the container). This leads LeafQueue to log metrics for the > same preemption multiple times. -- This message was sent by Atlassian Jira (v8.3.4#803005) - To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
[jira] [Comment Edited] (YARN-10724) Overcounting of preemptions in CapacityScheduler (LeafQueue metrics)
[
https://issues.apache.org/jira/browse/YARN-10724?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17311957#comment-17311957
]
Anup Agarwal edited comment on YARN-10724 at 3/31/21, 1:31 AM:
---
BTW, I have tested these changes with the discrete-time simulator
(https://issues.apache.org/jira/browse/YARN-1187) but have not tested these on
a real testbed. It would be good to get it reviewed.
In the patch, I am unsure about the following two things:
(1) LeafQueue originally checked: {{ContainerExitStatus.PREEMPTED ==
containerStatus.getExitStatus()}}; while FSAppAttempt checks:
{{containerStatus.getDiagnostics().equals(SchedulerUtils.PREEMPTED_CONTAINER)}}
for determining if the container was preempted. Ideally, I think 'logical and'
of these two conditions should be taken.
(2) It may be the case that someone deliberately logged preemptions metrics in
LeafQueue due to some reason that I don't know about. Having said that the
change in the patch is pretty much consistent with code already in FSAppAttempt.
was (Author: 108anup):
BTW, I have tested these changes with the discrete-time simulator
(https://issues.apache.org/jira/browse/YARN-1187) but have not tested these on
a real testbed. It would be good to get it reviewed.
In the patch, I am unsure about the following two things: # LeafQueue
originally checked: {{ContainerExitStatus.PREEMPTED ==
containerStatus.getExitStatus()}}; while FSAppAttempt checks:
{{containerStatus.getDiagnostics().equals(SchedulerUtils.PREEMPTED_CONTAINER)}}
for determining if the container was preempted. Ideally, I think 'logical and'
of these two conditions should be taken.
# It may be the case that someone deliberately logged preemptions metrics in
LeafQueue due to some reason that I don't know about. Having said that the
change in the patch is pretty much consistent with code already in FSAppAttempt.
> Overcounting of preemptions in CapacityScheduler (LeafQueue metrics)
>
>
> Key: YARN-10724
> URL: https://issues.apache.org/jira/browse/YARN-10724
> Project: Hadoop YARN
> Issue Type: Bug
>Reporter: Anup Agarwal
>Assignee: Anup Agarwal
>Priority: Minor
> Attachments: YARN-10724-trunk.001.patch, YARN-10724-trunk.002.patch
>
>
> Currently CapacityScheduler over-counts preemption metrics inside
> QueueMetrics.
>
> One cause of the over-counting:
> When a container is already running, SchedulerNode does not remove the
> container immediately from launchedContainer list and waits from the NM to
> kill the container.
> Both NODE_RESOURCE_UPDATE and NODE_UPDATE invoke
> signalContainersIfOvercommited (AbstractYarnScheduler) which look for
> containers to preempt based on the launchedContainers list. Both these calls
> can create a ContainerPreemptEvent for the same container (as RM is waiting
> for NM to kill the container). This leads LeafQueue to log metrics for the
> same preemption multiple times.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
-
To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
[jira] [Comment Edited] (YARN-10724) Overcounting of preemptions in CapacityScheduler (LeafQueue metrics)
[ https://issues.apache.org/jira/browse/YARN-10724?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17311802#comment-17311802 ] Anup Agarwal edited comment on YARN-10724 at 3/30/21, 8:43 PM: --- Addressed checkstyle issues: [^YARN-10724-trunk.002.patch] was (Author: 108anup): Addressed checkstyle issues: [^YARN-10724-trunk.002.patch] > Overcounting of preemptions in CapacityScheduler (LeafQueue metrics) > > > Key: YARN-10724 > URL: https://issues.apache.org/jira/browse/YARN-10724 > Project: Hadoop YARN > Issue Type: Bug >Reporter: Anup Agarwal >Assignee: Anup Agarwal >Priority: Minor > Attachments: YARN-10724-trunk.001.patch, YARN-10724-trunk.002.patch > > > Currently CapacityScheduler over-counts preemption metrics inside > QueueMetrics. > > One cause of the over-counting: > When a container is already running, SchedulerNode does not remove the > container immediately from launchedContainer list and waits from the NM to > kill the container. > Both NODE_RESOURCE_UPDATE and NODE_UPDATE invoke > signalContainersIfOvercommited (AbstractYarnScheduler) which look for > containers to preempt based on the launchedContainers list. Both these calls > can create a ContainerPreemptEvent for the same container (as RM is waiting > for NM to kill the container). This leads LeafQueue to log metrics for the > same preemption multiple times. -- This message was sent by Atlassian Jira (v8.3.4#803005) - To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
[jira] [Comment Edited] (YARN-10724) Overcounting of preemptions in CapacityScheduler (LeafQueue metrics)
[ https://issues.apache.org/jira/browse/YARN-10724?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17311649#comment-17311649 ] Anup Agarwal edited comment on YARN-10724 at 3/30/21, 5:13 PM: --- I have added a unit test that triggers the overcounting issue along with a fix [^YARN-10724-trunk.001.patch]. The fix also updates FairScheduler to log other preemption metrics including preemptedMemorySeconds and preemptedVcoreSeconds. was (Author: 108anup): I have added a unit test that triggers the overcounting issue along with a fix [^YARN-10724-trunk.001.patch]. The fix also updates FairScheduler to log preemptedMemorySeconds and preemptedVcoreSeconds. > Overcounting of preemptions in CapacityScheduler (LeafQueue metrics) > > > Key: YARN-10724 > URL: https://issues.apache.org/jira/browse/YARN-10724 > Project: Hadoop YARN > Issue Type: Bug >Reporter: Anup Agarwal >Assignee: Anup Agarwal >Priority: Minor > Attachments: YARN-10724-trunk.001.patch > > > Currently CapacityScheduler over-counts preemption metrics inside > QueueMetrics. > > One cause of the over-counting: > When a container is already running, SchedulerNode does not remove the > container immediately from launchedContainer list and waits from the NM to > kill the container. > Both NODE_RESOURCE_UPDATE and NODE_UPDATE invoke > signalContainersIfOvercommited (AbstractYarnScheduler) which look for > containers to preempt based on the launchedContainers list. Both these calls > can create a ContainerPreemptEvent for the same container (as RM is waiting > for NM to kill the container). This leads LeafQueue to log metrics for the > same preemption multiple times. -- This message was sent by Atlassian Jira (v8.3.4#803005) - To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
