10-10-2023 03:53 AM
Hi folks,
We have 2 node VCS cluster
RDM is used for shared storage.
DEVICE TYPE DISK GROUP STATUS
san_vc2_000270 auto:cdsdisk wdg0101 wdg01 online thinrclm clone_disk
We are moving storage from IBM to HPE.
I have added new rdm disk and have added this disk to the disk group.
3pardata0_235 auto:cdsdisk wdg0102 wdg01 online thinrclm clone_disk
I am going to evacuate disk01 to disk02 by the following and will all volumes.
vxevac -g wdg01 wdg0101 wdg0102
Do I need to freeze/stop the clusters before I start evacuating?
Is there a note/documentation that I can follow?
After the volumes are moved, how do I delete the old disk? Do I need to shutdown cluster for this?
Do i need to make any changes from VCS cluster after the above is done?
If anybody can list the steps/procedure to do, will be highly appreciated.
I have looked at the note 100009768, but this doesn't answer my questions.
Regards
10-10-2023 07:32 AM
Anyone please?
Thanks
10-11-2023 05:28 PM
Hello,
Please see my answers/comments in line below start with "<<<"
...
Do I need to freeze/stop the clusters before I start evacuating?
<<< freezing the service groups with the volume resourses on which the storage will be migrated is highly recommended. Stopping the cluster is no a MUST as stopping the cluster may resautl in dg deport
Is there a note/documentation that I can follow?
<<< there are some technotes covering this topic. eg.
After the volumes are moved, how do I delete the old disk? Do I need to shutdown cluster for this?
Do i need to make any changes from VCS cluster after the above is done?
<<< you do not need to shutdown the cluster to remove the old storage.
<<< once the data is evacuated, to free up the old disks, run the command below
<<< vxdg -g dg_name rmdisk disk1 disk2 ... (refer to vxdg command manpage for the command syntax)
<<< prior to performing data migration, consider the followings
1. collect a fresh DataCollector/VRTSexplorer report
2. a full data backup
Good luck!