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-9457) Master could fail start if region server with system table is down
Date Mon, 09 Sep 2013 20:34:52 GMT

    [ https://issues.apache.org/jira/browse/HBASE-9457?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13762248#comment-13762248

Jimmy Xiang commented on HBASE-9457:

Yes, it is hard to ensure that, especially when those region server recovery cases are considered,
even harder if log replay is considered.

To undo the special namespace system table assignment, we need to change how namespace is
initialized.  Currently ZKNamespaceManager can be initiated only after the namespace table
is assigned.  However ZKNamespaceManager is not used any where. If we are ok to (1) put ZKNamespaceManager
aisde for now or just remove it, (2) initiate TableNamespaceManager lazily only when it is
needed, then we can undo it.
> Master could fail start if region server with system table is down
> ------------------------------------------------------------------
>                 Key: HBASE-9457
>                 URL: https://issues.apache.org/jira/browse/HBASE-9457
>             Project: HBase
>          Issue Type: Bug
>          Components: master, Region Assignment
>            Reporter: Jimmy Xiang
>            Assignee: Jimmy Xiang
>            Priority: Critical
> In the region server holding the system table is killed while master is starting, master
will hang there waiting for system table to be assigned which won't happen.

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