Clusterware trace Files

As we have gone through the processes so far, we have mentioned where the processes write their trace files, but I would like to spend a few lines talking about the logs that are included in Oracle 11g Release 2. In Oracle 11g Release 1, RDBMS instances, listeners, and so on, moved to using the Automatic Diagnostic Repository (ADR). The default repository home location is the ORACLE_BASE/diag directory, and the various process types have directories under that, such as asm, rdbms, and tnslsnr. When an error occurs, the ADR framework creates an incident and gives you the option of packaging up that incident using the ADR Command Interpreter (ADRCI) to send to Oracle Support.

Although I will not go into ADR here, because ASM is now under our GI home and may be a critical part of the clusterware, you need to know where to go for troubleshooting. The clusterware has not as yet followed the ADR model; however, the tracing has evolved significantly since 10 g Release 1. Most of the trace files are fairly low level and for use by Oracle Support to diagnose problems—only the cluster alert log is meant for user consumption in the same way as the RDBMS alert log is.