Skip to main content
Computer Networking & Repair

Dell

Go Search
Home
Photo Blog
Documents and Lists
Support
Photography
Help
Golf Courses
  

Michael Angelo Consulting Co. (609) 641-6392 | NJ Computer Networking and Repair > Dell
Bolging about Dell Poweredge Server RAID
Creating Dell Utility Partition
 If you want an actual utility partition, then you have to use the DOM Server Assistant CD on the LD first before creating any other partitions. As it is, any placeholder partition will do as far as the boot.ini is concerned. With informed editing you should be able to fix the boot.ini without needing to place a placeholder partition. I advise learning that skill, it will help you deal with clients that delete the utility partition and then call you to get their system to where it will boot again.
Creating the correct partition numbering
So your saying that the 100GB Volume I created on the new Array should be split into 2 Volumes, the 1st holding the Dell Utility Partition, with Windows on the second volume.  I can fix that by deleting the 100GB volume and creating 2 new volumes.
 
Should I have used the Dell Open Manage install CD to create the Dell Fat32 Utility Partition, or will any partitioin/volume due as a place holder?
Moving the logical drives
Michael,
   Moving the logical drives involves physically removing the drives from the server, going into the controller BIOS and deleting the logical drives, powering back down, installing only the drives for the array you want as LD0 (slot position does not matter to the controller, though for future maintence work you should put LD0 in ID 0,1,and 2, so some repair tech does not replace the wrong drive.) You then power up and tell the controller to use the configuration from the hard drives, save, exit and power down, then install the other set and repeat the procedure.
  Obviously you'll need to fix the boot.ini before you proceed, as your dual boot version will want the old install to be on LD0, and there could be an issue with the partition numbering as well, as you did not put the utility partition on the new array. Look at the multi statement in the boot.ini to figure out the edits you'll need to make. A fault tolerant boot floppy is a good tool to use when doing this kind of work, see the M/S Technet article on how to create such a floppy.
Moving Windows and Changing which array windows boots from.
Hi Michael Angelo,
 
  In respone to your question on how to make array(1) be the boot drive. There are multiple ways to move the data from one array to the other. My personal recommendation is to go ahead and make a backup of it using a tape drive, so that you can review the backup logs for files that were damaged and not backed up, replacing the files and running consistency checks on the drives and repeating the backup till everything is backed up and you are sure that everything is clean. I then do a parallel install to the new array, and save the new boot.ini somewhere else, like on a floppy, then I restore the backup over the new install, rename the boot.ini to something like boot_partition_0.old, and copy the saved boot.ini to the new install. You are now prepared for the following:
   Changing the new array to be logical drive 0, and the old drives to logical drive 1.
 
   In case you're wondering, you use the attrib command to see and change the readonly, system and hidden attributes on the boot.ini so you can make the changes, then put the attributes back on it when done.
 
  If the new drives are capable of faster access, and they are on a bus that has ANY device that is slower, the speed of the new drives will be changed to match the slowest device on the bus. This includes, controllers, and backplanes as well as drives, so I don't expect all that much of a speed improvement other than that due to a faster RPM on the new drives.
 
Regards,
Dell-GaryS
Moving Windows from Array(0) to Array(1)
Your solution sounds simple; how, or what is the procedure for Changing the new array to be logical drive 0, and the old drives to logical drive 1?    Do I just reverse the possition of the drives in the case, or is the change done in the Perc Controller setup?
 
FYI: what I did Sunday was replace drives 3,4,5, (or array(1)), with 3 new drives;
additionally, in stead of  creating a single Volume (which is all I needed), I first created a 100GB Volume that is unused, and allocated the remaining 500GB to a second Volume that I restored the client's data to.  That second 500GB volume is my new Data Volume drive F:\, and is working just fine.  The customer is happy. 
 
What I hope to do next is follow test your procedure on a test server to practice moving Windows into the unused 100GB Volume on array(1), then configure array(1) to be array(0) and boot from it.
 
On my Powedge 2400 test server there is array(0), and array(1).  I completed the paralell install of Windows to a free volume on array(1).   Windows setup modified the boot.ini on array(0) so it dispays a dual boot menu.  So I can boot into either copy of Windows on array0 or array1. 
 
