Data Guard monitoring: setup and troubleshooting
How to set up the optional Data Guard management pack, what it monitors, and how to solve common problems, including mounted standby databases.
Applies to: Opslogix Oracle Management Pack (all versions)
Last updated: October 2026 · Latest version: 26.4.4517.0
See also: Upgrade guide for 26.4.4517.0 · Version support · Release notes
Setup
- Import the management pack "OpsLogix.IMP.Oracle.DataGuard" from the Management Packs folder of the release.
- Make sure the monitoring account has the Data Guard grants listed in KB: How to create the monitoring account for the Oracle monitoring.
- Add the primary and the standby instances in the "Oracle IMP Configuration Dashboard" (Administration pane, Add/Remove Monitoring Instance tab). Select the DBA role SYSDBA or SYSDG. This privilege is needed to monitor standby instances.
- Assign a license to each instance (Assign tab).
The Data Guard primary and standby objects are discovered on the Oracle instances every 5 minutes.
Mounted standby databases
From V26.4.4517.0, mounted standby databases are supported:
- For SYSDBA connections to a mounted standby database, enter the SID of the standby database in the Instance Name field, not the service name.
- Oracle does not allow ALTER SESSION or DBMS_APPLICATION_INFO on a mounted standby (ORA-01219). The pack handles this; you don't need to do anything.
What is monitored
| Monitor | Runs on | Default |
|---|---|---|
| Apply lag | Standby | Critical when the apply lag is 30 seconds or more |
| Transport lag | Standby | Critical when the transport lag is 2 seconds or more |
| Apply finish time | Standby | Critical when the apply finish time is 1 second or more |
| Archive gap | Standby | Critical when there is 1 or more archive gap |
| Standby data files | Standby | Status of the standby data files |
| Media recovery process | Standby | Status of the media recovery process |
| Logical recovery process | Standby | Status of the logical recovery process |
| Archive destination status | Primary | Status of the archive destinations |
| Standby redo log files | Primary | Status of the standby redo logs |
All monitors run every 5 minutes. Apply lag, transport lag and apply finish time are also collected as performance data every 15 minutes, and shown in the Data Guard performance view. You can change thresholds and intervals with overrides.
Troubleshooting
- No Data Guard objects are discovered: check that the DataGuard management pack is imported, that the instance was added with the SYSDBA or SYSDG role, and that the instance has a license assigned.
- The standby instance shows a connection error: for a mounted standby, use the SID instead of the service name (see above). The connection alert contains the Oracle error message; see the article "Troubleshooting: Instance connection alerts and common ORA errors".
- Properties such as DB_UNIQUE_NAME or PROTECTION_MODE are empty: upgrade to V26.4.4517.0 or later; earlier versions did not discover these properties.