[
https://issues.apache.org/jira/browse/HBASE-10566?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13908493#comment-13908493
]
Nicolas Liochon commented on HBASE-10566:
-----------------------------------------
The difficult point is to correlate this with the cancel. The RpcController
supports the cancellation, but we don't really show this class to the client
code. However, doing a cancel is a client decision. So we can be protobuf
compliant with the timeout, as it's something we manage as a conf parameter, so
we don't have to show the RpcController to the client, but we can't easily be
protobuf compliant with the cancel, as it would require to expose this
RpcController.
the path for this jira is:
clean rpc timeout -> shared thread pool for the connection instead of 2 threads
per region server -> cancellation server side
likely, this would help to put Netty or byteBuffer in the loop as well.
And this would allow to have shorter socket timeout: today, if we want to
support an operation that can last a few minutes (a scan with a lot of filters
under heavy load for example), we need to have a socket timeout of a few
minutes as well.
So I'm interested in any feedback. [~saint....@gmail.com],
[~andrew.purt...@gmail.com]; [~enis]; [~devaraj] ?
> cleanup rpcTimeout in the client
> --------------------------------
>
> Key: HBASE-10566
> URL: https://issues.apache.org/jira/browse/HBASE-10566
> Project: HBase
> Issue Type: Bug
> Components: Client
> Affects Versions: 0.99.0
> Reporter: Nicolas Liochon
> Assignee: Nicolas Liochon
> Fix For: 0.99.0
>
> Attachments: 10566.sample.patch
>
>
> There are two issues:
> 1) A confusion between the socket timeout and the call timeout
> Socket timeouts should be minimal: a default like 20 seconds, that could be
> lowered to single digits timeouts for some apps: if we can not write to the
> socket in 10 second, we have an issue. This is different from the total
> duration (send query + do query + receive query), that can be longer, as it
> can include remotes calls on the server and so on. Today, we have a single
> value, it does not allow us to have low socket read timeouts.
> 2) The timeout can be different between the calls. Typically, if the total
> time, retries included is 60 seconds but failed after 2 seconds, then the
> remaining is 58s. HBase does this today, but by hacking with a thread local
> storage variable. It's a hack (it should have been a parameter of the
> methods, the TLS allowed to bypass all the layers. May be protobuf makes this
> complicated, to be confirmed), but as well it does not really work, because
> we can have multithreading issues (we use the updated rpc timeout of someone
> else, or we create a new BlockingRpcChannelImplementation with a random
> default timeout).
> Ideally, we could send the call timeout to the server as well: it will be
> able to dismiss alone the calls that it received but git stick in the request
> queue or in the internal retries (on hdfs for example).
> This will make the system more reactive to failure.
> I think we can solve this now, especially after 10525. The main issue is to
> something that fits well with protobuf...
> Then it should be easy to have a pool of thread for writers and readers, w/o
> a single thread per region server as today.
--
This message was sent by Atlassian JIRA
(v6.1.5#6160)
