[
https://issues.apache.org/jira/browse/YARN-4308?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15286840#comment-15286840
]
Sunil G commented on YARN-4308:
-------------------------------
Thanks for weighing in the thoughts [~templedf] and [~Naganarasimha Garla]
bq.I'm fine with just making sure that anyone hunting reasons for missing
reports will trip over a pointer to the likely cause
Metric can only give a suggestion and a possible error cause. So once this
symptom is hit, it produces some indications (runtime metric). And it can help
the admin to enable debug log for this class alone (as [~Naganarasimha Garla]
mentioned). I am not very much sure about internally changing log level. Seems
like it can be little bit of too much complex snippets to handle a defined
scenario (comments/javadoc can define the protocol or standard). So mostly
agreeing to [~Naganarasimha Garla]'s view and still thinking that we can have a
metric.
> ContainersAggregated CPU resource utilization reports negative usage in first
> few heartbeats
> --------------------------------------------------------------------------------------------
>
> Key: YARN-4308
> URL: https://issues.apache.org/jira/browse/YARN-4308
> Project: Hadoop YARN
> Issue Type: Bug
> Components: nodemanager
> Affects Versions: 2.7.1
> Reporter: Sunil G
> Assignee: Sunil G
> Attachments: 0001-YARN-4308.patch, 0002-YARN-4308.patch
>
>
> NodeManager reports ContainerAggregated CPU resource utilization as -ve value
> in first few heartbeats cycles. I added a new debug print and received below
> values from heartbeats.
> {noformat}
> INFO
> org.apache.hadoop.yarn.server.nodemanager.containermanager.monitor.ContainersMonitorImpl:
> ContainersResource Utilization : CpuTrackerUsagePercent : -1.0
> INFO
> org.apache.hadoop.yarn.server.nodemanager.containermanager.monitor.ContainersMonitorImpl:ContainersResource
> Utilization : CpuTrackerUsagePercent : 198.94598
> {noformat}
> Its better we send 0 as CPU usage rather than sending a negative values in
> heartbeats eventhough its happening in only first few heartbeats.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
