[
https://issues.apache.org/jira/browse/YARN-309?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13615605#comment-13615605
]
Bikas Saha commented on YARN-309:
---------------------------------
why is this check needed when it has already been done during init()?
{code}
+ if (getNextHeartBeatInterval() <= 0) {
+ LOG.warn("The HeartBeat Interval should be larger than 0. " +
+ "Use default value instead.");
+ setNextHeartbeatInterval(YarnConfiguration.
+ DEFAULT_NM_HEARTBEAT_INTERVAL_MS);
+ }
{code}
Is this setter really needed?
{code}
+ public void setNextHeartbeatInterval(long nextHeartbeatInterval){
+ this.nextHeartBeatInterval = nextHeartbeatInterval;
+ }
{code}
Something that we could do, possibly in another jira would be to send
nextheartbeat randomized between 1-1.2 times the actual values. This helps a
lot in smoothing out the incoming requests over time.
> Make RM provide heartbeat interval to NM
> ----------------------------------------
>
> Key: YARN-309
> URL: https://issues.apache.org/jira/browse/YARN-309
> Project: Hadoop YARN
> Issue Type: Sub-task
> Components: resourcemanager
> Reporter: Xuan Gong
> Assignee: Xuan Gong
> Attachments: YARN-309.1.patch, YARN-309.2.patch, YARN-309.3.patch,
> YARN-309.4.patch
>
>
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
