A warning message is logged every 30 seconds "CM plugin status not found for <CM UUID>, plugin may not be running on this node" on 2 Manager nodes for a Manager cluster configured with 3 nodes
search cancel

A warning message is logged every 30 seconds "CM plugin status not found for <CM UUID>, plugin may not be running on this node" on 2 Manager nodes for a Manager cluster configured with 3 nodes

book

Article ID: 384007

calendar_today

Updated On:

Products

VMware NSX

Issue/Introduction

  • You configure 3 node Manager cluster.
  • On 2 Manager nodes out of 3, you see a warning message is logged every 30 seconds.

    /var/log/cm-inventory/cm-inventory.log
    <Timestamp>  WARN inventoryTasksScheduler4 CmInventoryService 84750 FABRIC [nsx@6876 comp="nsx-manager" level="WARNING" subcomp="cm-inventory"] CM plugin status not found for <CM UUID>, plugin may not be running on this node.

    /var/log/syslog
    <Timestamp> <Hostname> NSX 84750 FABRIC [nsx@6876 comp="nsx-manager" level="WARNING" subcomp="cm-inventory"] CM plugin status not found for <CM UUID>, plugin may not be running on this node.

  • If you have multiple compute managers configured, the warning is logged for each compute manager on 2 Manager node.
  • The nodes may vary by compute manager.

Environment

VMware NSX 4.x

Cause

Only one NSX Manager node runs the CM plugin for each Compute Manager. The other two nodes log a message every 30 seconds saying the plugin is not running. This is expected by design.

Resolution

Starting with NSX 4.2.2 and VCF 9.0, the log level for this message has been changed from "WARNING" to "INFO" to reflect that it is not an issue.
In earlier versions, if the warning is seen on 2 out of 3 NSX Manager nodes for a specific Compute Manager, it is expected behavior and can be safely ignored.

Additional Information

Please contact Broadcom support in cases below which are unexpected.

  • The warning is logged for a specific compute manager on all the Manager nodes at the same time.
  • The warning is NOT logged for a specific compute manager on 2 or more Manager nodes at the same time.
  • It's fine that the warning for different compute managers are logged on different 2 Manager nodes.