Database agents such as SQL Server and Oracle jump between displaying OS metrics and showing no OS metrics in the workload of the database dashboard, but when you go into the homepage the database information the OS metrics are be available.
There may also be incorrect cluster node names displayed in the Global View's System Utilization Host column or the associated popup.
The most common reason for an agent not to function as expected is topology duplication. In this scenario, the agent was probably created more than once.
For example, the user has created the agent by using the installer wizard, then deleted the agent from Agent Status and created it again by using the Agent Status page. If not properly deleted from the topology, the agent will maintain two topologies:
Global View will address the first instance it finds, meaning the old and non/updated topology. The old topology is not active, and because of that the agent will not be shown in Global View.
Note that The Topologies are created in the FMS process and agent is running on the FglAM process. That explains why the agent is marked in green in Agent Status, while there is a failure on the FMS side.
This can also occur if a second topology is created due to a different hostname being registered by the DNS in (a result of a nslookup having being run in earlier DB cartridges) when agents were created.
RESOLUTION 1
Delete the duplicate topology using the Database Technical Support dashboard
RESOLUTION 2
Hostname issues, multiple hosts are responding to the same requests, especially in cluster environments
RESOLUTION 3
Intermittent collection issues or connection failures reaching the monitored host.
© ALL RIGHTS RESERVED. Feedback Terms of Use Privacy Cookie Preference Center