We migrated a lot of old and stable applications from an Application Sever of a different Vendor to JBoss AS 7.1.1.final.
On the previous application server the same applications ran for many years without raising any problems.
Please find below initial info about the configuration:
• OS: Linux Red-Hat
• Application Server: JBoss AS 7.1.1.final
• Ver. JDK: 188.8.131.52
• JAVA_OPTS="$JAVA_OPTS -Xms2048m -Xmx2048m -XX:MaxPermSize=2048m -Dsun.rmi.dgc.client.gcInterval=3600000 -Dsun.rmi.dgc.server.gcInterval=3600000"
On an Application Server of the production environment on which about 42 applications (EJB) run with periodic and constant scheduling, a complete application server crash occurred. In this situation, the application server is completely unavailable and inactive.
Analyzing the log file (client log) we can find the following error.
08/12/2020 10:00:43 70179 No EJB receiver available for handling [appName:BatchProd,modulename:ModuleEJB,distinctname:] combination for invocation context org.jboss.ejb.client.EJBClientInvocationContext@2d17d41d
From the error, according to our understanding, it is clear that the problem is inherent to the saturation of JNDI resources.
In order to identify in a deterministic manner the cause that is generating this application server failover, I could kindly ask for the procedures to be activated to analyze the logs and how to monitor the usage and consumption of JNDI resources and/or identify the proper configuration that should avoid the problem.
This is a blocking problem as it occurs randomly in the production environment.