Brocade Communications Systems StorageWorks 4400 - Enterprise Virtual Array Manual de usuario descargar pdf (Pagina 21)

ISL link connectivity resumed

When an ISL link is restored, data replication automatically resumes if the suspend on links down

parameter is disabled. If the write history log becomes full, the data replication group is marked to do

a full copy from Site A to Site B, which occurs when connectivity is restored. If the link is restored

before the data replication group is marked for full copy and resynchronizes, a delta (merge)

resynchronization occurs rather than a full copy.

If the suspend on links down option is enabled, it is necessary to manually resume replication by

setting the Replication I/O DR property to resumed, as shown in Figure 10.

Site A access to storage resumed

When connectivity to the storage on Site A is restored, the EVA in Site A will show the DR group

mode in the destination state after a site failover to Site B has occurred. However, depending on the

nature of the storage failure, in some cases the data replication group on both sites may be in source

state in some cases.

When this happens, the steps to perform the failback operation depend on the settings of the DR

group within HP Command View EVA.

If Suspend on links down is set to Enabled, the data replication group needs to be manually

resynchronized by resuming replication I/O before it can fail back. This is achieved by setting the

Replication I/O parameter to Resumed, as shown in Figure 10.

Once replication resumes, the data replication group changes from source/source mode to

source/destination mode, with source in Site B replicating to the destination volumes in Site A. A full-

copy or merge resynchronization takes place, as previously described.

For maximum SQL Server instance availability, wait until resynchronization completes before

attempting to move the SQL cluster group back to Site A, using Windows Failover Manager.

Otherwise, SQL Server and SharePoint will remain offline until resynchronization completes.

Site A failback

Failback to Site A is performed identically to the steps described in Site A access to storage resumed,

with the exception that the cluster nodes in Site A must also be brought online before attempting a

failback operation.

Additional steps must be taken to complete returning SharePoint to its original state and location. As

previously stated, the following functions must be performed:

 WFE redirection

 Restore the index server role to the Site A index server

 Recrawl the content databases

During testing, these steps are performed in 56 minutes, using online backups.

Failback during resynchronization

The recovery process for failback during resynchronization is identical to the Site A cluster node

recovery. SQL Server can be manually moved back to Site A by using Microsoft Failover Cluster

Manager. If the Failover/Failback property of the HP CLX EVA resource is set to RESYNCWAIT, HP

EVA waits until the resynchronization completes before successfully coming online on Site A.

Otherwise, it is best to wait until resynchronization is complete before attempting a fail back to Site A.

1 2 ... 16 17 18 19 20 21 22 23

Sin comentarios

Brocade Communications Systems StorageWorks 4400 - Enterprise Virtual Array Manual de usuario Pagina 21