Dell PE2650 - RAID 0 - 5 x 33.9Gb 1

[KLewisBPM]

I have a DELL PE2650 with PERC 3/Di inbuilt. There are 5 x 33.9Gb drives in a RAID 0 configuration with total capacity 169.5Gb as drive C:

I cannot believe it is set up like this! (We inherited it)

Only 26gb is being used, it is running Exchange Server and is a Domain Member, DNS Server, Backup Server.

Is there anyway that this RAID 0 conversion can be converted into RAID 5 and retain the data?

Or alternatively can anyone think of a way I could do this, i.e. plug an external HDD, ghost the drive etc etc

I am very twitchy at the moment, Backup exec has been backing up the exchange environment and on 3 occassions the system has crashed with one of the HDD's blinking orange! as soon as you pull the power and plug back in it boots up fine.

Any suggestions on how to solve this would be great!!




KLewisBPM

 

[LawnBoy]

the stress relief alone, was worth the price

I agree wholeheartedly. If you're spread as thin as I am it's a lifesaver.


--
The stagehand's axiom: "Never lift what you can drag, never drag what you can roll, never roll what you can leave.

 

[KLewisBPM]

Bad News for me!

I ran the Acronis True Image and within 2 minutes the software reported a read error and then multiple delayed write failures.

I have got the system back up and running after a couple of reboots, so at least we can keep running for now.

But it appears this drive error is gonna keep haunting me until I do a complete rebuild.

I have a spare poweredge 2550 dual 1400 zeon so I can prepare this offline, my only concern is the exchange first storage group that causes the drive to fall over.


I'm gonna do a full system backup now excluding the first storage group... Can anyone see any problems here?

Any alternative suggestions would be greatly received!!

cheers

KLewisBPM

 

[technome]

I obviously have to change the boot sequence so it sees the new partition first but just wondered if Acronis will indeed make it bootable."
Acronis makes the new volume bootable as a default, the new array will automatically boot first(which I do not like).

If your getting delayed write errors, could be the disks or the firmware revision on the disks or raid adapter. Is the raid adapter at the latest revision ?

When was the last time a consistency check run on the volume? Sounds like you exchange install/data are residing in a mine field of disk surface error(s). The issue of running a CC is that there are possibly too many errors for the adapter to correct, at one time, which might fail the array. If it gets close to the point of a rebuild, I would try a CC as a last resort, it might succeed in marking the bad areas and allow you to use Acronis.




........................................
Chernobyl disaster..a must see pictorial

http://www.kiddofspeed.com/default.htm

 

[blantonlewis]

You might consider how you want it to look in the end. My suggestion (from just skimming) would be to put your exchange store(s) on your external box or just another array in the same box.

If I were you, I would FIRST do a consistency check of the volume (through openmanage) and make SURE it's backed up.

Then, and only then, I would move your exchange store to another array to improve performance.

You can use the reconfigure command in openmanage to go from RAID 0 to RAID 5, according to this:

http://support.dell.com/support/edocs/software/svradmin/1.9/en/stormgmt/vdmgmt.html#280170

Hope that helps

 

[KLewisBPM]

hi blantonlewis

thanks for the response

one of the main problems I have is that I cannot do a Full backup. as soon as i try to back up the exchange first storage group the drive falls over. Strange how the server runs with no problems whatsoever until then.



KLewisBPM

 

[technome]

Blantonlewis...
The main issue is the raid 0 has disk errors, a consistency check has a good chance of failing the array. Since data resides in the areas of disk errors, the raid adapter should have marked the error areas as bad and moved the data elsewhere, obviously it is not doing this. Since the error correction is not taking place, it is assumable there are many errors to be corrected, very likely more than the adapter can handle, that is why, at this point, I consider this a last resort.

A direct conversion to raid 5, with the raid 0 in it's present state, with disk errors, will almost surely fail, leaving Klewis with zip

........................................
Chernobyl disaster..a must see pictorial

http://www.kiddofspeed.com/default.htm

 

[KLewisBPM]

Its a tough one... I just hope that people reading this will point blank refuse RAID 0 forever lol

At the moment I am left with a couple of options, the main one being to purchase a new server install all the relevant applications including exchange, then move the mailboxes to the new server.

