Monitoring subsystems on Kubernetes

Use kubectl get to check the health of the API Connect clusters in your Kubernetes deployment.

The APIConnect operator controls the life-cycle of management subsystem microservices. When you install, you create the following custom resources (CRs) in order to install API Connect and all subsystems:

  • ManagementCluster or mgmt
  • AnalyticsCluster or a7s
  • PortalCluster or ptl
  • GatewayCluster or gw

    Every CR has a status section. The APIConnect operator regularly updates this section with the current status of the subsystem.

For example, use the following command to obtain the status of ManagementCluster:

kubectl get mgmt -n <namespace>

Example:

kubectl get mgmt
NAME   READY   STATUS    VERSION      RECONCILED VERSION      AGE
m1     16/16   Running   10.0.8.0      10.0.8.0-44   24m

Output:

  • NAME: Name of the Management Subsystem
  • READY: Current number of management subsystem which are up, followed by the expected number of management subsystem components which should be up. For example: 16/16.
  • STATUS: Current status of Management Subsystem

    During fresh install, upgrade or any maintenance operation, such as changing configuration for S3 backups, Management CR status can go to different states depending on the action, but ultimately it should always reach Running status.

    If the status says Warning, then run kubectl describe mgmt -n <namespace> and check for more details in the Status section.

  • VERSION: Version chosen in spec.version in the management CR.
  • RECONCILED VERSION: Current management subsystem version. Includes a build number.
  • AGE: Age of the management subsystem CR.