hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jimmy Xiang (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HBASE-8233) Poke the cache and let the system reload cached information
Date Mon, 01 Apr 2013 18:25:16 GMT

    [ https://issues.apache.org/jira/browse/HBASE-8233?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13619008#comment-13619008

Jimmy Xiang commented on HBASE-8233:

ZK watcher is an alternative. But we don't want too many ZK watchers, which may have performance
For the ZKTable state specific, the cache should match the znode all the time since the updates
are expected to from the code, not from zkcli. So generally, a ZK watcher is not needed. 
If we have a watcher just for this scenario (HBASE-6469), it may be a waste.
> Poke the cache and let the system reload cached information
> -----------------------------------------------------------
>                 Key: HBASE-8233
>                 URL: https://issues.apache.org/jira/browse/HBASE-8233
>             Project: HBase
>          Issue Type: New Feature
>            Reporter: Jimmy Xiang
> Sometimes, cached information is stale and we don't have a way to get rid of it except
restarting the server.  If we can poke the cache and let the system reloads, it will much
easier and quicker.

This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira

View raw message