Tek-Tips is the largest IT community on the Internet today!

Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!

  • Congratulations strongm on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!

Exchange 2003 recovery

Status
Not open for further replies.

lanceja

MIS
May 23, 2002
181
US
Here is my scenario: Exchange 2003 on a hard drive. Hard drive is actually 3 - 74Gb drives setup for Raid5 so have approximately 150Gb available but only 7Gb free. Want to replace the 3 - 74Gb drives with 3 - 146Gb drives.

1. Have looked at the documents on the Recovery Storage Group information and it looks to me (and I could be wrong) that I need to add on some added options for my Exchange and then create the Storage Group. If I do this, any idea of how big this Storage Group will be and can I create this group on an external drive?

2. Tried using Ghost 2003 to make an image of the hard drive so I can just reimage the new drives. When I tried this, 15 hours after start, was only up to 5% so do not see this working.

3. Want to try Acronis but not sure if can justify price tag. Has anyone used Acronis to image an Exchange 2003 server and did this work?

Thanks for any and all information
 
Why not just replace the 74GB drives with the 146GB drives one at a time, letting the raid controller rebuild the array between each disk change, then expand the array to use the new space.

You should also be able to do this live so no downtime required.

[small]have you turned it off and on again?[/small]
 
We thought about this but was not sure it would work. My thoughts were that with Raid5, each disk had to be the same. If I put a 146Gb disk into a Raid5 that were utilizing 74Gb drives, it would not recognize the 146Gb drive or worse, bring everything down.

Have you done this before and if so, what was the procedure you did?

Thanks for your input.
 
Each drive must be the same size or larger.

We recently performed this on an array with 14 drives, upgrading them from 72.8GB to 300GB drives. Each drive took 12 hours to rebuild so it took a long time but we got there in the end.



[small]have you turned it off and on again?[/small]
 
It depends on the RAID controller...some, like Dell/HP/Compaq, allow for array expansion through a GUI or BIOS utility. Others, like 3Ware, require a script (offered free by the manufacturer) to fully expand the array to take up all the new space. Some will do it on their own.

I would question the wisdom of moving from 74GB to 146GB drives. With the storage market what it is today, why not opt for larger drives like SjrH did, 300GB or even larger? My point is that it will be a much longer time before you run low on space again (and you will!) and huge drives are super-cheap these days.


Tony

Users helping Users...
 
You still need to be aware of the IOPS load. Putting larger drives in doesn't necessarily mean performance to add more users/mail. And the performance hit by using RAID 5 is bad enough (over RAID 10).

Just adding bigger drives to an existing array and then doing some GUI Voodoo means that the storage isn't likely aligned to the correct boundaries, which means you're going to potentially increase work for the array.

so have approximately 150Gb available but only 7Gb free
You started with rougly 136.16. What's the event logs say about whitespace?

Pat Richard MVP
Plan for performance, and capacity takes care of itself. Plan for capacity, and suffer poor performance.
 
We finally used Acronis Enterprise server to clone the hard drive. We did not go with purchasing 300Gb drives due to the fact we already had the 146Gb drives and it is lot less painful going through a root canal than trying to get our company to purchase anything.

I will check on the event logs and whitespace. Prior to us changing out the drives, we did delete a couple of mailboxes which was using up to 2Gb of space according to the System manager. Even with this did not see a decrease in space. I thought the maintenance that occurred over night would fix this, but did not happen.

Thanks for all of your suggestions.
 
Even with this did not see a decrease in space. I thought the maintenance that occurred over night would fix this, but did not happen.
Only an offline defrag would do that.

Pat Richard MVP
Plan for performance, and capacity takes care of itself. Plan for capacity, and suffer poor performance.
 
Have you taken a look at the Exchange best practices? Do you have enough room in your server to utilize your 73 GB drives and your 146 GB drives. You might not be able to create a RAID 10 (as Pat suggested) but you could separte your OS (RAID 1) from your Exchange data and log files (RAID 5). You could use your last 73 GB drive as a single drive for your page file. This would certainly help with the IOPS. Just a thought.
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor

Back
Top