Sign in
Nodes

Remove a node from the cluster entirely. The node must be drained first so its current containers have been rescheduled. Containers no deployment needs that may still exist on the node (a failed deployment's, ones retired while it was offline, or ones a redeploy replaced while it was unreachable) are removed first; if any cannot be, the removal is refused unless `force=true`, which records them as orphaned.

Changed on

The node's ID is taken out of every environment's (and project default's) target nodes, and the response lists them in placements_updated. When the node is the only target of a placement the removal is refused with 409 naming those environments, since emptying it would let them run anywhere.

delete/internal/nodes/{node_id}

Request

  • Base URL: /api (relative, and the document was published on a file host, not at the API’s origin)
  • URL: /api/internal/nodes/{node_id}
  • Auth: HTTP bearer

Path parameters

node_idinteger required

Node ID

Query parameters

forceboolean

Remove the node even if containers Temps placed on it could not be confirmed removed (e.g. the host is gone for good). Those containers are recorded as orphaned instead of being re-homed.

Response

Node removed

idinteger required
messagestring required
orphaned_containersstring[]

Containers Temps could not confirm were removed from the node before it was removed (only with force=true). They are recorded as orphaned; remove them on that host by hand if it still exists.

Changes

  • 1bfd419deb2fRevision changes
    • info

      added the optional property to the response with the status

  • 8fff5a8151baRevision changes
    • info

      added the new optional query request parameter force

    • info

      added the optional property to the response with the status