[
https://issues.apache.org/jira/browse/HDFS-16415?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
yangzhanling updated HDFS-16415:
--------------------------------
Description:
NameNode has stoped when it wrote metadata to JournalNode happend timeout
exception,it is ok normally, _because have been well protected with meta data
in namenode._
_But when wait the respoonse for JorunalNode, it will be Full GC as quick as a
flash and it spent a lot time,then the timeout exception has happened ,and
Namenode has stoped, so it was unstable, and unreliable._
_as follows:_
org.apache.hadoop.hdfs.qjournal.client.QuorumCall#waitFor
//when happend FULL GC spent some time ,rem < 0,so timeoutException, namenode
stoped
if (rem <= 0)
{ throw new TimeoutException(); }
was:
NameNode has stoped when it wrote metadata to JournalNode happend timeout
exception,it is ok normally, _because have been well protected with meta data
in namenode._
_But when wait the respoonse for JorunalNode, it will be Full GC as quick as a
flash and it spent a lot time,then the timeout exception has happened ,and
Namenode has stoped, so it was unstable, and unreliable._
> Namenode has stoped when it wrote metadata to JournalNode happend timeout
> Exception
> -----------------------------------------------------------------------------------
>
> Key: HDFS-16415
> URL: https://issues.apache.org/jira/browse/HDFS-16415
> Project: Hadoop HDFS
> Issue Type: Improvement
> Components: namanode
> Affects Versions: 2.7.7
> Environment: org.apache.hadoop.hdfs.qjournal.client.QuorumCall#waitFor
> //when happend FULL GC spent some time ,rem < 0,so timeoutException, namenode
> stoped
> if (rem <= 0) {
> throw new TimeoutException();
> }
> Reporter: yangzhanling
> Priority: Major
>
> NameNode has stoped when it wrote metadata to JournalNode happend timeout
> exception,it is ok normally, _because have been well protected with meta data
> in namenode._
> _But when wait the respoonse for JorunalNode, it will be Full GC as quick as
> a flash and it spent a lot time,then the timeout exception has happened ,and
> Namenode has stoped, so it was unstable, and unreliable._
> _as follows:_
> org.apache.hadoop.hdfs.qjournal.client.QuorumCall#waitFor
> //when happend FULL GC spent some time ,rem < 0,so timeoutException, namenode
> stoped
> if (rem <= 0)
> { throw new TimeoutException(); }
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
