Asynchronous SnapMirror disaster recovery basics

SnapMirror is disaster recovery technology, designed for failover from primary storage to secondary storage at a geographically remote site. As its name implies, SnapMirror creates a replica, or mirror, of your working data in secondary storage from which you can continue to serve data in the event of a catastrophe at the primary site.

If the primary site is still available to serve data, you can simply transfer any needed data back to it, and not serve clients from the mirror at all. As the failover use case implies, the controllers on the secondary system should be equivalent or nearly equivalent to the controllers on the primary system to serve data efficiently from mirrored storage.

Data protection relationships

Data is mirrored at the volume level. The relationship between the source volume in primary storage and the destination volume in secondary storage is called a data protection relationship. The clusters in which the volumes reside and the SVMs that serve data from the volumes must be peered. A peer relationship enables clusters and SVMs to exchange data securely.

Cluster and SVM peering express configuration

The figure below illustrates SnapMirror data protection relationships.



Scope of data protection relationships

You can create a data protection relationship directly between volumes or between the SVMs that own the volumes. In an SVM data protection relationship, all or part of the SVM configuration, from NFS exports and SMB shares to RBAC, is replicated, as well as the data in the volumes that the SVM owns.

You can also use SnapMirror for two special data protection applications:

How SnapMirror data protection relationships are initialized

The first time you invoke SnapMirror, it performs a baseline transfer from the source volume to the destination volume. The SnapMirror policy for the relationship defines the contents of the baseline and any updates.

A baseline transfer under the default SnapMirror policy MirrorAllSnapshots involves the following steps:

How SnapMirror data protection relationships are updated

Updates are asynchronous, following the schedule you configure. Retention mirrors the Snapshot policy on the source.

At each update under the MirrorAllSnapshots policy, SnapMirror creates a Snapshot copy of the source volume and transfers that Snapshot copy and any Snapshot copies that have been made since the last update. In the following output from the snapmirror policy show command for the MirrorAllSnapshots policy, note the following:

cluster_dst::> snapmirror policy show -policy MirrorAllSnapshots -instance

                     Vserver: vs0
      SnapMirror Policy Name: MirrorAllSnapshots
      SnapMirror Policy Type: async-mirror
                Policy Owner: cluster-admin
                 Tries Limit: 8
           Transfer Priority: normal
   Ignore accesstime Enabled: false
     Transfer Restartability: always
 Network Compression Enabled: false
             Create Snapshot: true
                     Comment: Asynchronous SnapMirror policy for mirroring all snapshots 
                              and the latest active file system.
       Total Number of Rules: 2
                  Total Keep: 2
                       Rules: SnapMirror Label     Keep  Preserve Warn Schedule Prefix
                              ----------------     ----  -------- ---- -------- ------
                              sm_created              1  false       0 -        -
                              all_source_snapshots    1  false       0 -        -

MirrorLatest policy

The preconfigured MirrorLatest policy works exactly the same way as MirrorAllSnapshots, except that only the Snapshot copy created by SnapMirror is transferred at initialization and update.

                       Rules: SnapMirror Label     Keep  Preserve Warn Schedule Prefix
                              ----------------     ----  -------- ---- -------- ------
                              sm_created              1  false       0 -        -