Tech Tip: Monitoring Oracle Database Servers
This month Anders provides insight on monitoring Oracle databases using Nagios.
Since Oracle’s database server is designed to meet a very broad range of usage scenarios, it can be quite time-consuming to maintain Oracle installations. The database administrator has to keep track of the current usage of several kinds of resources and extents, make sure that the SID-resolution works as expected, tune cache-settings, make sure that log archiving works properly, keep track of backup-modes, manually run test-queries to verify that the database works as expected for specific applications, and verify that transactions work as they should.
The individual tasks are not very time-consuming, but the hours add up when executed over and over again on several installations.
Luckily, there’s a Nagios plugin which can do most of the job for you – check_oracle.pl.
There are several Nagios plugins for Oracle available on the web. The one described in this tech tip was designed in co-operation with an Oracle DBA.
There are several nagios-plugins for Oracle available on the web. The one described in this tech tip was designed in co-operation with an Oracle DBA.
check_oracle.pl depends on Oracle Instant Client and a user with sufficient access to the databases you want to monitor. The plugin can monitor:
- database availability (tnsping)
- login or dummy login
- cache hit ratio
- tablespace usage
- datafile usage
- total number of datafiles
- available extents per table or tablespace
- log archiving enabled or not
- user defined query, match for string or numeric value
- broken transactions
- failed transactions
- transactions in the deferred error queue
- invalid transactions
The plugin supports placing authentication credentials in an auth-file and can handle hostname as input instead of using a .ora-file for SID-lookups.
The upstream version of the plugin is available from the original authors here: