You are viewing a plain text version of this content. The canonical link for it is here.
Posted to hdfs-issues@hadoop.apache.org by "Colin Patrick McCabe (JIRA)" <ji...@apache.org> on 2012/07/13 00:53:35 UTC
[jira] [Created] (HDFS-3651) optionally, the NameNode should invoke
saveNamespace after getting a SIGTERM
Colin Patrick McCabe created HDFS-3651:
------------------------------------------
Summary: optionally, the NameNode should invoke saveNamespace after getting a SIGTERM
Key: HDFS-3651
URL: https://issues.apache.org/jira/browse/HDFS-3651
Project: Hadoop HDFS
Issue Type: Improvement
Components: name-node
Affects Versions: 2.0.0-alpha
Reporter: Colin Patrick McCabe
Assignee: Colin Patrick McCabe
Priority: Minor
It would be nice if the NameNode could be configured so that it did a saveNamespace and then shut down cleanly after receiving a SIGTERM signal. In general, it is a good practice to call saveNamespace when doing an orderly shutdown, to ensure that all of the information in the namespace is on disk. Of course, this should not be necessary if the SecondaryNameNode or StandbyNameNode is operating correctly. However, when there are bugs in these daemons, a saveNamespace can prevent disaster.
Currently, we don't catch SIGTERM, but just shut down immediately, without doing any cleanup. Of course, it will always be possible to shut down the NameNode without doing a saveNamespace, simply by sending SIGKILL, which is un-catchable.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira