FAQ for NetApp Replication
This FAQ answers common questions about NetApp Replication. It focuses on concepts, terminology, and behavior that are useful when setting up, managing, and troubleshooting replication relationships between ONTAP systems, Cloud Volumes ONTAP, and Amazon FSx for ONTAP.
General concepts
Replication supports Amazon FSx for ONTAP, Cloud Volumes ONTAP, and on-premises ONTAP. You can configure replication between any of the three systems.
Replication supports three configurations:
-
Simple: replication occurs from volume A to volume B.
-
Fanout: replication occurs from volume A to multiple destinations.
-
Cascade: replication occurs from volume A to volume B, and then from volume B to volume C.
The baseline transfer includes a full copy of the source data. Subsequent transfers contain only differential (incremental) copies of the source data.
NetApp doesn't charge you for using Replication itself, but you should check with your cloud provider for applicable data ingress and egress charges. The NetApp Console incurs charges when you deploy Console agents to connect to your storage systems or data services; some other NetApp data services accessible from the Console are licensed or subscription-based.
You don't need a license or subscription to start using the Console or Replication. For on-premises ONTAP configurations, an active SnapMirror license must be installed.
Replication policies
There are three types of replication policies:
-
Mirror: replicates newly created snapshot copies to a destination volume, useful for disaster recovery preparation or one-time data replication; you can activate the destination volume for data access at any time.
-
Backup: replicates specific labeled snapshot copies to a destination volume and typically retains them longer than on the source volume, useful for restoring corrupted or lost data and for compliance/governance purposes.
-
Mirror and Backup: combines both disaster recovery and long-term retention, and is the default policy included with each system.
If the default policies don't meet your needs, you can create custom policies using System Manager.
A backup policy only replicates snapshot copies whose labels match the labels defined in the source volume's snapshot policy. If the labels in your custom backup policy don't match any labels in the snapshot policy, the system cannot replicate any snapshot copies. For example, a backup policy with the labels "daily" and "weekly" only replicates snapshot copies with those labels.
The default snapshot policy creates hourly, daily, and weekly snapshot copies, retaining six hourly, two daily, and two weekly copies. The default backup policies replicate daily and weekly snapshot copies; it retains 7 daily and 52 weekly snapshot copies.
Setting up replication
Before replicating data, confirm the following:
-
The source and target systems already exist (or create them).
-
Verify that the source and destination volumes are running compatible ONTAP versions; see compatible ONTAP versions.
-
Ensure you've met the system-specific requirements for data replication requirements.
The intercluster network should be configured so that cluster peers have pair-wise full-mesh connectivity, meaning each pair of clusters in a cluster peer relationship has connectivity among all of their intercluster LIFs. This applies when an ONTAP cluster with multiple LIFs is the source or destination.
Several default schedules are available when setting up replication. If you want a different schedule, you must create a new schedule on the destination cluster using System Manager.
Managing replication relationships
This can occur when the destination system fails and then comes back online. You must initialize the relationship from the destination system using System Manager or the CLI in order for data replication to resume according to the defined schedule.
These actions have different effects:
-
Break: breaks the relationship between source and destination volumes and activates the destination volume for data access. Typically used when the source volume can't serve data due to corruption, deletion, or an offline state.
-
Resync: reestablishes a broken relationship and resumes replication on schedule; the contents on the destination volume are overwritten by the source volume's contents.
-
Reverse Resync: reverses the roles of source and destination volumes, overwriting the original source volume's contents with the destination's contents. Any data written to the original source volume between the last replication and when it went offline is not preserved.
Deleting the data protection relationship between the source and destination volumes ends data replication between volumes, deletes the SnapMirror relationship from the destination site, and releases the relationship from the source site. It also breaks the SnapMirror relationship, briefly making the destination volume the primary site for read/write operations so the destination can be used independently after replication is removed.
An unlimited transfer rate might negatively impact the performance of other applications and your internet performance. You should limit the maximum transfer rate (specified in megabytes per second) when setting up replication, and you can edit it later from the Replication Status page.