You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@activemq.apache.org by "Hiram Chirino (Commented) (JIRA)" <ji...@apache.org> on 2012/04/12 14:11:20 UTC

[jira] [Commented] (APLO-190) Add support for socket level tuning

    [ https://issues.apache.org/jira/browse/APLO-190?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13252366#comment-13252366 ] 

Hiram Chirino commented on APLO-190:
------------------------------------

Apollo always sets SO_LINGER to 0, so that it can control when the TCP connection is finally reset.  It also always enables TCP_NODELAY since apollo handles optimally buffering multiple writes into bigger IP packets.

The SO_RCVBUF and SO_SNDBUF are configured via the tcp transport's 'receive_buffer_size' and 'send_buffer_size' (I'll update the docs so that's more clear).

I'll shortly add an option for configuring the listen backlock.
                
> Add support for socket level tuning
> -----------------------------------
>
>                 Key: APLO-190
>                 URL: https://issues.apache.org/jira/browse/APLO-190
>             Project: ActiveMQ Apollo
>          Issue Type: Improvement
>            Reporter: Lionel Cons
>
> In order to better control Apollo's network behavior, it would be very good to be able to configure some socket parameters via Apollo's configuration file. The ones that come to my mind now are:
> listen()'s backlog
> setsockopt()'s SO_LINGER, SO_RCVBUF and SO_SNDBUF
> FWIW, SO_KEEPALIVE is already tunable in Apollo and ActiveMQ has these already (see http://activemq.apache.org/tcp-transport-reference.html).

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira