Change disks in SQL Cluster.


Here are the general steps recommended for the migration of SQL Storage volumes, for instance:Ensure that there are Full backups of all databases in the system, when just prior to migration.

Stop all SQL Server services (Cluster Manager – Take Offline Services SQL only).

Assigning the new storage volumes and Drive Letter’s temporary copies.

Full copy of all existing data, keeping the ACLs of existing volumes to new ones.

Removal of dependencies on the SQL Service of old records.

Removing the Drive Letter’s of old records and assignment of these new discs to Drive Letters (attention has to be guaranteed that all drive Letters are kept for each volume replaced)

We recommend that prevented access by the operating system on both nodes to the old disk (disable the making of the Path, or another method which ensures that the OS does not access the old disk), but keeping the data intact in case you need to RollBack.

Allocation of new volumes in the SQL Cluster Group.

Making even the services of SQL “Offline” Move SQL group for each of the physical nodes in the cluster, must raise all funds, with the exception of SQL Services. With this step we ensure that the new volumes are correctly recognized by the OS of each physical node.

Assign dependencies removed in section 5 to the SQL Service, should be exactly the same as previously existing.

Making again the SQL Failover Group, even with the SQL services “Offline”, among all physical nodes in the cluster.

Making the Bring Online Services and SQL Move Group from all physical nodes in the cluster to ensure that the SQL service is started successfully on all physical nodes in the cluster.

You can follow the following article concerning optimization of the discs to SQL Server 2008:

Advertisements

Disk Management Extend Fails “The parameter is incorrect.”


I have a couple of cluster nodes that I failed to size the C: drive
adequately and began to run out of disk space. No problem! They’re all virtual
machines, so I just need to expand the disks and away I go! Well, I’ve run into
this situation several times since my lab has a lot of machines that I’ve
upgraded over the years and while 15 – 25 GB volumes were fine prior to Windows
2008, it just doesn’t get it now and I like to use 40 GB volumes. That said,
I’ve seen several cases where the virtual disks grow correctly and show up in
the OS correctly, but when you try and use the Windows 2008 disk extend feature
then you get this error from Logical Disk Manager, “The parameter is
incorrect.”

What’s worse is that the Logical Disk Manager sees the volume as 40 GB, while
the file system still only recognizes it as the original size. See the diagram
below where I’ve laid a Windows Explorer view of the C: file system over the
(supposed) 40 GB volume after extending the volume. Weird huh? What’s worse is
that reboots don’t correct the issue. LDM is firm in it’s belief that this
volume is now 40 GB and the File System doesn’t seem to care.

Here’s the work around that I’ve found to correct this situation. Go back
into Logical Disk Manager (Disk Management in Server Manager) and Shrink the
offending volume by 1 MB. After a minute or so, this will revert the volume back
to it’s original size.

Now you can choose to extend it again and all will be right with the world.
Notice in the screen shot below that both the volume and the file system now
agree on the correct sizing.

Understand – Windows Server 2008 Failover Clusters: Networking


  1. http://blogs.technet.com/b/askcore/archive/2010/02/25/windows-server-2008-failover-clusters-networking-part-1.aspx
  2. http://blogs.technet.com/b/askcore/archive/2010/02/25/windows-server-2008-failover-clusters-networking-part-2.aspx
  3. http://blogs.technet.com/b/askcore/archive/2010/02/25/windows-server-2008-failover-clusters-networking-part-3.aspx

Cluster Disk With Identifier (identifier) has a Persistent Reservation on it – Windows 2008


Cluster disk with identifier (identifier) has a persistent
reservation on it ,the disk might be part of other cluster. removing the disk
from other validation set.

This error due to the fact that the LUNs still keeping the old identifiers
from the old cluster, You have to use Cluster command line to clear the
reservation by that command:

cluster.exe node %nodename% /clear:disknumber

Original post: http://fawzi.wordpress.com/2009/11/27/cluster-disk-with-identifier-identifier-has-a-persistent-reservation-on-it/

Hello world!


Hello,
This is the last way to get to all of you.

We will put here problems we will encounter in the day to day and their resolutions involving the world of IT.

I´ll hope everyone can help in this new blog.

Best regards.