Remote Hosting of Mail During a Crises

Case Study 2
Scenario:
New Orleans software development company (requested anonymity) running Exchange 2003 Enterprise on an IBM Netfinity Server. EDB size 39GB aprox. 60 users. The server was submerged for 2 weeks along with backup tapes. The tapes were damaged beyond repair. The drives were removed out of the server. Their configuration was Exchange running on the boot drive (C) and the Information Stores were on a RAID 1 (D). Not a bad set up if the IS was also backed up outside the server on something other than tapes.

We took the mirrored drives and was able to pull off the EDB and STM files after soaking the platters in a solution to remove the corrosion. The question is what to do with them.

Disasters aside we see this type of situation constantly. Veritas or Arcserve fail for whatever reason or backup corrupted edb files and the customer is out of luck. We all know that ESEUTIL is useless for most recovery situations. This year alone we have recovered over 9000 corrupted databases. These do not include failures due to viruses or exceeding the 16gb limit (about 3500 of those types of failures).

Resolution:
This company was hurting big time. Everything was in their employee's email and public folders (9GB BTW). They didn't have any offices to operae out of and most of them are still homeless. The owner was able to get a couple of suites in Oklahoma and set up a few machines so her top programmers could finish their main projects and the accounting dept. could do some billing.

We decided to approach one of our solution partners Rackspace and asked them for some pro-bone server space running Noteworthy a quick and fast email system. The complied and we had them up and running in a matter of hours. The cool thing about Noteworthy was they could have a shared folder so all the PUB1.edb files were accessible. We had to split them into 5 1.7GB pst files, but they were still able to get organized.

Thanks go out to Rackspace for stepping up to the plate on this one. Our policy has been to recover first and get paid when these people are back on their feet again.

Exchange Data Recovery

More About Hurricane Data Recovery

Restoring Backed Up Priv Files Into New Exchange Server

Case Study 1
Scenario:
New Orleans accounting firm (requested anonymity) running Exchange 2000 Enterprise on a Dell Power Edge 4400 Server. EDB size 23GB aprox. 250 users.

The failure was at the RAID 5 level. The server shutdown and 2 out of 5 hard drives (78GB Seagate SCSI) failed. The Exchange server was backed up using Veritas and a Quantum tape library. The tapes were damaged by water.

The company is being temporarily housed by Regus in Houston.

Resolution:
The data recovery division of MAS Technology was called in to recover the data off of the Power Edge. The hard drives were immediately cloned then returned with 2 replacements to the customers temporary headquarters. The IT staff was able to get Exchange back online and the user accounts active. Of course they had no mail.

DTI Data's RAID Data Recovery Division was able to repair 1 of the damaged drives and get the RAID volume online virtually. DTI immediately transferred the EDB and STM files to the client for attempted disaster recovery. At the same time copies of the files were transferred to the exchange recovery division of DTI, in case the disaster recovery was unsuccessful.

Here are the steps taken for the disaster recovery:

1) ESEUTIL /P E:\EXCHSRVR\MDBDATA\PRIV.EDB (After you finish this command please delete all the log (EDB.LOG, EDBXXX.LOG) files from E:\EXCHSRVR\MDBDATA Folder)
2) ESEUTIL /P E:\EXCHSRVR\MDBDATA\PUB.EDB
After Finishing the above command please type the below command.
3) ESEUTIL /D E:\EXCHSRVR\MDBDATA\PRIV.EDB
4) ESEUTIL /D E:\EXCHSRVR\MDBDATA\PUB.EDB
5) Go to C:\EXCHSRVR\BIN folder and run the next command.
6) ISINTEG -PRI -FIX -TEST ALLTESTS (Run this command maximum three (3) time to get Error=0, Warning=0, Fixes=0)
7) ISINTEG -PUB -FIX -TEST ALLTESTS(Run this command maximum three (3) time to get Error=0, Warning=0, Fixes=0)

Unfortunately due to the dirty shutdown the files were unable to be recovered in this manner.

DTI was able to repair the other copies and transform the mailboxes into pst files. The pst's were delivered on an external hard drive within 48 hours to be ex-merged into a new database. For details on how to use Ex-Merge for disaster recovery click here.

They now have access to all their mail and are back in business. We can only hope that they will remain safe through Hurricane Rita.