[
https://issues.apache.org/jira/browse/HDFS-16979?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17714313#comment-17714313
]
ASF GitHub Bot commented on HDFS-16979:
---------------------------------------
virajjasani commented on code in PR #5552:
URL: https://github.com/apache/hadoop/pull/5552#discussion_r1171870853
##########
hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/ipc/CallerContext.java:
##########
@@ -47,6 +47,7 @@ public final class CallerContext {
// field names
public static final String CLIENT_IP_STR = "clientIp";
public static final String CLIENT_PORT_STR = "clientPort";
+ public static final String DFSROUTER_PORT_STR = "dfsrouterPort";
Review Comment:
@LiuGuH if this comment is addressed, then also things look good IMO.
The point is both `dfsrouterPort` and using `isClientPortInfoAbsent()` to
set it in the audit logs, together seems fragile because anyone in future
passes clientPort in the context and will get dfsrouterPort printed, which is
not correct for non-router usecases.
> RBF: Add dfsrouter port in hdfsauditlog
> ---------------------------------------
>
> Key: HDFS-16979
> URL: https://issues.apache.org/jira/browse/HDFS-16979
> Project: Hadoop HDFS
> Issue Type: Improvement
> Reporter: liuguanghua
> Priority: Major
> Labels: pull-request-available
>
> When remote client request through dfsrouter to namenode, the hdfsauditlog
> record the remote client ip and port ，dfsrouter IP，but lack of dfsrouter port.
> This patch is done for this scene.
>
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
