[jira] Closed: (UIMA-1704) UIMA AS aggregate doesnt handle service broker restart

["Jerry Cwiklik (JIRA)" <ui...@incubator.apache.org>]

     [ https://issues.apache.org/jira/browse/UIMA-1704?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Jerry Cwiklik closed UIMA-1704.
-------------------------------

    Resolution: Fixed

Modified to send CAS through ProcessErrorHandler when a connection to a broker is lost and subsequently recovered. When retries are allowed, the code creates a new temp queue to a remote delegate and instantiates a new listener. Once the temp queue and a listener are created, the CAS is send to a delegate for processing. 

> UIMA AS aggregate doesnt handle service broker restart
> ------------------------------------------------------
>
>                 Key: UIMA-1704
>                 URL: https://issues.apache.org/jira/browse/UIMA-1704
>             Project: UIMA
>          Issue Type: Bug
>          Components: Async Scaleout
>            Reporter: Jerry Cwiklik
>             Fix For: 2.3AS
>
>
> If a delegate service broker is restarted, the aggregate client doesnt attempt to recreate a listener to a temp reply queue. This may cause a hang or a timeout in the aggregate client. In this scenario there are two brokers: one managing aggregate service queue and a second that manages delegate service temp reply queue. When the latter is restarted, the dispatch code does not attempt to create a new listener. The aggregate code should detect a failure of the broker, mark a delegate as FAILED and kill the listener. On subsequent CAS request, the aggregate should attempt to create a new listener and a new JMS Connection to a broker. Failure to re-establish the connection is subject to error handling as defined in deployment descriptor Process error configuration for the delegate. 

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