So if I follow your process, I would next:
 
1. copy the current boot.ini to a floppy
2. backup my boot volume from array(0)    (That's the volume holding my primary Windows)
3. restore that backup to array(1)    (overwritting the parallel copy of windows)
4. place the saved boot.ini on the array(1) root directory.
 
Here is where I'm not clear....
 
5. Changing the new array to be logical drive 0, and the old drives to logical drive 1
 
   How too.....
 
Enter the controller (Ctrl-A) array manager and set the first drive in the new array to be 0:0 boot.  ????
 
Thank you again for all your help. 
RAID(5) Drive Upgrade on a Dell PE2500se & 2400.
RAID5 upgrade on a Dell PE2500se.
Well, the day has come and I'm into it.  Here is what I have.
The server has two Raid5Arrays (3 drives each).  The Windows 2000 server boots from C:
 
Array(0) is Drives 0-2 and holds the boot C: and E:. partitions
The array is 33GB made into two partitions C:=8GB   and  E:=25GB  
 
(The 8GB C: drive is a problem and the point to this question.)
 
Array(1) is Drives 3-5 and hold the F: drive partition.
Drive F: was 139GB and is the array I am upgrading today with 3 new 300GB drives.
 
At this point, the new F: Array drives are installed and just completed initializing.
That took 12 hours, and now I'm ready to create Volume and restore my data.
 
Everything to this poine is fine.
BUT: In thinking about how some day soon I will need to also replace the Array(0) that holds the C: and,E: partitions; it dawns on me that this new Array is so large that it could hold the content of both Arrays and still have plenty of free space.
 
So I'm wondering;  If, on the new Array(1), instead of creating a single volume, I create 3 Volumes; one for each drive C:, E:, and F:.  How might I proced to move C:, and E:, over to the new volumes. 
 
Now I don't have time to complete that part of the job today because the client needs his server tomorrow; but i'm wondering if it would be smart to created 3 Volumes on the new Array and leave the first partition empty (unused) so at a later time I can Ghost the C: partition on Array(0) to Array(1), then mark the first partition on Array(1) as the new boot partition.
Could that be a way to move all data to the newer, larger, faster Array and eliminate the original slow 16GB array(0) drives?
 
And if I could, would I want to; meaning, is there a drawback to having just 1 array, as oposed to 2 arrays.
 
I'm calling Dell support to ask the above question;..  It's sunday 9:50am and I only have about 2 hours to decide this question.  I hope you see this in time to respond.
 
Thanks for everything; your input over the past 2 weeks has been valuable and most appreciated.
 
Regards,
 
Michael Angelo
Array Configuration
The RAID configuration is stored BOTH on the hard drives as well in the NVRAM on the controller.
 
  Assuming you still had a partition you were booting to, yes Array Manager would have been complaining about the missing drives and offline array.
 
Message Edited by DELL-GaryS on 08-01-2007 10:38 AM
Replacing RAID Drives with new larger drives

While waiting for a reply: I did the following test:

Powered PE3400 off, removed 2 raid(5) drives. Then estarted entered RAID Config Utility, MESSAGE: configuration Change Detected. [accept] [reject] ; I selected [accept] and upon entering the Config utility, find the container that was holding my array is gone.

Now on this unit I didn't have any replacment drives to install then kill, so I just powered the server off, re-inserted the original 3 drives, power the server on and Ctrl-M into the config utility agan to get a MESSAGE: saying something like 'Drives 1,2,3 previously offline now restored. Presto the removed Container is back and the 3 drive Array is restored.

I Booted into Windows and the drives are working.

Question: I wonder where the RAID configuration data is kept. Is the container information stored on the 3 drives, or in NVRAM, or what?  After removing the drives, the Array Manager didn't list the container and drives as Missing; but I chose [accept] and NOT [reject] when the RAID config warned me that the configuration changed.

Hmmm, I'll try again and this time select [reject] and see if the container will be there, and if so, what it's status is.

Question: I also wonder if the Windows Array Manager would  have shown the missing drives as OFF-Line?

Welcome to the Dell Raid Blog!
To begin using your site, click Create a Post under Admin Links to the right.

This Blog is about upgrading drives in a RAID Array on a Dell Poweredge 2500sc.  I tested my procedures on a spare Dell Poweredge 2400.  As guidance, my support came from the Forums on support.dell.com.

 ‭(Hidden)‬ Admin Links