Zhe Zhang created HADOOP-13657:
----------------------------------
Summary: IPC Reader thread could silently die and leave NameNode
unresponsive
Key: HADOOP-13657
URL: https://issues.apache.org/jira/browse/HADOOP-13657
Project: Hadoop Common
Issue Type: Bug
Components: ipc
Reporter: Zhe Zhang
Priority: Critical
For each listening port, IPC {{Server#Listener#Reader}} is a single thread in
charge of moving {{Connection}} items from {{pendingConnections}} (capacity
100) to the {{callQueue}}.
We have experienced an incident where the {{Reader}} thread for HDFS NameNode
died from run time exception. Then the {{pendingConnections}} queue became full
and the NameNode port became inaccessible.
In our particular case, what killed {{Reader}} was a NPE caused by
https://bugs.openjdk.java.net/browse/JDK-8024883. But in general, other types
of runtime exceptions could cause this issue as well.
We should add logic to either make the {{Reader}} more robust in case of
runtime exceptions, or at least treat it as a FATAL exception so that NameNode
can fail over to standby, and admins get alerted of the real issue.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
