hadoop-common-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Todd Lipcon (Updated) (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (HADOOP-8212) Improve ActiveStandbyElector's behavior when session expires
Date Mon, 26 Mar 2012 04:58:31 GMT

     [ https://issues.apache.org/jira/browse/HADOOP-8212?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Todd Lipcon updated HADOOP-8212:
--------------------------------

    Attachment: hadoop-8212.txt

Attached patch fixes the behavior to not notifyFatalError when the session is expired. The
existing code already handled rejoining.

I also fixed a race bug I turned up where, after rejoining with a new zkClient, some old notifications
from the previous zkClient could end up getting through. The watchers and callbacks now pass
along the zkClient used to set them, and then in the callback, we check to make sure it is
still current.

I also simplified the test case to no longer be multi-threaded, since it's much easier to
follow as a linear progression, and the threads didn't buy us anything. I added test coverage
around session expiration to cover the new code.
                
> Improve ActiveStandbyElector's behavior when session expires
> ------------------------------------------------------------
>
>                 Key: HADOOP-8212
>                 URL: https://issues.apache.org/jira/browse/HADOOP-8212
>             Project: Hadoop Common
>          Issue Type: Improvement
>    Affects Versions: 0.23.3, 0.24.0
>            Reporter: Todd Lipcon
>            Assignee: Todd Lipcon
>         Attachments: hadoop-8212.txt
>
>
> Currently when the ZK session expires, it results in a fatal error being sent to the
application callback. This is not the best behavior -- for example, in the case of HA, if
ZK goes down, we would like the current state to be maintained, rather than causing either
NN to abort. When the ZK clients are able to reconnect, they should sort out the correct leader
based on the normal locking schemes.

--
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

        

Mime
View raw message