This post is more than 5 years old

2 Posts

2250

September 28th, 2010 07:00

Expanding a drive best practices

I would like to increase three hard LUNS but don't know how it will affect the cluster administrator. Here is my problem:

I have an exchange 2003 cluster that is using three SAN Drives. I want to increase their size.  All LUNS are stripped. Is there a benefit to use migrate or expand. How can I do it properly and how to have windows cluster recognize the new sized drives?

Thanks for your help.

Jesus Hernandez

isjh0@tamuk.edu

1 Rookie

 • 

66 Posts

September 28th, 2010 08:00

I tried doing this, if you have a striped LUN from the Array, the best solution is that, you map a bigger LUN on to the server and then trying copying or migrate the data from old smaller LUN to newer LUN, in simple words HOST COPY.  The reason why, when you increase a striped LUN there is a chace of corrupting the data on it ( I may be wrong but it happened to me, data on my LUN simply went blank and i had to restore it using my backup). The best way to increase a striped LUN is if you have mirror view configured , then your work is done pretty simple. Even if something goes bad, you can sync it up from you Mirrror view copy

Regarding Concatenated LUN, I havent done any expansion, but if I am not wrong its pretty straight forward. You can increase it whenever you want.

Thank You

Rama

6 Operator

 • 

5.7K Posts

September 28th, 2010 07:00

Is this a local cluster or do you use Mirrorview Cluster Enabler ? (clusters using 2 Clariions, one on each site)

If you need to enlarge LUN's on 2 sites, you need to bring the cluster down, stop the mirrorview sync on those LUN's, enlarge the LUN's on both sides, re-establish the sync again and bring up the cluster.

131 Posts

September 28th, 2010 07:00

When you say they are striped, do you mean on the host level or the array (i.e. software RAID in Windows?)

If you mean you're just striping on the array and not on Windows then read below...  If you're using Windows for RAID - then stop!  You have a CLARiiON for this.

If you don't need the space immediately, then migrate is a good option.  Just migrate it to a new larger LUN.

Expand will create a MetaLUN.

     Striped - Usually you'd do this if you need to span a large number of spindles for I/O or a lot of capacity.

     Concatenated - Usually you'd do this if you need to extend a LUN in a time crunch, then you'd migrate it off to either a larger LUN or a striped MetaLUN

Extend the LUN using migrate the LUN - Since it's exchange, use a lower priority (medium is probably fine for Flare 29+, high is probably fine for 28 and under)

Then go into diskpart on your active node.  You don't need to bring anything down to extend a LUN, but do it during a maintenance window just to be safe.

     Select disk #

     Select Partition #

     Extend

http://support.microsoft.com/kb/304736

2 Posts

September 28th, 2010 11:00

Thanks for the info. Has anyone Migrated a LUN from a smaller to a bigger LUN that is part of a cluster?  I don't want my production to break. As soon as I start the process, there is no turning back. That is why I am asking all the experts what you guys have done and how it can benefit me. I don't want to migrate and then my exchange cluster don't work. Or I don't want to find out that I have to turn off my server and use gparted or some disk partition tool to expand the drive.

Can you guys help me understand or point me where I can get windows cluster articles?

Thanks for your help.

Jesus Hernandez

131 Posts

September 28th, 2010 12:00

My previous post had a link on how to extend a partition in use by MSCS in Windows.

What you're asking is an easy and common process for any storage admin.

Windows handles LUN expansion very well.

Taken from the link I posted.

How to Extend an Existing Drive into Free Space if the Hardware Supports Capacity Extension

We recommend the following procedure for performing an offline extension of the partition during a scheduled maintenance window. You can also extend volumes that are online. (In that case, steps 2 and 3 are optional.) Before you perform an online extension of a production partition, we recommend that you test the specific hardware environment and hardware configuration to confirm that it will behave correctly.

  1. Back up the shared disk (or disks) that that you want to extend.
  2. Power off all but one node in the cluster.
  3. Take the entire group that the physical disk resource is located in offline. Bring only the physical disk resource that is to be extended online. This process should close any open handles to the disk.

    NOTE: If you have any disk or Host Bus Adapter (HBA) utilities that access the disk, you may need to quit them or stop the services so that they will release any handles to the disk.
  4. Add the additional physical drives and extend the additional disk or disks as free space by using the instructions that are included with the hardware vendor documentation.
  5. Click All Programs, point to Administrative Tools, click Computer Management, and then start the Disk Management console. Verify that the new free space is added to the end of the proper drive. Right-click the existing partition, and then click Properties. On the General tab, type a unique name for the partition. This name will be used to identify the partition you want to extend in Diskpart.exe. Quit Computer Management.

    NOTE If you encounter any problems with the preceding two steps while you are extending the drive, contact your hardware vendor for assistance.
  6. At a command prompt, type DISKPART, and then press ENTER to start Diskpart.exe.
  7. Type LIST VOLUME, and then press ENTER to display the existing volumes on the computer.
  8. Type SELECT VOLUME volume number, and then press ENTER, where volume number is the number of the volume that you want to extend. Note that the volume will have the unique name that you created in step 5, and will have been listed in the output of the command you ran in step 7.
  9. Type EXTEND, and then press ENTER to extend the partition into all of the available disk space to the end of the drive.
  10. Type EXIT, and then press ENTER.
  11. Now that the volume has been extended, you can bring the entire group that contains the physical disk resource online, and then power up all of the other nodes in the cluster.
  12. Verify that the group can come online and failover to all other nodes in the cluster

190 Posts

September 28th, 2010 12:00

You are talking shared clustering and exchange...assuming you are still 2003...this may not apply to 2007/2010

For drives other than the first SG that was created during the install (which is where some unmoveable system mailboxes live - thanks microsoft), I generally add new disks (of course this assumes you have enough space to do this) and migrate the mailboxes to the new storage.  The mailbox moves can be done online.  Of course if you have 10,000 mailboxes this might not be an option....  While it can take a while, it removes the whitespace/defrags the data (and your mailbox databases tend to shrink a good bit) and you only have two blips on the cluster - one when you add the disk, another when you remove the old disk.  MS has plenty of articles on adding/removing disks from clusters.  This will also give you a chance to align the partition correctly if it wasn't done during your original install.

Acronis makes a server tool to resize disks (among other things) - never tried this on a cluster disk though.

Dan

1 Rookie

 • 

66 Posts

September 28th, 2010 12:00

My suggestion would be using PPME Host Copy, if you dont want to take the entire system down. Its pretty easy and good tool to use and sync's up data quite well. I used it in my environment and I hardly took 5 minutes of downtime on my production server when it was done. And every thing was same, like original but this time with a big drive space.

Thank You

Rama

0 events found

No Events found!

Top