hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "stack (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (HBASE-9461) Some doc and cleanup in RPCServer
Date Fri, 13 Sep 2013 05:58:52 GMT

     [ https://issues.apache.org/jira/browse/HBASE-9461?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

stack updated HBASE-9461:
-------------------------

    Attachment: 9461v2.txt

More doc and untangling.  More to do but this ok for now.

RpcServer had internal CallRunner class.  The scheduler was reaching over to use this inner
class.  Broke it out of RpcServer and made it Standalone.  Made it take an RpcServerInterface
instead of RpcServer.

Undid CallRunner implementing Runnable (confusing -- how it is run is internal function of
Scheduler implementation)

Cleaned up RpcServerInterface explaining why a start and a startThreads and an openServer.
 Added a few methods to support CallRunner being outside of RpcServer.

Added into CallRunner the managment of the call queue size.  As it was, callQueueSize was
incremented before we created a CallRunner but internal to CallRunner it was managing the
decrement.

Shutdown access on CallRunner constructor and CR#getStatus and other methods only for use
in this package.

Added javadoc all around to explain the cryptic.

RpcSchedulerContextImpl was like CallRunner, a class used by the Scheduler only it was an
inner class of RpcServer.  Moved it out.

Removed unused thread local SERVER.

Add simple test to demo being able to instantiate a CallRunner outside of RpcServer context.a
                
> Some doc and cleanup in RPCServer
> ---------------------------------
>
>                 Key: HBASE-9461
>                 URL: https://issues.apache.org/jira/browse/HBASE-9461
>             Project: HBase
>          Issue Type: Bug
>            Reporter: stack
>            Assignee: stack
>         Attachments: 9461.txt, 9461v2.txt, ipc2.ucls
>
>
> RPC is a dog to follow.  I want to do buffer pooling for reading requests but its tough
drawing the diagram of who is doing what when.  HBASE-8884 seems to have made it more involved
still.  This issue is about doing a bit of untangling.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira

Mime
View raw message