[
https://issues.apache.org/jira/browse/HBASE-16505?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15474668#comment-15474668
]
Enis Soztutar commented on HBASE-16505:
---------------------------------------
Sorry to come in late, but I don't think we should have committed the patch as
it is. Where is the +1's from the other reviewers?
AsyncRegion is not really async as of now. It is a misnomer. Realistically,
there will only be 1 implementation of HRegion (whether it is completely
blocking or not), so no need to differentiate Region / AsyncRegion I think.
This is probably intended only for 2.0, so we can and should change the Region
interface instead of building on top. There are no guarantees for coprocessor
compatibility between 1.x and 2.0.
I like the idea for having timeouts and a context carried around for the region
operation. It helps with handlers being stuck and and the regionservers
stalling. What are the plans for checking the deadline? Not every operation can
throw timeouts in the middle. The plan is to hand code timeout checks in most
places in get / put code ?
Why listeners instead of Future / CompletableFutures ?
> Add AsyncRegion interface to pass deadline and support async operations
> -----------------------------------------------------------------------
>
> Key: HBASE-16505
> URL: https://issues.apache.org/jira/browse/HBASE-16505
> Project: HBase
> Issue Type: Sub-task
> Reporter: Phil Yang
> Assignee: Phil Yang
> Fix For: 2.0.0
>
> Attachments: HBASE-16505-v1.patch, HBASE-16505-v10.patch,
> HBASE-16505-v10.patch, HBASE-16505-v2.patch, HBASE-16505-v3.patch,
> HBASE-16505-v4.patch, HBASE-16505-v5.patch, HBASE-16505-v6.patch,
> HBASE-16505-v7.patch, HBASE-16505-v8.patch, HBASE-16505-v9.patch
>
>
> If we want to know the correct setting of timeout in read/write path, we need
> add a new parameter in operation-methods of Region.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
