Recovering objects on the Minio server

Recovering buckets and objects on the Minio server in cases of disk/node failure

Recover

The DC/OS Minio service allows you to heal buckets and objects on the DC/OS Minio server, such as in the case of disk/node failure. You must heal the DC/OS Minio cluster if you intend to add new disks to the cluster.

Recover

  1. Run the following command to execute the recover plan:

    {
    dcos miniod --name=<service_name> plan start recover
    }
    
  2. The ‘recover’ task will run the mc heal command.

    mc admin heal -r minioclient
    

    Recover1

NOTE: The mc heal command is only applicable for a DC/OS Minio erasure coded setup (standalone and distributed). See Erasure Coding Scheme in Minio or the Minio Command Reference for more information.

Node replace

If you have replaced a node using node replace, then when the pod starts on a different node, the recover task will be executed.

Node failure

In case of node failure, when the node comes up again, then the recover plan will be executed. Specifically, the number of parity blocks and number of data blocks will be restored.

Recover

Figure 1. - Recovering the disk