11-09-2012 11:49 AM
Are these the steps:
1)Ask storage team to present LUN's
2)LUN's get presented
3)NOTE LUN's from top of format output
example: c7t50060161106023B8c43
4)run devfsadm on all THREE nodes
5)LABEL Luns using format utility(NO NEED TO PARTITION LUN's)
example: format c7t50060161106023B8c43
6)Run vxdctl enable on all THREE nodes
7)Use vxdiskadm to initialize the LUN
disk to initialize: c7t50060161106023B8c43
8)Use vxdiskadm to add LUN to diskgroup
disk to add: c7t50060161106023B8c43
9)use vxresize to increase size of VxFS (vxresize -F vxfs -g <vol group name> <volume name> +Xg)
Questions: Will the disk name change after I do a vxdctl enable? Do I need to get VxVm disk names?
Solved! Go to Solution.
11-12-2012 09:50 AM
c7t50060161106023B8c43 is cXtXdXsX format (you're just missing the dXsX parts)
The resize needs to be done on the node which has the filesystem mounted.
Basically in your procedure:
Steps 1, 2, 3, 4, 6 need to be done on all 3 nodes.
Steps 5, 7 only need to be done once, can be done on any node.
Step 8, 9 need to be done on the system where the diskgroup and filesystem are online.
So if SG1 is running on node1, you run steps 8, 9 on node1. If SG2 is running on node2, you need to run steps 8, 9 on node2, they can't be done from node1.
Hope that helps.
Also: moved this to the Storage Foundation forum rather than the Storage and Clustering Documentation forum.
11-09-2012 12:02 PM
Your steps are fine and can be performed on the active node.
About VxVM disk name: it all depends on your current disk naming scheme.
What does 'vxdisk list' show?
11-09-2012 12:28 PM
Well currently vxdisk list shows disk names in CxTxDxSx format, not in c7t50060161106023B8c43 format
second question: one SG is online on one node and another SG is online on another node. I have to increase size of filesystems belonging to both SG's, do I need to go to each node and do this or can I do it from one node?
11-12-2012 09:50 AM
c7t50060161106023B8c43 is cXtXdXsX format (you're just missing the dXsX parts)
The resize needs to be done on the node which has the filesystem mounted.
Basically in your procedure:
Steps 1, 2, 3, 4, 6 need to be done on all 3 nodes.
Steps 5, 7 only need to be done once, can be done on any node.
Step 8, 9 need to be done on the system where the diskgroup and filesystem are online.
So if SG1 is running on node1, you run steps 8, 9 on node1. If SG2 is running on node2, you need to run steps 8, 9 on node2, they can't be done from node1.
Hope that helps.
Also: moved this to the Storage Foundation forum rather than the Storage and Clustering Documentation forum.