[
https://issues.apache.org/jira/browse/HBASE-2414?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
stack updated HBASE-2414:
-------------------------
Attachment: master2.txt
Failed attempt 2: Starts a master with disabled scanners and then fakes it out
by invoking master methods directly calling region start and region server
heartbeat methods, passing back the region open, etc. messages. All works
pretty good coming up but then master wants to connect to the non-running meta
to add edis tot the table itself. We could blank this networking out but in at
least a few of the tests we want to build updates to fail legitimately because
.meta. is not available. This tack may be worth further dev but we need
something better.
Looking now at bringing up the minihdfscluster and then doing direct injection
as per the above patch on the individual daemons to get what I want. Also
looking at changing the proxy so its not over rpc but goes local instead.
> Enhance test suite to be able to specify distributed scenarios
> --------------------------------------------------------------
>
> Key: HBASE-2414
> URL: https://issues.apache.org/jira/browse/HBASE-2414
> Project: Hadoop HBase
> Issue Type: Bug
> Components: test
> Affects Versions: 0.20.3
> Reporter: Karthik Ranganathan
> Attachments: directcluster.txt, master2.txt
>
>
> We keep finding good cases that are reasonably hard to test, yet the test
> suite does not encode these.
> For example:
> HBASE-2413 Master does not respect generation stamps, may result in meta
> getting permanently offlined
> HBASE-2312 Possible data loss when RS goes into GC pause while rolling HLog
> I am sure there are many more such "scenarios" we should put into the unit
> tests.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
