Troubleshooting the Control Plane

The Control Plane produces symptoms the exit code alone does not name, and the troubleshooting guide carries the exit-code contract itself.

Timeouts on a log follow

One exit 3 timeout is specific to the Control Plane:

• A controlplane --follow poll hitting its own fixed 30-second bound. The follow as a whole has no bound and ignores --wait-timeout, but each of its log polls stops at 30 seconds regardless of --timeout.

A failed asynchronous operation

One Control Plane write leaves the reason somewhere other than the resource it created:

• A create that appeared to succeed and produced nothing usable. Read the task, not the resource. A failed Control Plane database create in particular leaves a failed database with no reason attached to it, so pgedge controlplane task get --database <id> <task-id> is what prints the reason, and task logs prints the step trace.

The tasks and async operations guide covers finding that task and reading its outcome, and managing Control Plane databases covers creating the database. The incident runbooks take a failed operation, a stopped instance, a full disk, slow queries, replication lag and an unreachable host or Control Plane through the reads that narrow each one, and the monitoring and alerts guide covers catching them on a schedule.