New to Oracle (and OEM), not new to SQL Server, I am testing our new installation of the SQL Server Plug-in on our development server.
The plug-in is the latest version 18.104.22.168.0. The OEM is version 10.2.0.5.0. The Agent is Oracleagent10gAgent version 10.2.0.5.0. The SQL Server is a standalone SQL Server 2005 – RTM (NOT SP3 – don’t ask!)
Monitoring of the SQL Server in the OEM is working well except for two conditions. I have opened cases with Oracle, but thought I’d see if anyone else has encountered these problems.
1) Alert not generated in OEM when data file hits threshold or is full. The datafile is set to 4MB, with restricted growth to 4MB. I’ll insert rows into this database, get the "1105 can’t allocate space error", but nothing is reflected in the OEM. Interesting point – thresholds/full-conditions are working for the log file. In the OEM, I have the Database metric - Database File Space Available (%) – set to <20 (warning) and <10 (critical). Is this correct metric?
2) Alert not generated in OEM when an error message is generated in the Windows Event application log. The "1105" error above, and others as well, aren’t being displayed in the OEM. The database metric – "Event Severity – Application" is set to ".*warning.*" and ".*error.*" respectively. A similar metric in the host metrics for the "System" Windows event log is working properly using these values.
Any thoughts or suggestions? Do you think applying SP3 may make a difference? If Oracle Support resolves these issues, I’ll post the resolution.