[
https://issues.apache.org/jira/browse/HADOOP-18885?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17767401#comment-17767401
]
ASF GitHub Bot commented on HADOOP-18885:
-----------------------------------------
simbadzina commented on code in PR #6043:
URL: https://github.com/apache/hadoop/pull/6043#discussion_r1332537239
##########
hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/ipc/Server.java:
##########
@@ -2945,6 +2945,9 @@ private void processRpcRequest(RpcRequestHeaderProto
header,
}
}
} catch (IOException ioe) {
+ if (ioe instanceof RetriableException) {
Review Comment:
Consider putting a try catch blocked around receiveRequestState. This makes
it easier to see where the exception is coming from, instead of checking the
type of exception later. So something like
```java
try {
stateId = alignmentContext.receiveRequestState(header, getMaxIdleTime());
} catch (RetriableException re) {
rpcMetrics.incrRcRejectedByObserverCalls();
throw re;
}
```
> Add rpcRejectedByObserverCalls metric to quantify the number of rejected
> RPCs by Observer NameNode
> ---------------------------------------------------------------------------------------------------
>
> Key: HADOOP-18885
> URL: https://issues.apache.org/jira/browse/HADOOP-18885
> Project: Hadoop Common
> Issue Type: Improvement
> Reporter: Haiyang Hu
> Assignee: Haiyang Hu
> Priority: Major
> Labels: pull-request-available
>
> When observer Node's serverStateId is too far behind clientStateId will
> rejected client RPC request,
> mabe we need add rpcRejectedByObserverCalls metric to RpcMetrics to quantify
> the number of rejected RPCs by Observer NameNode to help determine how well
> the server.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
