[
https://issues.apache.org/jira/browse/HBASE-18204?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16056886#comment-16056886
]
Enis Soztutar commented on HBASE-18204:
---------------------------------------
I was testing this patch together with a couple of other patches. It seems that
we are still not there yet in terms of handling regionserver failure. Testing
with 2 regionservers, 10 regions, and writing 500M of data. When the server is
killed the client hangs:
{code}
I0621 02:36:04.355890 6477 simple-client.cc:110] Sent 90000 Put requests in
33961 ms.
I0621 02:36:08.138052 6477 simple-client.cc:110] Sent 100000 Put requests in
37743 ms.
W0621 02:36:09.939910 6485 HandlerContext-inl.h:186] readException reached end
of pipeline
{code}
> [C++] Cleanup outgoing RPCs on connection close
> -----------------------------------------------
>
> Key: HBASE-18204
> URL: https://issues.apache.org/jira/browse/HBASE-18204
> Project: HBase
> Issue Type: Sub-task
> Reporter: Enis Soztutar
> Assignee: Enis Soztutar
> Attachments: hbase-18204_v1.patch
>
>
> Our client-dispatcher maintains a map of outgoing RPCs per server with the
> promises.
> In case the server goes down, or TCP connection is closed, we should complete
> the outgoing RPCs with exceptions so that higher level waiters can unblock
> and retry.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
