A stand-alone MinIO server would go down if the server hosting the drives goes offline. In contrast, a distributed MinIO setup with _m_ servers and _n_ drives will have your data safe as long as _m/2_ servers or _m*n_/2 or more drives are online.

For example, an 16-server distributed setup with 200 drives per node would continue serving files, up to 4 servers can be offline in default configuration i.e around 800 drives down MinIO would continue to read and write objects.

drives.  Therefore, the number of drives you present must be a multiple of one of these numbers.  Each object is written to a single erasure-coding set.

Minio uses the largest possible EC set size which divides into the number of drives given. For example, *18 drives* are configured as *2 sets of 9 drives*, and *24 drives* are configured as *2 sets of 12 drives*.  This is true for scenarios when running MinIO as a standalone erasure coded deployment. In [distributed setup however node (affinity)...

	driveUsedInodesMD = NewGaugeMD(driveUsedInodes,
		"Total used inodes on a drive", allDriveLabels...)
	driveFreeInodesMD = NewGaugeMD(driveFreeInodes,
		"Total free inodes on a drive", allDriveLabels...)
	driveTotalInodesMD = NewGaugeMD(driveTotalInodes,
		"Total inodes available on a drive", allDriveLabels...)
	driveTimeoutErrorsMD = NewCounterMD(driveTimeoutErrorsTotal,

| `minio_cluster_health_drives_offline_count`        | Count of offline drives in the cluster. <br><br>Type: gauge         |        |
| `minio_cluster_health_drives_online_count`         | Count of online drives in the cluster. <br><br>Type: gauge          |        |
| `minio_cluster_health_drives_count`                | Count of all drives in the cluster. <br><br>Type: gauge             |        |

|:------------------------------------|:--------------------------------------|
| `minio_cluster_drive_offline_total` | Total drives offline in this cluster. |
| `minio_cluster_drive_online_total`  | Total drives online in this cluster.  |
| `minio_cluster_drive_total`         | Total drives in this cluster.         |

## Cluster ILM Metrics

          "intervalFactor": 1,
          "legendFormat": "",
          "metric": "process_start_time_seconds",
          "refId": "A",
          "step": 60
        }
      ],
      "title": "Total Drives",
      "type": "stat"
    },
    {
      "datasource": {
        "type": "prometheus",
        "uid": "${DS_PROMETHEUS}"
      },
      "description": "",
      "fieldConfig": {

		percUtil:          "Percentage of time the disk was busy",
		usedBytes:         "Used bytes on a drive",
		totalBytes:        "Total bytes on a drive",
		usedInodes:        "Total inodes used on a drive",
		totalInodes:       "Total inodes on a drive",
		cpuUser:           "CPU user time",
		cpuSystem:         "CPU system time",
		cpuIdle:           "CPU idle time",
		cpuIOWait:         "CPU ioWait time",

	diskReserveFraction = 0.15

	// diskAssumeUnknownSize is the size to assume when an unknown size upload is requested.
	diskAssumeUnknownSize = 1 << 30

	// diskMinInodes is the minimum number of inodes we want free on a disk to perform writes.
	diskMinInodes = 1000

	// tlsClientSessionCacheSize is the cache size for client sessions.
	tlsClientSessionCacheSize = 100
)

func init() {

	beforeDrives := formatsToDrivesInfo(s.endpoints.Endpoints, formats, sErrs)

	res.After.Drives = make([]madmin.HealDriveInfo, len(beforeDrives))
	res.Before.Drives = make([]madmin.HealDriveInfo, len(beforeDrives))
	// Copy "after" drive state too from before.
	for k, v := range beforeDrives {
		res.Before.Drives[k] = v
		res.After.Drives[k] = v
	}

	if countErrs(sErrs, errUnformattedDisk) == 0 {

	return nil, fmt.Errorf("%w (offline-disks=%d/%d)", errErasureReadQuorum, disksNotFound, len(p.readers))
}

// Decode reads from readers, reconstructs data if needed and writes the data to the writer.
// A set of preferred drives can be supplied. In that case they will be used and the data reconstructed.