How to migrate or move storage from one array to another or change the RAID level

The method that Veritas recommends with InfoScale is to follow the below steps. There are other methods to use as far as hardware mirroring, etc., but by using the below described methods, the grey areas can be eliminated. If there are problems with the migration, please open a case with Veritas so further assistance can be provided. 

The steps below can be performed while everything is online and running as no downtime is required.  

 

There are some preliminary steps that need to be taken in an InfoScale Enterprice configuration (VCS cluster) to avoid a potential fault of the MountV resource during migration. 

 

These steps will allow a change to the RAID level of the back-end if you wanted to test performance with different RAID levels.  This method also applies if you are trying to migrate off of one array to another (because it is out-dated or any other reason).  It should also be mentioned that with VVR, everything would be done the same with the addition that we need to make sure to move off the SRL log volume as well.  This can either be mirrored or a simple drag and drop of the SRL log volume in the Disk View tab to a disk that makes up the new SAN.

1) Present (Zone) storage that is identical in size (can be greater in size) as current storage to machine.

2) Add disks to diskgroup that are of same size as disks in the disk group to allow the creation of a mirror (Example: A diskgroup with 2 disks of 100GB a piece, 2 or more disks that are equal to or greater than 100GB will need to be added. 

3) Right click on the Volume(s) in the diskgroup and select Mirror > Add.  Make sure to select the appropriate disks from the new storage to add the mirror. Follow the same steps for all volumes in diskgroup.

4) Wait until the Mirror(s) completely synch up. This can be viewed by going to Tasks at the bottom left of the VEA GUI.  This can take a while depending on size of data so please let it finish.

   PRELIMINARY STEPS BEFORE STEP 5 IN A VCS CLUSTER:

 

POST STEPS AFTER STEP 5 IN A VCS CLUSTER:

 

a. Unfreeze the Service group by right-clicking on the group > Unfreeze

 

b. If the MountV is faulted, clear the fault by right-clicking on the SG and selecting Clear Fault - Auto. 

c. Bring the MountV online by right-clicking the resource and select Online.  
 

5. Verify that all access to data is the same and confirm the application is working properly. 

6. Once it is confirmed that all is well, right-click and delete the broken off mirrored volume(s) from the old array.  After deleting the mirror(s), Right click on the old disk arrays (nothing should be on them now) and select "Remove Disk from Dynamic Disk Group". This is a VERY important step, if you forget to do this you will see missing disks when they are zoned away and possibly have issues with your applications.

7. Zone away the old disks.