Use APIs to monitor CPU usage and network latency.

Checking CPU Usage

These APIs work similar to the top command:

  • Run the following API to check process status for all appliances:
    GET /api/v1/systemhealth/appliances/process/status
  • Run the following API to check the process status of a specific appliance.
    GET https://<nsx-manager-ip>/api/vi/systemhealth/appliances/<appliance-id>/process/status
    

Checking Network Latency

  • Run the following API to check for network latency values for all appliances:
    GET /api/v1/systemhealth/appliances/latency/status
  • Run the following API to check for network latency values for a specific appliance:
    GET https://<nsx-manager-ip>/api/v1/systemhealth/appliances/<appliance-id>/latency/status

See the latest version of the NSX API Guide at https://code.vmware.com/ for API details.