[
https://issues.apache.org/jira/browse/HBASE-10357?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Devaraj Das updated HBASE-10357:
--------------------------------
Attachment: 10357-4.3.txt
Addresses the comments last raised.
The main change is that the "close" of the unneeded scanners are queued up in
the executor service to make sure all scanners (primary or otherwise) are
closed when not needed.
Also a semantic change - when a scanner is obtained, the underlying table
shouldn't have been closed. Updated some unit tests to reflect this. So for
example, one pattern is - one would create a table, do a bunch of puts, and
then close the table. Subsequently, a scanner would be obtained from that
table. In the approach on this jira, since the scanner uses the pool always, if
the table is closed, the underlying pool would be shut down as well. This
would lead to issues. In earlier versions of the patch, I would create a new
pool if the passed pool was shut down but it makes handling the closing of the
pool, etc. an issue. I think for 1.0 we should make the change that a table
shouldn't be closed if we want to invoke methods on that table.
> Failover RPC's for scans
> ------------------------
>
> Key: HBASE-10357
> URL: https://issues.apache.org/jira/browse/HBASE-10357
> Project: HBase
> Issue Type: Sub-task
> Components: Client
> Reporter: Enis Soztutar
> Fix For: 0.99.0
>
> Attachments: 10357-1.txt, 10357-2.txt, 10357-3.2.txt, 10357-3.txt,
> 10357-4.2.txt, 10357-4.3.txt, 10357-4.txt
>
>
> This is extension of HBASE-10355 to add failover support for scans.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
