Hi, Darren
Was there any exception/stack trace in the server.log when you
deployed application 2 first? And was there any exception/stack trace in
the server when the server was restarted and loading application 2 (for
the scenario where the application 2 was deployed first)?
- Hong
>I have an unusual deployment problem with Glassfish UR1 P01, which happens on both Solaris and Mac OSX.
>
>I have 2 applications that I deploy onto the same instance of Glassfish. They both have their own HTTP listener and virtual server. Application 1 is a customer facing website, application 2 processes incoming messages and do scheduled processing at varous intervals, fired by some cron jobs. The method of communication between the 2 applications is via JMS and the shared MySQL DB.
>Both these applications were ported over from SJSAS 8.2 PE, and ran without problems.
>
>The problem I have is that when I deploy these applications I have to deploy application 1, before application 2. If I deploy application 1 after 2 then next time I restart Glassfish application 2 does not start correctly. The servlets and EJBs do not start in application 2, even though the servlet filters in that applications webapp are running as I can see the output to the log4j log when the crons are firing.
>If I disable application 2, and then enable it again then the application starts correctly, and if I redeploy application 2 after the restart then it also starts correctly.
>
>Does anyone have any ideas as to why this might happen? Has anyone else had a similar problem and get anywhere with investigating it? Any light anyone can shed on this problem is greatly appreciated.
>
>Thanks in advance,
>Darren.
>[Message sent by forum member 'xerces' (xerces)]
>
>http://forums.java.net/jive/thread.jspa?messageID=213653
>
>---------------------------------------------------------------------
>To unsubscribe, e-mail: users-unsubscribe_at_glassfish.dev.java.net
>For additional commands, e-mail: users-help_at_glassfish.dev.java.net
>
>
>