Video Screencast Help

Deporting/Importing Disk Group in VMware.

Created: 18 Nov 2011 • Updated: 21 Nov 2011 | 7 comments
Umair K's picture
This issue has been solved. See solution.

Hello.

I am running SFHA 5.1 in a VMware environment. I created a 2 node cluster and the nodes failover perfectly. However when i create a disk group on a shared bus and deport it from the current node. It does not show up for importing in the other node. 

I have created the IP, NIC and VmDG resources and they are all online on the node the disk group was created on. However when i try to fail them over. the VmDG faults as it is not able to import itself on the other node.

I tried both Windows server 2003 and 2008 and the problem persisted.

Please note that i ran the '' Vxclus usesystembus ON '' command to enable the shared bus.

I would really appreciate your help in this regard.

 

Thanks

Umair.

Comments 7 CommentsJump to latest comment

mikebounds's picture

Can you see the disks on the failover node in Windows Devices.  If you can't see them at this low-level in Windows then SFW will not be able to import them and this reflects a problem in VMWare.

Mike

UK Symantec Consultant in VCS, GCO, SF, VVR, VxAT on Solaris, AIX, HP-ux, Linux & Windows

If this post has answered your question then please click on "Mark as solution" link below

Umair K's picture

Hi Mike.

Thanks for a quick response. What disks are you talking about? I can see the hard disks i have added throught Vmware in Windows Devices if thats what you asked?

mikebounds's picture

That's what I meant - to check that you can see them in Windows Devices from both nodes.  If you can then you can try to do a rescan (in GUI or vxassist rescan) to to if it finds diskgroup.  If it doesn't and there is no data on the disks, then you could try to create a new diskgroup from the failover node to see what the error is as this may gives clues as to what is wrong - from the GUI "create new diskgroup" may be greyed out, so you could try from command line using "vxdg -g dg_name init harddisk_name".

Mike

UK Symantec Consultant in VCS, GCO, SF, VVR, VxAT on Solaris, AIX, HP-ux, Linux & Windows

If this post has answered your question then please click on "Mark as solution" link below

Wally_Heim's picture

Hi Umair K,

If you create the clsutered dynamic disk group on node 1 and deport it, then you should be able to rescan on node 2 (if it sees the same disks) to see the disk group.  If the disk group does not show up on node 2 then it sounds like the two nodes are not seeing the same disks.

It will not matter where you create the disk group if both nodes are seeing the same disks. However, if the diskgroup is imported and you are using SCSI-2 reservations then the other node will not see the disk group until the disk group is deported. 

If SCSI-3 reservations are used, then both nodes should see the diskgroup even when it is imported and reserved on one node.  A rescan on the passive node will be needed after the disk group is initially created.

Keep in mind that the VMware docs for setting up shared disks on a MCSC cluster will be the same steps that you use to setup the disks for a SFW-HA cluster in vmware.  Have you followed the VMWare MSCS documentatin when setting this up?

Thanks,

Wally

Umair K's picture

Hi Wally.

I havent followed any documentation while setting this up. I set it up as a standard setup which i have done with SF and SFHA in the past. However this is the first time i am sharing a disk between them.

Please link me up to any documentation that you might know of to set SCSI-3 reservations on both nodes.

Thanks

Umair.

Wally_Heim's picture

Hi Umair,

VMWare should have documentation on how to setup shared disks for MSCS or WFOC.  The steps are slightly different depending on the exact VMWare product and version that you are running.  If memory serves me correctly, it comes with the media but if not a quick google search with your VMWare product/version should find what you are needing.

Thanks,

Wally

Umair K's picture

Hi Wally. I found a guide to install Remote CLI and zeroing the disks before adding them with a new SCSI controller in VMware. Fixed my issue. Thanks a bunch.

Umair.

SOLUTION