Cannot add a Ceph cluster

Troubleshoot failures when adding a Ceph cluster to the Multi-Cluster Manager.

Problem

Adding a Ceph cluster fails or results in a connection timeout.

Explanation

The Ceph cluster agent pushes data to the Multi-Cluster Manager. If the agent is not running or cannot reach the API endpoint, the cluster cannot be added.

Resolution

  1. Verify network connectivity from the Ceph cluster.
    ping -c 3 <mcm-appliance-ip>
  2. Verify that the API endpoint is reachable.
    curl -k https://<mcm-appliance-ip>/api/health
  3. Identify the agent service on the Ceph cluster.
    ceph orch ps | grep -i mcm
  4. Review the logs, as needed, to identify exceptions and errors.
    1. Review the agent log to identify onboarding, communication, and cluster-side errors.
      cephadm logs --name <service-name> --follow
    2. Review the API application log to identify any Python exceptions and application-level errors.
      tail /var/log/mcm/api/api.log
    3. Review the API service log to identify service startup and systemd-related issues.
      journalctl -u api.service