[
https://issues.apache.org/jira/browse/HBASE-17576?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15867141#comment-15867141
]
Enis Soztutar commented on HBASE-17576:
---------------------------------------
- This does not look correct:
{code}
+ actions_by_region_[hregion_location->HashCode()]->AddAction(action);
{code}
you should construct the map by supplying the hashcode and comparison functions
(look at RegionLocationMap at location-cache.h).
- If you notice in the java side, we do not have a Callable for the Batch
caller. The single request caller is used for various types of requests (get,
scan, etc) while the batch is only used for doing ClientService.multi() call.
So you can simplify it by removing the callable:
{code}
+using BatchCallable =
std::function<folly::Future<RESP>(std::shared_ptr<HBaseRpcController>,
{code}
The consumers of the BatchCaller will be the RawAsyncTableImpl.batch() call,
which takes a list of Row objects.
- Let's discuss offline in what happens with the concurrent scatter-gather
logic inside GroupAndSend(). I'll give you and overview for the java side,
because it is non-trivial.
> [C++] Implement request retry mechanism over RPC for Multi calls.
> -----------------------------------------------------------------
>
> Key: HBASE-17576
> URL: https://issues.apache.org/jira/browse/HBASE-17576
> Project: HBase
> Issue Type: Sub-task
> Reporter: Sudeep Sunthankar
> Assignee: Sudeep Sunthankar
> Attachments: HBASE-17576.HBASE-14850.v1.patch,
> HBASE-17576.HBASE-14850.v2.patch, HBASE-17576.HBASE-14850.v3.patch
>
>
> This work is based on top of HBASE-17465. Multi Calls will be based on this.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
