Details
Description
This JIRA proposes to add a counter called RpcSlowCalls and also a configuration setting that allows users to log really slow RPCs. Slow RPCs are RPCs that fall at 99th percentile. This is useful to troubleshoot why certain services like name node freezes under heavy load.
Attachments
Attachments
Issue Links
- breaks
-
HADOOP-12457 [JDK8] Fix a failure of compiling common by javadoc
-
- Resolved
-
- is related to
-
HADOOP-15840 Slow RPC logger too spammy
-
- Open
-
-
HDFS-5180 Output the processing time of slow RPC request to node's log
-
- Resolved
-