[jira] Commented: (GBUILD-16) Improve GBuild agent JMS reconnect logic

["Kevan Miller (JIRA)" <de...@geronimo.apache.org>]

    [ http://issues.apache.org/jira/browse/GBUILD-16?page=comments#action_12369714 ] 

Kevan Miller commented on GBUILD-16:
------------------------------------

Here's some info on what ActiveMQ provides -- http://activemq.codehaus.org/Failover+Transport+Reference

It seems to provide for the configuration of a retry backoff strategy. Assuming it all works, it should provide the desired function -- fast recovery for short-term problems and long-term recovery for long outages. I don't view the backoff capabilities as super critical, anyway. We could set a fixed retry rate of 30 seconds with few negative consequences... 

Apologies if I was overly harsh in my summary of the problems...

> Improve GBuild agent JMS reconnect logic
> ----------------------------------------
>
>          Key: GBUILD-16
>          URL: http://issues.apache.org/jira/browse/GBUILD-16
>      Project: GBuild
>         Type: Bug
>   Components: agent
>     Versions: 1.0.0
>     Reporter: Kevan Miller
>     Assignee: Kevan Miller
>     Priority: Minor

>
> The gbuild agent reconnect logic has several problems.
> User configuration settings for reconnect processing are ignored. Max retries and delay time are all hard-coded. 
> If the initial connection fails, it seems that the client will never connect.
> Also, we should be able to configure agents to attempt to reconnect forever. There should be some retry backoff scheme or bimodal reconnect interval... 
> I seem to recall that ActiveMQ has similar features in their client api. It might be simpler to remove what GBuild currently does and use the built-in ActiveMQ support...
>   

-- 
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
   http://issues.apache.org/jira/secure/Administrators.jspa
-
For more information on JIRA, see:
   http://www.atlassian.com/software/jira