Skip to content
PrexorCloud

Cluster Commands

These commands give you situational awareness across the cluster: the list of attached nodes, an at-a-glance overview, and the controller + CLI version pair.

What you’ll learn

  • How to list and drain nodes.
  • The difference between the dashboard-style status overview and a raw REST query.
  • How to obtain controller and CLI versions for support tickets and CI.

prexorctl status

Cluster-wide overview. Renders three summary columns (cluster, nodes, instances), a live-metrics card with sparklines, and a per-group status table. With --json, returns the byte-identical /api/v1/overview response so scripts continue to work.

Terminal window
prexorctl status
prexorctl status --json

Flags: none — uses the global --json / --controller / --token.

prexorctl node

Cluster-node management. A “node” is a daemon process registered with the controller; one daemon = one node.

node list

Terminal window
prexorctl node list
prexorctl node list --state ONLINE
prexorctl node list --state DRAINING --json

Flags:

  • --state <state> — filter by state. Valid values: ONLINE, DRAINING, UNREACHABLE, OFFLINE.

The default rendering shows id, status pill, CPU %, memory used/total, running-instance count, and connected-since timestamp. Footer counts group nodes by state.

node info

Terminal window
prexorctl node info <id>

Detailed view: status, connected-since, resource card (CPU, memory, free disk, instance count), and the per-instance table for instances currently running on this node.

node drain

Terminal window
prexorctl node drain <id>

Marks the node as DRAINING. The controller stops scheduling new instances onto it; existing instances keep running until they drain naturally or you stop them explicitly.

node undrain

Terminal window
prexorctl node undrain <id>

Returns a draining node to ONLINE. The scheduler resumes placing new instances.

prexorctl version

Prints the CLI version, Go runtime, OS/arch, and — when authenticated — the controller’s reported version. Useful when filing bugs.

Terminal window
prexorctl version
prexorctl version --json

The JSON form merges CLI and controller_* keys into a single object suitable for diagnostics bundles.

Scripting examples

Drain a node, wait for instances to clear, then power it down:

Terminal window
prexorctl node drain node-3
while [ "$(prexorctl node info node-3 --json | jq '.instanceCount')" -gt 0 ]; do
  sleep 5
done
prexorctl node info node-3

Bail out of CI if the cluster isn’t healthy:

Terminal window
HEALTHY=$(prexorctl status --json | jq -r '.degraded')
[ "$HEALTHY" = "false" ] || exit 1

Next up