• Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 1040
  • Last Modified:

Exchange 2007 single-server disaster recovery

I am running Exchange 2007 single-server on Windows Server 2008 with Backup Exec 12.5 backups to tape nightly.  I have a corrupted raid configuration that Dell is sending a new hard drive for.  Dell says we will have to destroy the logical drive and recreate it to clear up a punctured stripe that has corrupted the array.  Thus, I will be wiping and restoring this server.

Backup Exec support sent me the following to steps.  Can anyone verify these?  Are there any gotchas that I'm missing in these two documents?

Step 1 - Reload server and restore from backup.  Steps 1-14 only, non-DC.
http://seer.entsupport.symantec.com/docs/309696.htm

Step 2 - Restore the Exchange database.
http://seer.entsupport.symantec.com/docs/288635.htm

Thank you for any feedback before I embark on this once Fedex delivers my drive.
0
kitus
Asked:
kitus
  • 11
  • 11
  • +1
1 Solution
 
endital1097Commented:
If at all possible, I would create a recovery storage group and test a restore of your exchange databases

is this server also a dc? if not, which step 1 leads me to believe, your worries would be minimal after a test restore
0
 
endital1097Commented:
worst case scenario if the system restore fails on a non DC is you would need to run Exchange setup with the /RecoverServer switch which pulls the configuration data of your exchnage server from AD
0
 
jeremycushingCommented:
If this was me I would get a trial copy of Symantec Backup Exec System Recovery 2010 and image your server.  You can image it to a USB drive.  

I would also do a full backup to tape (like you do).

It might be even quicker to restore your server using BESR then with a tape.  No need for the recovery of the Exchange, etc...

This way you are covered twice.

I have recovered many a server with BESR and have had no issues.  

0
Concerto Cloud for Software Providers & ISVs

Can Concerto Cloud Services help you focus on evolving your application offerings, while delivering the best cloud experience to your customers? From DevOps to revenue models and customer support, the answer is yes!

Learn how Concerto can help you.

 
R--RCommented:
Check this.
http://www.msexchange.org/tutorials/Disaster_Recovery.html 
Setup.exe /DisasterRecovery
0
 
kitusAuthor Commented:
endital1097 - This is not a DC.  I like the idea of /Recoverserver as a backup plan.  Thanks!

jeremy - BESR is not an option right now because my existing database appears to be corrupt after the server went down once already.  I will have to restore from tape after recreating the raid array.  I like the idea of using BESR in the future, though.  Thank you for the suggestion.
0
 
endital1097Commented:
i highly recommend trying to create the recovery storage group tonight while you wait for your new drive and test the restore

that way you know if you need to prepare for the worst
0
 
kitusAuthor Commented:
While waiting for the drive, I ran ESEUTIL /p and ISINTEG - FIX against the database and all seems repaired.  I have not brought the information store up again yet but am instead copying the database folder to a USB drive for an extra backup (90GB).  Will a flat copy like this be useable for worst case scenario?

I have hopes of bringing up the information store and getting everything up to date, then running a final backup to tape before rebuilding the array.  If time, I will test the rsg as you suggest, endital1097.  Thank you so much for the help so far.  
0
 
kitusAuthor Commented:
I meant to say storage group, not information store in the previous post.
0
 
endital1097Commented:
yes, copying files while the database is offline is safe

run eseutil /mh c:\path\priv.edb
verify the state is clean shutdown
0
 
kitusAuthor Commented:
We have clean shutdown according to eseutil /mh.  That's the first thing that's gone well.  Excellent. Thank you for the reminder to double-check.

Database is copied off to USB.  Going to now mount the storage groups.  If all is good, will let the users back into Outlook temporarily for outgoing queued email (incoming email is on hold at provider till recovery is complete), and run a final backup of the server with BE.
0
 
endital1097Commented:
excellent news
0
 
kitusAuthor Commented:
There is still one corrupt log file that I cannot copy to USB and causes Backup Exec to stop.  This even after eseutil /mh said clean shutdown.  I have sent the file to the recycle bin and now the second backup is running longer so I'm guessing it sees the database is now OK.

After running eseutil and isinteg last night, what does Exchange think of this corrupt log file?  Is it just ignoring it?  Will that file come back to haunt me later?  It was dated 4:00am so was probably just spam anyway.
0
 
endital1097Commented:
when did the database mount
0
 
kitusAuthor Commented:
Database mounted this morning at 9:00am much later than that log file, in fact a full day later.  Database is mounted now and running fine.
0
 
endital1097Commented:
that is most likely an old log file that is no longer used

when you run the eseutil /mh there is a log required value
0
 
kitusAuthor Commented:
Damn.  The server just dropped during the tape backup due to the corrupt Raid array.  Now dead in the water till new drive arrives via fedex and I can rebuild.

I now have two copies of the Exchange database -
1.  Backup to tape from two nights ago.
2.  Flat copy to USB drive from rebuild this morning.

What are your suggested Exchange steps after I rebuild Windows and restore the server from tape?  Should restore the storage group from tape first and then copy over the newer database from USB using RSG?  Or can I fly straight from the USB copy and save time?
0
 
endital1097Commented:
once you rebuild you can use the usb copy since it is in a clean shutdown state

you could also copy the logs after the database mount to the usb
0
 
kitusAuthor Commented:
Good to hear, thanks.

Any need for RSG in that case?  It seems to me that if I restore the server and then copy the exchange database from USB back to its' original location, everything should run as before.  Am I missing something here?
0
 
endital1097Commented:
no need for rsg
just put the files back in their original location
you most likely need to select the option this database can be overwritten by a restore
0
 
kitusAuthor Commented:
enital1097 - My server restore keeps going to blue-screen after reboot after tape restore so I am going with the Recoverserver option.  However, the setup /m:recoverserver command says I need Exchange Server 2007 build 8.3.83.6 or later.  Must be SP3.  Do you know where I can easily download an SP3 installation CD?
0
 
endital1097Commented:
sorry, just got out of the new datacenter and into sunlight
0
 
kitusAuthor Commented:
All up and running now.  The /recoverserver switch works amazingly well.  After fighting a full server restore from tape that was blue screening, I went with the recoverserver route.  Worked smoothly and easily.
0

Featured Post

Hire Technology Freelancers with Gigs

Work with freelancers specializing in everything from database administration to programming, who have proven themselves as experts in their field. Hire the best, collaborate easily, pay securely, and get projects done right.

  • 11
  • 11
  • +1
Tackle projects and never again get stuck behind a technical roadblock.
Join Now