Detailed steps required for upgrading Veritas Volume Replicator (VVR) from Storage Foundation for Windows 4.x to 5.0.x (SFW).

Article:TECH56322  |  Created: 2007-01-19  |  Updated: 2008-01-02  |  Article URL http://www.symantec.com/docs/TECH56322
Article Type
Technical Solution


Environment

Issue



Detailed steps required for upgrading Veritas Volume Replicator (VVR) from Storage Foundation for Windows 4.x to 5.0.x (SFW).

Solution



Preparing VVR for upgrade from 4.3.x versions to 5.0
Follow the procedures in the given order while upgrading from 4.3.x versions to 5.0.

Warning: A full autosynchronization is required if the procedures listed below are not performed in the given order.

To prepare the primary site
1 Take the application service group offline on the primary site.

From the Cluster Manager (Java Console), right-click the application service group and select the Offline menu option.

or

From the command line, type

hagrp -offline <group_name> -sys <system_name>

2 Take the RVG resource offline.

From the Cluster Manager console, right-click the RVG resource and select the Offline option.

or

From the command line, type

hares -offline <resource_name> -sys <system_name>

3 To get the list of RVG and RLINK records, type the following from the command line:

vxprint -lVP [-g <diskgroup_name>]

4 Verify that the data on the Replicator Log is written to the secondary site by
running the following command on the primary site.

vxrlink [-g <diskgroup_name>] status rlink_to_secondary

This command displays the replication status of the secondary represented by the specified RLINK.
This command can be run only on the primary site. Verify that the data volumes on the secondary are consistent and up-todate with the primary before proceeding to the next step.

5 Detach the RLINK to prevent VVR from replicating data to the secondary site.

From the VEA console, right-click the secondary RVG and select the Stop Replication option to stop VVR from replicating to the secondary
site. The Stop Replication option is available only when selecting the secondary RVG. When this operation is performed the connection
between the primary and secondary RVG is broken.

or

From the command line, type

vxrlink [-g <diskgroup_name>] det <rlink_name>

This command detaches the RLINK from the RVG.

6 Dissociate the Replicator Log volume from the RVG.

From the VEA console, right-click the Replicator Log volume and select the Dissociate Replicator Log option from the menu.

or

From the command line, type

vxrvg [-f] [-g <diskgroup_name>] [-r <rvg_name>] dis <volume_name>

This command dissociates the specified Replicator Log from the RVG. If the -f option is used, it forcefully dissociates the volumes.

Warning: Use the -f option only when you are sure that the secondary is completely up-to-date with the primary. Otherwise this option can cause data corruption.

7 Bring the Disk Group and IP resources offline in a VCS cluster setup.

From the Cluster Manager console, right-click the Disk Group resource and select the Offline option from the menu.

or

From the command line, type

hares -offline <resource_name> -sys <system_name>

Repeat the above step to offline the IP resource.

To prepare the secondary site

1 Take the RVG resource offline.

From the Cluster Manager console, right-click the RVG resource and select the Offline option.

or

From the command line, type

hares -offline <resource_name> -sys <system_name>

2 To get the list of RVG and RLINK records, type the following from the command line:

vxprint -lVP [-g<diskgroup_name>])

3 If you used the VEA console to stop replication on the primary site, then you need not perform this step on the secondary site. However, if you used the command line to detach the RLINK while preparing the primary site, then you must perform this action on the secondary site as well.

From the command line, type

vxrlink [-g <diskgroup_name>] det <rlink_name>]

This command detaches an RLINK from the secondary RVG.

4 Dissociate the Replicator Log volume from the RVG.
From the VEA console, right-click the Replicator Log volume and select the Dissociate Replicator Log option from the menu that appears. This option is available for selection only if the Replicator Log is associated with an RVG.

or

From the command line, type

vxrvg [-f] [-g <diskgroup_name>] [-r <rvg_name>] dis <volume_name>

This command dissociates the specified Replicator Log from the RVG. If the -f option is used, it forcefully dissociates the volumes.

Warning: Use the -f option only when you are sure that the secondary site is completely up-to-date with the primary. Otherwise this option can cause data corruption.

5 Offline the Disk Group and IP resources in a VCS cluster setup.

From the Cluster Manager console, right-click the Disk Group resource and select the Offline option from the context-menu that appears.

or

From the command line, type

hares -offline <resource_name> -sys <system_name>

Repeat the above step to offline the IP resource.


Re-enabling VVR after the upgrade from 4.3.x versions to 5.0

Follow the procedures in the order given while re-enabling from 4.3.x versions to 5.0.

Warning: A full autosynchronization will be required if the procedures listed below are not performed in the given order.

To enable the updated objects

1 Bring the Disk Group Resource online.
From the Cluster Manager (Java Console), right-click the Disk Group Resource and click Online.

or

From the command line, type

hares -online <resource_name> -sys <system_name>

2 Bring the VVR IP Resource online:

From the Cluster Manager (Java Console), right-click the VVR IP Resource and click Online.

or

From the command line, type

hares -online <resource_name> -sys <system_name>

3 Re-associate the Replicator Log volume.

From the VEA console: On the primary site, expand the RDS to which you want to associate the Replicator Log volume. Right-click the primary RVG, and click the Associate Replicator Log option. Select the correct Replicator Log volume from the dialog box and click OK.

or

From the command line, type

vxrvg [-g <diskgroup_name>] aslog <rvg_name> <volume_name>

4 To attach the RLINK, open a command prompt and run the following command:

vxrlink -f att RLINK

5 Repeat steps 1-3 on the secondary site.

6 On both the primary and the secondary site, bring the RVG service group online.

From the Cluster Manager (Java Console), right-click the RVG service group and click Online.

or

From the command line, type

hagrp -online <group_name> -sys <system_name>

7 On the primary site, bring the application service group online.

From the Cluster Manager (Java Console), right-click the application service group and click Online.

or

From the command line, type

hagrp -online <group_name> -sys <system_name>

8 If bringing the service groups online does not start the application, perform any necessary tasks to start the application. Depending on options in your
environment, these tasks may include mounting databases or manually starting the application.



Legacy ID



295156


Article URL http://www.symantec.com/docs/TECH56322


Terms of use for this information are found in Legal Notices