I have received another couple of suggestions,

1. Close Down the Exchange services and try imaging the partition again.

2. Do a file level backup of exchange


If i run a consistency check on the drive there is a chance that the array could completely fail and like technome says i could be up the creek without a paddle

KLewisBPM

 

[technome]

If you back it up as much as you can, considering the next step is a rebuild, I would run a consistency check at this point, the raid 0 may survive, then you will be able to produce an image. Considering there are disks which have errors in the original, even if the errors are corrected, a conversion to raid 5 on the original array would not be wise, as most times, disks with physical errors will likely produce more in the near future.
Raid 0, the array choice of masochists/sadists.


........................................
Chernobyl disaster..a must see pictorial

http://www.kiddofspeed.com/default.htm

 

[blantonlewis]

OK, I didn't understand some of the details above, now I do (I think) .

Of course, the most important thing is your exchange mailbox data. You can probably get by with not keeping anything else. But you HAVE to stop putting mail on this thing, or you're as culpable as the fool that set up RAID 0!

You need to stop the exchange information store service, copy the data & the log files to another drive, and use those files to start a new exchange server. I would keep the old exchange server as-is.

To copy the files, open the exchange system manager & look under servers -> storage groups. Right-click on the mailbox stores & choose properties. Then choose the database tab and it will show you where the files are located. There will be a .edb file and a .stm file (probably named priv.edb & priv.stm). In any case, those are your crown jewels. The log files, those 5MB chunks that go away once you've run a backup... you can find the location of those by right-clicking on the storage group and noting where it says the "transaction logs" are located. Copy them also, and pray that "circular logging" was not checked.

Once you have a new exchange install with your data copied over, you'll have to point the information store at the new data. good luck

 

[jpm121]

Wonder what happened?

 

[josephb31]

This is great information, but did it fix the server problem?

 

[KLewisBPM]

Hi Guys,

I am still here and the problem is still here, for some strange reason I wasn't notified of any new posts one here!

Since my last post I have done the following! On a Dell 2550 2003 Server I have installed VMWARE, I have created a Virtual Windows 2000 environment and run the exchange 2003 setup ready to try and restore into as a test run. However I can't seem to get it working as a secondary server, i wanted to try and move some old mailboxes on to the virtual server directly. The system manager works but is connected to the original server.

I Dismounted the Mailbox Folders DB and Public Folders DB and copied the entire contents of the Exchsvr\MDBDATA folder including all the 5mb log files that have been accumulating since the backups were stopped!

So the Server did go down again whilst copying! with the same drive failing. I rebooted the server and checked what had been copied to the new ARRAY i previously created.

I have attached a picture that shows the original and new directory's.

The priv1.edb copied fine! Now there a few files missing.. I think the next thing for me to do is to try an copy the missing files one at a time to see if I can actually isolate a single file!

Sorry for the delay in getting back to you guys and thanks for your input!!



KLewisBPM

 

[KLewisBPM]

Pic is here

KLewisBPM

 

[KLewisBPM]

Ok Guys..

I have now found the one single file that causes the Drive to Fall over.

It is.. priv1.stm and it is 4GB in size!


KLewisBPM

 

[technome]

Great you were able to save most of the data

"I have now found the one single file that causes the Drive to Fall over."

An error in a file will not cause a raid disk to drop out, only a disk surface error, firmware bug, or bad PCB. Likely the disk which is dropping out has a bad surface defect causing a hole in your file. To complicate the situation, when a particular raid disk drops out, many times the disk with a problem is NOT the disk which drops out. In this situation, in the raid bios or raid management software, you need to look for drive errors on each individual disk..no guarantee on this, disk with no errors can still be bad.


........................................
Chernobyl disaster..a must see pictorial

http://www.kiddofspeed.com/default.htm

 

[KLewisBPM]

Hi technome..

I was not looking for an error in a file!! I was looking for the file that is laying on the bad surface of the drive. Now I am trying to discover if this file can be recovered without the drive falling over and whether it is safe to run the disk tools. If so then I plan to try and relocate the exchange first storage group to a new redundant ARRAY. Back it up and install it into a virtual exchange server outside of the network to ensure that it all works. Then I can go on to plan a server repair.

I am so wary of the fact that i cannot backup the first storage group at the moment that if i mess this up I could be up the creek without a paddle!


thanks for the response.

KLewisBPM

 

[technome]

Tough situation... like walking on thin cracking ice...you have to move, but which way.



........................................
Chernobyl disaster..a must see pictorial

http://www.kiddofspeed.com/default.htm