[jira] Updated: (SMXCOMP-847) stflow handling/behaviour in the http consumer processor is wrong

["Thomas Termin (JIRA)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/SMXCOMP-847?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Thomas Termin updated SMXCOMP-847:
----------------------------------

    Attachment: SMXCOMP-847.patch

> stflow handling/behaviour in the http consumer processor is wrong
> -----------------------------------------------------------------
>
>                 Key: SMXCOMP-847
>                 URL: https://issues.apache.org/jira/browse/SMXCOMP-847
>             Project: ServiceMix Components
>          Issue Type: Bug
>          Components: servicemix-http
>    Affects Versions: servicemix-http-2010.01
>            Reporter: Thomas Termin
>            Assignee: Thomas Termin
>         Attachments: SMXCOMP-847.patch
>
>
> The stFlow behaviour is wrong in the consumer processor. I call it behaviour because it is actually not only used if st flow is selected but also if the repsonse thread is faster then the request thread could go into suspend. It that latter race condition occurs it wont never work correctly again without a redeploy. The isSTFlow Parameter cannot be a global attribute because that would lead to wrong behaviour for all parallel requests. What makes it more bad is, that this attribute is never resetet back to false. So all follow requests will fail because the response is not present.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.