[jira] [Updated] (APLO-315) Apollo should protect itself against clients that send many frames with receipt requests, but which do not read the socket for those receipts.

["Hiram Chirino (JIRA)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/APLO-315?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Hiram Chirino updated APLO-315:
-------------------------------

    Summary: Apollo should protect itself against clients that send many frames with receipt requests, but which do not read the socket for those receipts.  (was: Apollo does not send back receipts under load)
    
> Apollo should protect itself against clients that send many frames with receipt requests, but which do not read the socket for those receipts.
> ----------------------------------------------------------------------------------------------------------------------------------------------
>
>                 Key: APLO-315
>                 URL: https://issues.apache.org/jira/browse/APLO-315
>             Project: ActiveMQ Apollo
>          Issue Type: Bug
>         Environment: apollo-99-trunk-20130322.033735-216
>            Reporter: Lionel Cons
>            Assignee: Hiram Chirino
>         Attachments: APLO-315.out, APLO-315.pl, APLO-315.pl.patch, APLO-315.pl.patch2
>
>
> It seems that, under load (even a moderated one, with CPU usage < 100%), Apollo does not send back receipts to the producers waiting for them. The attached script shows producers having stopped sending new messages and waiting several minutes without getting the expected receipts.
> The setup is quite particular (1 topic, 1 direct dsub and 1 dsub with selector) because this comes from trying to reproduce APLO-284.
> See the attached script and its annotated output for more information.

--
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