[jira] [Commented] (PROTON-2339) Introduce 'Drain buffers' event to proactor raw connection

classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view
|

[jira] [Commented] (PROTON-2339) Introduce 'Drain buffers' event to proactor raw connection

ASF GitHub Bot (Jira)

    [ https://issues.apache.org/jira/browse/PROTON-2339?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17293974#comment-17293974 ]

ASF subversion and git services commented on PROTON-2339:
---------------------------------------------------------

Commit 92072237908668d0117f1b6bf5fabe7328b6589b in qpid-proton's branch refs/heads/master from Andrew Stitcher
[ https://gitbox.apache.org/repos/asf?p=qpid-proton.git;h=9207223 ]

PROTON-2339: Proactor raw connection: Introduce DRAIN_BUFFERS event

This will supercede the use of READ/WRITTEN events to tell the
application to retrieve its buffers before DISCONNECT.

This more specific event should be clearer in purpose.

For backwards compatibility with existing code, if the application
doesn't drain the buffers in response to the drain event, we will
send the read/written events (as applicable) as well. This behaviour
should be removed when the existing major application (dispatch) is
fixed to use this new event.


> Introduce 'Drain buffers' event to proactor raw connection
> ----------------------------------------------------------
>
>                 Key: PROTON-2339
>                 URL: https://issues.apache.org/jira/browse/PROTON-2339
>             Project: Qpid Proton
>          Issue Type: Improvement
>          Components: proton-c
>    Affects Versions: proton-c-0.33.0
>            Reporter: Andrew Stitcher
>            Assignee: Andrew Stitcher
>            Priority: Major
>             Fix For: proton-c-0.34.0
>
>
> In discussions with users of th raw connection API there is constant (if often brief) puzzlement over receiving read/written events after the connectino is closed but before the disconnect event.
> These events are to tell the application to retrieve all the buffers that they have given the raw connection before the connection becomes completely inaccessible and hence the buffre would leak.
> It would be clearer (and invite less confusion) to have a specific event to tell the application to drain all its buffers.



--
This message was sent by Atlassian Jira
(v8.3.4#803005)

---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]