Administrator Guide

• Both groups must be running PS Series rmware version 7.0 or later, and the clusters on those groups must be running FS Series
rmware version 3.0 or later.
To perform single-step failback to primary:
1. Click NAS, expand NAS Cluster, and expand Local Containers.
2. Select the recovery container.
3. Click Failback to primary. The Failback to Primary message is displayed.
4. Click Yes. The Replicate Recovery Container message is displayed.
5. Type the account name and password for the array on the primary site and click OK.
To monitor the failback operation on the recovery container, display the Alarms and Operations toolbar and click the Failback
Operations tab.
About NAS Replica Containers
NAS replica containers reside on the remote (destination) replication partner cluster and contain replicated data from the source
container. As with replica volumes, you can temporarily or permanently promote replica containers to recovery containers and grant
hosts access to recovery container data.
Unlike source containers, you cannot create snapshots of replica containers. However, any replicas of the source container are
replicated to the destination partner.
The following considerations apply to replica containers:
• The NAS container reserve on the replication partner must have enough free space to store the replica.
• Unlike volume replication, which retains a replica set of previously created replicas, the system retains only the current replica of
the source container in a NAS container replication conguration. To roll back to a previous point in time, you must use
snapshots.
• You can either replicate the container to a new replica container or you can specify that the container be replicated to an existing
replica container.
• The system assigns names to new replica containers as follows:
– The group gives the replica container the name of the source container, with a randomly generated four-digit string
appended to the end of its name. For example, if you replicate a container named ContainerA, the replica might be named
ContainerA_8970.
– If the length of the container name, combined with that of the appended string, is more than 229 characters, the system
shortens the replica container name to be 229 characters long. Fewer characters are accepted for this eld if you type the
value as a Unicode character string, which takes up a variable number of bytes, depending on the specic character.
• If you replicate to an existing container:
– The destination container must either be an empty container containing no data, or a container to which the source container
has previously replicated.
– The source and replica containers do not need to be the same size. Whenever replication occurs, the group automatically
adjusts the size of the replica container so that it either shrinks or grows to match that of the source container.
– All data previously residing on the container is overwritten and cannot be recovered.
• If read-only access is enabled on a replica container, it appears in the list of containers on the destination cluster.
NOTE: Replica containers count toward the total number of containers in the group. See the Dell EqualLogic PS
Series Storage Arrays Release Notes for the maximum number of containers allowed in a group.
• Container Replication of conguration information about SMB shares and NFS exports from a NAS container to a replica
container might be delayed for several minutes. This delay will have no eect on the replication of the data itself, but might aect
your ability to access the data because you might have to recreate recent conguration updates. If you replicate immediately, the
rst replication might not have all the data, but subsequent replications will.
Replicate to Another Container in a Cluster
If the original source container is no longer available, you can congure the recovery container to replicate to another container in the
original source cluster. However, if the original source container is available, Dell recommends failing back to it. Failing back to the
original source container takes less time than failing back to a new container.
296
About Data Recovery