NSX-T Bare Metal Edge Node CPU alarm with UUID of all 0's

Products

• VMware NSX Networking

Issue/Introduction

Symptoms:

• You receive alarms for an NSX-T bare metal edge node, indicating the CPU is very high.
• The NSX-T bare metal edge node UUID for these alarms is all 0's:
  • 00000000-0000-0000-0000-000000000000
• These alarms are present, even when the NSX-T bare metal edge node is placed in maintenance mode.
• There is little or no traffic on the NSX-T bare metal edge node.

Environment

VMware NSX-T Data Center 3.x
VMware NSX-T Data Center

Cause

• During registration of the NSX-T bare metal edge node, the CPU of the NSX-T bare metal edge node may become high and cause an alarm to be generated.
• Before registration is complete the NSX-T bare metal edge node UUID is all 0's:
  • 00000000-0000-0000-0000-000000000000
• When registration is complete, the NSX-T bare metal edge node will receive a correct UUID.
• Later, even after registration is complete, trying to acknowledge the alert may fail.
• This is because the NSX-T bare metal edge node with UUID of all 0's no longer exists.

Resolution

This issue is resolved in NSX-T 3.1.2.

Workaround:

• You can use an API to remove these stale alarms.
• To run this API call, log into the NSX-T manager as root and run the following API call to get the alarms:
  • curl -X GET -H 'X-NSX-Username:admin' -H 'content-type: application/json' -ik http://localhost:7449/phonehome-coordinator/api/v1/alarms
• From the results identify the stale alarm and use the UUID of the alarm in the next Delete API call:
  • curl -X DELETE -H 'X-NSX-Username:admin' -H 'content-type: application/json' -ik http://localhost:7449/phonehome-coordinator/api/v1/alarms/<alarm-id>
• After triggering DELETE alarm, please restart the napi by using the following command:
  • "/etc/init.d/nsx-mp-api-server restart" cmd