Running manual intervention for Investigation Dashboard issues

Understand when you need to run manual intervention for Investigation Dashboard issues, and which intervention steps to take.

Monitoring and automatic recovery attempts to fix the issues it finds by running several recovery actions. If these attempts fail, you need to intervene manually.

To figure out when to intervene and how, first go to the Investigation dashboard open issues report. In the report, check the most recent automatic repair action that displays next to each open issue. Then, refer to the manual intervention recommendation for that repair action in the following table. In cases where your intervention still does not resolve the issue, contact IBM Guardium support.

Last automated repair action Recommendation
Quick Search is disabled on the managed unit. The hardware requirements are not met. Awaiting manual action.

For more information, see minimum hardware requirements.
Quick Search is disabled on the unit. Awaiting manual action.

Enable the Investigation Dashboard with the GuardAPI command.

For more information, see this topic.
Quick Search is down. Attempt to restart Quick Search.

Restart Solr manually. Run the restart_solr grdapi.

For more information, see the restart_solr topic.

Verify that the managed unit is set up correctly on the central manager.

Unregister and register the unit. For more information, see the topics about registering and unregistering units.
Attempt to ping Quick search ports (8983, 9983).

Test connectivity in the cluster by running the GuardAPI test_solr_connectivity on the central manager.

For more information, see the test_solr_connectivity topic

Verify connectivity for ports 8983 and 9983 between the central manager and managed units.

Check on both internal (Guardium) and external firewalls.
Verify that the core hostname is set up correctly. Verify that the unit hostname is set up correctly. There are multiple ways to do so.
  • CLI command of store system hostname. For more information, see this topic.
  • Verify that the correct hostname appears on the central manager.

    On the central manager, go to Manage > Central Management > Central Management and check the hostname for your unit. If the hostname is incorrect, run Refresh Unit Info.

  • If neither of these methods resolve the issue, contact IBM support to run a clean Solr upgrade.
Validate that the unit type is set up correctly. If the issue remains unresolved after a few hours, contact IBM Guardium support.
The core was replicated and one of the cores can’t be reached. Attempt to remove the unreachable core. If the issue remains unresolved after a few hours, contact IBM Guardium support to re-create the core.
The core is down and the collection can’t be reached. Attempt to create a new core instead of the old one without copying the data. If the issue remains unresolved after a few hours, contact IBM Guardium support to re-create the core.
The core is down and the collection can’t be reached. Attempt to create a new core instead of the old one with the data from the old core.

If the issue remains unresolved after a few hours, contact IBM Guardium support to re-create the core.
The core was replicated and both cores can’t be reached. Attempt to create a new core instead of the old one without copying the data.

If the issue remains unresolved after a few hours, contact IBM Guardium support to re-create the core.
The core was replicated and both cores can’t be reached. Attempt to create a new core instead of the old one with the data from the old core. If the issue remains unresolved after a few hours, contact IBM Guardium support to re-create the core.
The cores are down for the specified collection. Attempt to create a new core instead of the old one without copying the data. If the issue remains unresolved after a few hours, contact IBM Guardium support to re-create the core.
The cores are down for the specified collection. Attempt to create a new core instead of the old one with the data from the old core. If the issue remains unresolved after a few hours, contact IBM Guardium support to re-create the core.
The unit is not listed in the collection’s cluster. Attempt to create a new core instead of the old one without copying the data. If the issue remains unresolved after a few hours, contact IBM Guardium support to re-create the core.
The unit is not listed in the collection’s cluster. Attempt to create a new core instead of the old one with the data from the old core. If the issue remains unresolved after a few hours, contact IBM Guardium support to re-create the core.
Attempt to restart quick search.

Unregister and register the unit. For more information, see the topics about registering and unregistering units.

The creation of the collection failed because the collection configuration is incorrect. Awaiting manual fix.

 If the issue remains unresolved after a few hours, contact IBM Guardium support.
The configuration manager is running on the managed unit. Attempt to stop the configuration manager process on the unit. If the issue remains unresolved after a few hours, contact IBM Guardium support.
The unit is listed in the incorrect configuration manager. Restart Solr manually. Run the restart_solr gradapi. For more information, see the restart_solr topic.
The indexing process is not scheduled. Attempt to resume or reschedule the indexing job. Restart the UI manually. For more information, see the restart GUI section in Configuration and control CLI commands .
Verify that the indexing process ran in the last hour. Restart the UI manually. For more information, see the restart GUI section in Configuration and control CLI commands.

The indexing process failed.

Awaiting manual fix.

 If the issue remains unresolved after a few hours, contact IBM Guardium support.