OBIA 11.1.10.2 configuration fails in Sun solaris 11 X86_64

Question

Hi - I am installing OBIA 11.1.1.10.2 with OBIEE 11.1.1.9 in Sun Solaris 11 Box.

1. The OBIEE configuration completed successfully.

2. When I try to configure the OBIA it is failing at 46% when it is trying to restart the bi_server1 Managed server.

3. When I try to start the manage server through commandline it is starting in running mode.

4. When the UI try to restart the manage services it is failing.

5. We have sufficicent memory SWAP space no issues.

6. We are using JDK1.7_79

7. Hit retry couple of times still the same issue.

java.lang.Thread.run(Thread.java:745)

java.lang.IllegalArgumentException: No enum constant oracle.as.install.bi.wls.ServerState.FAILED_NOT_RESTARTABLE

at java.lang.Enum.valueOf(Enum.java:236)

at oracle.as.install.bi.wls.ServerState.valueOf(ServerState.java:3)

at

Any thoughts ?

Krishna Srinivasan · Answer

Thanks Anjum. I appreciate your input.

We have resolved the issue. It seems the OBIA configuration was using 32 bit Java and it was not able to allocate more than 3 GB so during the restart process of managed server it failed. The below is the solution in solaris.

1. Make sure you are using 64 bit JDK.

2. Verify after install weblogic. In commenv.sh file it will update whether it is using 32 or 64 bit Java.

3. Make sure you set the JAVA_HOME env variable to 64 bit so the OBIA configuration will take it during the config.

Note bi_server1 managed server requires upto 9 GB os heap size so the JVM should be 64 bit.

Anjum Ara · Answer

Hi Krishna

We recently faced similar issue with exact same version.  I observed that if you wait for more than 1 hour before trying to start the service then the configurator does not pick the latest status of the service. Timing is very important.

I would recommend re-configuration of OBIEE and wait for step to reach "Starting BI_SERVER1". Quickly start the service and wait for 15-20min to pick the status as Running on its own. If it does not then click the check box on top left and then select retry.

Lastly, monitor the .out log file constantly (tail -f logfilename.out) for any issues .

Thanks

aa