[ 
https://issues.apache.org/jira/browse/HADOOP-16391?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16873827#comment-16873827
 ] 

Vinayakumar B commented on HADOOP-16391:
----------------------------------------

Thanks for hint [~xkrogen].

Yes, duplicate metrics is because of the exact metric names.

But, catch here is that, _deferred_ response approach is never used in any of 
the Hadoop Code, yet. Still there are duplicate entries in the metrics.

This issue is not common for all protocols. It comes only for protocols which 
have methods declared in them.

Example: 'ContainerManagementProtocolPB' have a method declared explicitly, 
even though same method is declared in Protobuf generated _BlockingInterface_.

In this case, explicitly declared method will be added as a metric for 
{{deferredRpcRates.init()}}. 
{code:java}
public interface ContainerManagementProtocolPB extends 
ContainerManagementProtocolService.BlockingInterface {

  SignalContainerResponseProto signalToContainer(RpcController arg0,
      SignalContainerRequestProto proto) throws ServiceException;
}{code}
This explicit declaration of method in interface is not-necessary IMO. And as a 
workaround fix, this will solve the current duplicate metrics.
{quote}We probably need to adjust addDeferredProcessingTime to add a prefix or 
suffix to all of the metric names.
{quote}
This will be the correct fix. But I am afraid, whether this will be considered 
as incompatible change?

Or should we say, user might be picking wrong metric all along because of 
duplicate metrics, this will be fixed now  and not an 'incompatible' change?

> Duplicate values in rpcDetailedMetrics
> --------------------------------------
>
>                 Key: HADOOP-16391
>                 URL: https://issues.apache.org/jira/browse/HADOOP-16391
>             Project: Hadoop Common
>          Issue Type: Bug
>            Reporter: Bilwa S T
>            Assignee: Bilwa S T
>            Priority: Major
>         Attachments: image-2019-06-25-20-30-15-395.png
>
>
> In RpcDetailedMetrics init is called two times . Once for deferredRpcrates 
> and other one rates metrics which causes duplicate values in RM and NM 
> metrics.
>  !image-2019-06-25-20-30-15-395.png! 



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

---------------------------------------------------------------------
To unsubscribe, e-mail: common-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: common-issues-h...@hadoop.apache.org

Reply via email to