[
https://issues.apache.org/jira/browse/HBASE-10926?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14030287#comment-14030287
]
Jerry He commented on HBASE-10926:
----------------------------------
The normal layout of the zk node for the 'reached' phase:
: |-/hbase/flush-table-proc
: |-reached
: |----cluster_test
: |-------hdtest010.svl.ibm.com,60020,1402634727563
: |-------hdtest011.svl.ibm.com,60020,1402634727508
procName = ZKUtil.getNodeName(ZKUtil.getParent(path)) --> cluster_test
member = ZKUtil.getNodeName(path) -->
hdtest010.svl.ibm.com,60020,1402634727563
But from the warning message, we seem to have the wrong layout (my deduction,
not actual test result):
: |-/hbase/flush-table-proc
: |-reached
: |----cluster_test
The leaf nodes were missing. Therefore we got:
procName = ZKUtil.getNodeName(ZKUtil.getParent(path)) --> reached
member = ZKUtil.getNodeName(path) --> cluster_test
Either the leaf node was not created or cleared due to race?
> Use global procedure to flush table memstore cache
> --------------------------------------------------
>
> Key: HBASE-10926
> URL: https://issues.apache.org/jira/browse/HBASE-10926
> Project: HBase
> Issue Type: Improvement
> Components: Admin
> Affects Versions: 0.96.2, 0.98.1
> Reporter: Jerry He
> Assignee: Jerry He
> Fix For: 0.99.0
>
> Attachments: HBASE-10926-trunk-v1.patch, HBASE-10926-trunk-v2.patch,
> HBASE-10926-trunk-v3.patch, HBASE-10926-trunk-v4.patch
>
>
> Currently, user can trigger table flush through hbase shell or HBaseAdmin
> API. To flush the table cache, each region server hosting the regions is
> contacted and flushed sequentially, which is less efficient.
> In HBase snapshot global procedure is used to coordinate and flush the
> regions in a distributed way.
> Let's provide a distributed table flush for general use.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
