MinIO — Operations¶

Scope

Production deployment, multi-tenant setup, erasure coding, lifecycle policies, and monitoring.

Deployment Patterns¶

Distributed Mode¶

# 4-node distributed deployment (minimum for erasure coding)
minio server http://minio{1...4}/data{1...4} \
  --console-address ":9001"

Topology	Nodes	Drives/Node	Erasure Set	Parity
Minimal HA	4	4	16 drives	EC:4 (default)
Production	8	8	64 drives	EC:4
Large scale	16+	12	192+ drives	EC:4-8

Kubernetes Operator¶

# Install MinIO Operator
kubectl apply -f https://raw.githubusercontent.com/minio/operator/master/resources/base/crds/minio.min.io_tenants.yaml
helm install operator minio/operator -n minio-operator --create-namespace

Bucket Lifecycle¶

{
  "Rules": [
    {
      "ID": "expire-old-versions",
      "Status": "Enabled",
      "NoncurrentVersionExpiration": {
        "NoncurrentDays": 30
      }
    },
    {
      "ID": "transition-to-tier",
      "Status": "Enabled",
      "Transition": {
        "Days": 90,
        "StorageClass": "GLACIER"
      }
    }
  ]
}

Performance Tuning¶

Setting	Default	Tuned	Impact
Erasure parity	EC:4	EC:2 (speed) or EC:8 (durability)	Write speed vs redundancy
Batch replication	OFF	ON	Cross-site data protection
Scanner speed	default	fast	Faster healing at cost of CPU

Monitoring¶

# MinIO client diagnostics
mc admin info myminio
mc admin prometheus generate myminio

# Key Prometheus metrics
minio_s3_requests_total
minio_node_disk_used_bytes
minio_node_disk_free_bytes
minio_inter_node_traffic_sent_bytes

Common Issues¶

Issue	Diagnosis	Fix
Drive offline	`mc admin info`	Replace drive, healing auto-starts
Slow uploads	Network or disk I/O	Check inter-node bandwidth, use NVMe
Bucket policy denied	`mc admin policy info`	Review IAM policy attachment
Disk full	`mc admin info --json`	Add drives, enable lifecycle deletion