[
https://issues.apache.org/jira/browse/HBASE-6987?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Gregory Chanan updated HBASE-6987:
----------------------------------
Attachment: HBASE-6987.patch
* Attached HBASE-6987.patch *
Code in trunk looks good (thanks again to nkeywal).
This patch makes a couple of changes:
- Simplifies some HBaseAdmin logic wrt checking if the master is available
- Adds a test for HBASE-6920 (it passes, since the code-under-test is good :)).
It is non-invasive to the RpcEngine layer, unlike the original test in
HBASE-6920.
The test isn't perfect; it only throws ServiceExceptions because that is all
that ProtobufRpcEngine throws. If that changes or another RpcEngine is used,
the RandomTimeoutRpcEngine may not properly mimic the exception types coming
from the RpcEngine. Putting hooks into any deeper layer (e.g. HBaseClient)
would get very tricky and invasive.
> Port HBASE-6920 to trunk (?)
> ----------------------------
>
> Key: HBASE-6987
> URL: https://issues.apache.org/jira/browse/HBASE-6987
> Project: HBase
> Issue Type: Bug
> Reporter: Gregory Chanan
> Assignee: Gregory Chanan
> Priority: Minor
> Fix For: 0.96.0
>
> Attachments: HBASE-6987.patch
>
>
> Need to investigate whether we need to port HBASE-6920 to trunk.
--
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
