Troubleshoot UCS device monitoring in the Orion Platform

This topic describes about issues you may encounter when monitoring UCS devices in the Orion Platform. For additional information, visit the SolarWinds Success Center and our online IT community, THWACK.

Be sure to review UCS device monitoring requirements.

Issue: Polling does not return data for UCS fabric interconnects.

Verify the following items:

  • Make sure a child node for the fabric interconnect exists beneath the parent node for the UCS controller. See Add nodes for devices connected to UCS controllers to the Orion Platform.
  • For fabric interconnects, the polling method must be SNMP.
  • Verify that fiber interconnects have external IP addresses. If the external gateway, external IP address, or external mask are set to, update the IP address.
  • The Orion Platform main polling engine or Additional Polling Engine (APE) — whichever polls the UCS cluster — must have access to the UCS API.

Issue: When setting up a UCS blade as a node in the Orion Platform, you select the Operating System Interface but no change occurs.

As you add a node for a UCS blade, you must specify which network interface to use to monitor the blade:

  • Management Interface: Use the Management Interface to monitor the blade. When you add the node, the IP address automatically populates with information captured from the parent BMC controller during SNMP polling.
  • Operating System Interface: Use this method to monitor the blade via its build-in operating system that returns a broader range of data, such as hardware health. To add the node, you'll need to provide its IP address because polling occurs in a layer of the OSI model that does not capture IP addresses.

If you select the Operating System Interface option but no change occurs, the blade probably has existing connections to other Orion Platform entities based on its IP address, as configured in the following product versions:

  • NPM 12.4 and earlier
  • SAM 6.7.1 and earlier
  • VMAN 8.3 and earlier

If the blade was already associated with other nodes in earlier product versions, you cannot switch the network interface used for the blade from the Management interface to the Operating System interface. This logic preserves existing node connections that are based on the blade IP address provided by the Management interface.

To work around this functionality, remove the blade node from the Orion Platform and then add it back again.