mc admin uncordon

Syntax

The mc admin uncordon command returns a cordoned node to active service after maintenance. This command notifies all other nodes in the cluster that the specified node is returning to service.

When you run mc admin uncordon, the command:

  • Removes the node from the cluster’s internal list of cordoned nodes.
  • Notifies all peer nodes that the target node is returning to service.
  • Allows peer nodes to re-establish grid connections to the node.

See Node Maintenance for a complete procedure on cordoning and uncordoning nodes.

Restart the cordoned node and run uncordon from a different node

The uncordon command only updates the cluster’s cordon state and notifies peer nodes. It does not start or restart the node.

A cordoned node persists its cordon state and re-applies the cordon on every startup. Simply restarting the MinIO AIStor process on a cordoned node does not rejoin it to the cluster — you must also run mc admin uncordon against the cluster (from a node other than the cordoned one) to clear the persisted state. The restart and the uncordon can occur in either order; both steps are required.

mc admin [GLOBALFLAGS] uncordon ALIAS NODE

Parameters

ALIAS

Required

The alias of the MinIO AIStor cluster containing the node to uncordon.

NODE

Required

The address of the node to uncordon, in the format hostname:port.

For example: node1.example.com:9000

Global flags

This command supports any of the global flags.

Examples

Uncordon a node after maintenance

The following example demonstrates the complete workflow for returning a cordoned node to service:

  1. Restart the MinIO AIStor process on the cordoned node:

    sudo systemctl restart minio

  2. Uncordon the node:

    mc admin uncordon myaistor node1.example.com:9000

The command returns:

Node uncordoned successfully

Verify node status after uncordoning

After uncordoning, verify the node has rejoined the cluster:

mc admin info myaistor

The node should display as Online in the command output.

Permissions

This command requires the admin:ServiceCordon permission.

Behavior

Error handling

If the uncordon operation encounters errors on some peer nodes, the command returns with state uncordoned-with-errors and lists the affected nodes. Common causes include:

  • The cordoned node was not restarted before uncordoning.
  • Network connectivity issues between nodes.
  • The target node is still initializing.

If you receive errors indicating the node must be restarted first, ensure the MinIO AIStor process is running on the target node and try the uncordon command again.

State persistence

After a successful uncordon, MinIO AIStor removes the node from the persisted cordon state in .minio.sys/cordoned.json. The node fully rejoins cluster operations after reconnecting to the grid.

Node state after uncordon

mc admin uncordon makes the cluster willing to accept the node back; it does not bring the node up. If you run mc admin uncordon before restarting the MinIO AIStor process on the target node, the node continues to display as Offline in mc admin info until the process is restarted and the grid connection is re-established.