Drain Node
Exclude a node from new placement and migrate its data off before removing it.
| Endpoint | Method | Admin Required |
|---|---|---|
/api/v1/admin/cluster/nodes/:nodeId/drain | POST | Yes |
Marks a node as draining. It's immediately excluded from new object placement, and a cluster-wide rebalance walk is triggered to migrate its data onto the remaining active nodes. Any outstanding replication task that exists to keep this node in sync as a responsible replica (a write straggler catch-up, or a rebalance copy) is cancelled at the same time, since draining excludes it from placement - continuing to retry those would just be wasted effort. Tasks for this node cleaning up its own now-unowned copies are left alone; that's exactly the mechanism that empties its disk before it's finally stopped.
Draining doesn't stop the node's process. Watch rebalance status drain to zero pending/failed tasks before actually stopping it - otherwise data it was still the sole holder of could be lost.
Path Parameters
nodeId
string required
The UUID of the node to drain, from List Nodes.
Response
Returns 200 OK on success.
Errors
| Status | Reason |
|---|---|
404 | Node not found |