RE: Cluster migrations

Our file clusters have one virtual name per cluster group, then we create
individual file share resources within that group. For the most part it works
fine unless whoever creates a new file share resource does not set the NTFS
permissions correctly. The user could potentially delete the entire share. If
our file clusters were not so large I would re-architect the way we do file
clusters. Mine have an average of 300 to 500 share resources on 1.75 to 3TB
of disk space. It would be easier to do it your way.
--
Michael Kemp
NBC Universal
Senior System Engineer
MCSE, ASE, CCNA


"MarkFox" wrote:

It sounds like that should work, we have not tried or tested it that way.
All our cluster resources use a virtual server name(network name) as part of
each file share resource, that way the name goes with the resource and the
client maps to that vitual server name, not the cluster name. Do you create
individual resources for each share? Unless you need more that 2TB per file
share it is better and easier to manage, to create one File Share resource
and have the cluster auto share the subfolders under it. For instance we
have a single File share resource that shares the thousands of home directory
subfolders under it. Then if you need to move it you just present the LUN
and then recreate the 4 required cluster resources, Phys. Disk, File Share,
IP Address and Network Name. Hope that helps.
--
Mark


"Michael Kemp MCSE,ASE,CCNA" wrote:

All of the suggestions great! However I don't need to move the data as it
will be presented as the existing LUN on the old cluster. I was hoping that
there would be a tool to replicate the file share resources. I think the best
path would be to evict one node, add the Win2k3 node with the new hardware,
OS and server name. Then move the presentations to the new node and evict the
old one, then bring in the second new node. I could then pause the cluster
and change the virtual name to reflect our current naming standard. This
sounds like the simplest path, what do you think?
--
Michael Kemp
NBC Universal
Senior System Engineer
MCSE, ASE, CCNA


"MarkFox" wrote:

You could also do a mixed mode upgrade where you evict one of the Win2k
cluster nodes, add a new Win2k3 cluster node, verify all the resources work
on that node and then evict the other Win2k node and repeat the process.
What we did was build a new cluster and then just take the resource offline
on the Win2k cluster node. Present the SAN LUN for that resource to the new
Win2k3 cluster node and create the required cluster resources on the Win2k3
cluster and test. If you run into any problems you can always go back to the
Wink2 cluster. One issue we discovered during our testing was that you must
run mountvol /e at the command on the Win2k3 node before you present the LUN
you are migrating. If you do not it will show up in disk management as RAW
and you cannot do anything but format it.
Once you run mountvol /e, present the LUN and rescan disks in disk
management. It should show up on the Win2k3 node with the next available
drive letter on that node. You can either keep that drive letter or change
it if needed. Once you are done run mountvol /n to turn mountvol off. Good
Luck!
--
Mark


"Michael Kemp MCSE,ASE,CCNA" wrote:

I have a number of file clusters that are on old hardware and using Windows
2000 Advanced server. Each cluster has between 125 and 300 file share
resources. There are no print or application resources. I am going to relace
the hardware and go to Windows 2003 Enterprise.
What I am looking for is an automated way to replicate the file share
resources on the new cluster. As far as the LUN presentations are concerned,
I can present them to the new cluster in about 2 minutes. I should be able to
write new signatures once I am going to cut over, but I want to stage most of
the work prior to any actual cutover.
Anyone performed this recently and have a procedure to share?

Thanks!
.

Relevant Pages

  • Re: Changing Node & Virtual IPs for Print Server Cluster
    ... It's a clustered print server. ... Cluster Node 1 Name: cluster1a.domain.com ... Move all resources to Node A. ... IP addresses for all virtual servers including the cluster itself. ...
    (microsoft.public.windows.server.clustering)
  • Re: Exch 2003 SP2 - applied on one node, but cant move resources
    ... resources to Node2, the failover did not complete because 'system attendant' ... Virtual Exchange server and failover occurred normally again upon taking ... cluster resources oline. ...
    (microsoft.public.exchange.admin)
  • Re: Failed cluster node confusion!
    ... Blue exclamation marks usually means that the cluster service has terminated ... If this fails then the heartbeat will go over the teamed NIC ... the second node did NOT failover the resources. ... working node when one node has completely died (blue screen, ...
    (microsoft.public.windows.server.clustering)
  • RE: Cluster migrations
    ... client maps to that vitual server name, not the cluster name. ... individual resources for each share? ... Then if you need to move it you just present the LUN ... run mountvol /e at the command on the Win2k3 node before you present the LUN ...
    (microsoft.public.windows.server.clustering)
  • Re: SAN Drive Presentation
    ... Since the LUN order on Node2 is what it should be, ... Unpresent the drives to Node1 ... Rejoin Node1 to the cluster. ... physical disks from the cluster, remove the disks from device manager, ...
    (microsoft.public.windows.server.clustering)