Replication & Disaster Recovery

Replicate volumes between MayaNAS or MayaScale nodes for disaster recovery and data protection. MayaNAS includes enterprise-grade DR features at no additional cost — same posture as legacy NAS appliance products that charge per-protected-TB.

Access

Navigate to Replication in the Web UI sidebar.

Replication Types

Type	RPO	How It Works	Best For
Snapshot Mirroring (SnapMirror equivalent)	Configurable from minutes to days	Asynchronous incremental ZFS replication via `zfs send/recv` over encrypted SSH with on-the-wire compression. Resumable on interruption. Per-volume retention policies on source and target	Recommended for DR. Cross-region, cross-cloud, on-prem capable. No per-TB license
Synchronous (DRBD)	Zero data loss (RPO 0)	Writes acknowledged only after stable write on both nodes via DRBD block-level replication	Same-region HA for block volumes (RDBMS, VMware datastores)
Asynchronous (safe)	~5 seconds	Writes acknowledged after secondary receives data in memory	Cross-region block replication with low-latency tolerance
Asynchronous (unsafe)	~5 seconds	Primary does not wait for secondary acknowledgment	Maximum performance, tolerates some data loss

Snapshot Mirroring is the recommended DR path for MayaNAS file/object workloads — it is the OpenZFS-native equivalent of NetApp SnapMirror, runs at the filesystem level, supports cross-region/cross-cloud/on-prem destinations, and is included with the product (no per-TB license fee).

Enable Replication

Select a volume by checking its checkbox
Click Enable
Configure the replication pair:

Field	Description
Remote Server	IP address of the secondary node
Remote Volume	Destination volume on the secondary node
Replication Type	`sync`, `async-safe`, `async`, or `snapshot`
Network Interface	Preferred interface for replication traffic
Sync Rate	Bandwidth allocation percentage (1–100%)

Click Finish

Schedule Replication

For snapshot-based replication, set a schedule:

Select a replicated volume
Click Schedule
Configure:

Field	Description
Interval	`frequent` (minutes), `hourly`, `daily`, or `monthly`
Period	Number of intervals between replications
Keep	Number of replication snapshots to retain

Click Save

Start / Stop Replication

Start — Run replication on demand. Status auto-refreshes while running.
Stop — Stop an active replication.

Status progresses through: Initializing > Replicating > Updated.

Disable Replication

Select the replicated volume
Click Disable
Confirm deletion of the replication pair

Planning

Synchronous Mirroring Requirements

Identical storage capacity on primary and secondary
Low-latency network between nodes
Dedicated network interface recommended for replication traffic

Asynchronous Mirroring

Secondary can have different storage performance characteristics
Safe mode ensures data received in memory before acknowledging
Typically ~5 seconds of data at risk

Snapshot-Based Replication

Works over any network (including WAN)
Minimum 1-minute RPO with frequent snapshot policy
Suitable for cloud object storage connections
No requirement for identical storage

CLI

# Create a synchronous replication pair
mayacli create replication tank/data \
  remote=server2:tank/data_mirror type=sync \
  controller=eth1 syncrate=50

# Create a snapshot-based replication pair
mayacli create replication tank/data \
  remote=server2:tank/data_mirror type=snapshot

# Show all replication pairs
mayacli show replication

# Show replication for a specific volume
mayacli show replication tank/data

# Start replication (on demand)
mayacli bind replication tank/data

# Stop replication
mayacli unbind replication tank/data

# Set a replication schedule (daily, keep 7)
mayacli set replication tank/data interval=daily period=1 keep=7

# Set frequent replication (every 5 minutes, keep 12)
mayacli set replication tank/data interval=frequent period=5 keep=12

# Disable replication schedule
mayacli set replication tank/data interval=none

# Delete a replication pair
mayacli delete replication tank/data

See CLI Guide for the full command reference.