DBCA failing to create database on RAC
We have a 2-node Linux RAC, using ASM storage. We have 2 10.2.0.3 ORACLE_HOME 's (dev and tst) with a total of 5 databases, plus ASM instance (in separate OH). We are trying to create another database and dbca appears to be failing to copy the database to 2nd node. DBCA creates the database instance on node 1 and then seems to hang for about 40 minutes, then does a shutdown abort. All datafiles are left on ASM storage, but we cannot start the database as there is no spfile or init file created.
Whe running dbca, we have looked at the default init parameters and it is set to cluster_instances=2, and we select ASM storage.
We have had an SR open with support for 2 weeks and they cannot reproduce in-house. From the dbca trace.log file, we find entries where it appears to be trying to create redo logs in $ORACLE_BASE/oradata/sid and not on ASM storage, even though the online redo logs are actually created on ASM.
I can create a single instance on each node, using ASM, so it is not system resources. I am wondering if it is the services on the 2nd node.
The crs_stat show all is online and available
Has anyone had this problem or offer any advice on where to search for other errors that may be the actual cause?