Enabling Cluster Support in VxVM (Optional)

This release includes an optional cluster feature that enables VxVM to be used in a cluster environment. The cluster functionality in VxVM allows multiple hosts to simultaneously access and manage a set of disks under VxVM control. A cluster is a set of hosts sharing a set of disks; each host is referred to as a node in the cluster.

  To enable the cluster functionality in VxVM

1. Obtain a license for the VxVM cluster feature.
2. Install the software packages onto each system (node) to be included in the cluster as described in Installing the VERITAS Storage Foundation 4.1 Software and Installing the VEA Client.
3. Initialize VxVM using the procedures described in Configuring VERITAS Volume Manager.
4. Start VEA as described in Starting the VEA Server and Starting the VEA Client.
5. Configure shared disks as described in Configuring Shared Disks.

Configuring Shared Disks

If you are installing VxVM for the first time or adding disks to an existing cluster, you need to configure new shared disks. If you are upgrading VxVM, verify that your shared disks still exist.

The shared disks should be configured from one node only. Since the VxVM software cannot tell whether a disk is shared or not, you must specify which are the shared disks.

Make sure that the shared disks are not being accessed from another node while you are performing the configuration. If you start the cluster on the node where you perform the configuration only, you can prevent disk accesses from other nodes because the quorum control reserves the disks for the single node.

Configuring New Disks

Use this procedure if you are installing and setting up VxVM for the first time.

  To configure new shared disks

1. Start the cluster on at least one node.
2. On one node, run the vxdiskadm program and choose option 1 to initialize new disks. When asked to add these disks to a disk group, choose none to leave the disks for future use.
3. On other nodes in the cluster, run vxdctl enable to see the newly initialized disks.
4. From the master node, create disk groups on the shared disks. To determine if a node is a master or slave, run vxdctl -c mode.

   Use the vxdg program or VEA to create disk groups. In the vxdg program, use the -s option to create shared disk groups.

5. From the master node only, use vxassist or VEA to create volumes in the disk groups.

   ---

     Note    The volumes must be of type gen. Do not create RAID-5 volumes. Before creating any log subdisks, read the section on DRL in the VERITAS Volume Manager Adminstrator's Guide.

   ---

6. If the cluster is only running with one node, bring up the other cluster nodes. Enter the vxdg list command on each node to display the shared disk groups.

Verifying Existing Shared Disks

Use this procedure if you are upgrading from a previous release of VxVM or a Storage Foundation product.

  To verify that your shared disk groups still exist

1. Start the cluster on all nodes.
2. Enter the following command on all nodes:
     # vxdg list

   This displays the existing shared disk groups.

Converting Existing VxVM Disk Groups to Shared Disk Groups

Use this procedure if you are upgrading from VxVM 3.x to VxVM 4.1 (or Storage Foundation 3.x to a Storage Foundation product at the 4.1 level) and you want to convert existing disk groups to shared disk groups.

  To convert existing disk groups to shared disk groups

1. Start the cluster on at least one node.

   For a two-node cluster, start the cluster on one node; for a four-node cluster, start the cluster on three nodes.

2. Configure the disk groups using the following procedure.

   To list all disk groups, use the following command:

   
     # vxdg list

   To deport disk groups to be shared, use the following command:

   
     # vxdg deport disk_group_name

   To import disk groups to be shared, use the following command on the master node:

   
     # vxdg -s import disk_group_name

   This procedure marks the disks in the shared disk groups as shared and stamps them with the ID of the cluster, enabling other nodes to recognize the shared disks.

   If dirty region logs exist, ensure they are active. If not, replace them with larger ones.

   To display the shared flag for all the shared disk groups, use the following command:

   
     # vxdg list

   The disk groups are now ready to be shared.

3. If the cluster is only running with one node, bring up the other cluster nodes. Enter the vxdg list command on each node to display the shared disk groups. This command displays the same list of shared disk groups displayed earlier.
4. For information on upgrading in a Cluster Volume Manager (CVM) environment, see the VERITAS Storage Foundation Cluster File System Installation and Administration Guide.

Upgrading in a Clustered Environment with FastResync Set

If there are volumes in the shared disk groups with FastResync set (fastresync=on), before beginning the upgrade procedure, reattach each snapshot to its data volume, using this procedure:

  To upgrade in a clustered environment when FastResync is set

1. You should run this procedure from the master node; to find out if you are on the master node, enter the command:

   # vxdctl -c mode

2. On the master node, list which disk groups are shared by entering:

   # vxdg -s list

3. Using the diskgroup names displayed by the previous command, list the disk groups that have volumes on which FastResync is set:

   # vxprint -g diskgroup -F "%name" -e "v_fastresync"

4. Reattach each snapshot:

   # vxassist -g diskgroup -o nofmr snapback snapshot_volume

5. If you are upgrading from VxVM 3.5 Maintenance Patch 3 or from VxVM 3.2 Maintenance Patch 5, set FastResync to off for each volume:

   # vxvol -g diskgroup set fastresync=off volume