Want to protect your cyber security and still get fast solutions? Ask a secure question today.Go Premium

x
  • Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 203
  • Last Modified:

Exchange 2003 Enterprise SP2 , After Power outage, Strange Behaviour HELP URGENT

We had a power outage, and the UPS devices wouldn't hold the server up so we gracefully shut it down.  After the power was restored, I brought up the DC's one at a time, 3 of them to be exact. then I brought up Exchange.  The backups ran last night with no problems. A hard drive in the array that the msg store is on died@ 4:22 this morning, it has been replaced and is rebuilding at the moment. Then at 4:28 this morning event log is beginning to fill with Cache Page Check sum errors.  But now things are really getting strange. Users are experiencing the following issues....

1)Users are no longer able to move or delete items
2) The deleted Items folder an most of the clients (outlook 2003) are filling with old deleted items from the past.
3) No one can sent an email with an attachment ( Operation Failed error message)
4) the Out box folder never refreshes items that have successfully been sent.

So I know what to do as far as the repair of the data base using eseutil and isinteg -s.... But I'm wondering if, when the server came back up, something else happened that is causing this odd behaviour for all my clients in outlook.
All the services are running that should be I think,....?  Both  stores ( pub/priv) are running and mounted....

Please help me diagnose this ASAP
TIA to all experts
0
CCNPwanabe
Asked:
CCNPwanabe
  • 3
  • 2
  • 2
  • +1
1 Solution
 
CCNPwanabeAuthor Commented:
Another point is, if I do runn the regeim of eseutil /p and /d will this fix the problems or is there something else that could be causing this madness.
Why do things like this only happen when the Sr, Net admin is out of town????? :-(
0
 
Beachdude67Commented:
Man, that's a tough spot. But this is your big chance to be a hero.

OK, first things first: to really gracefully bring down an exchange server you need to stop the exchange services running. You can do that next time by running services.msc and killing the applicable services.

Now for the big problem: verifying the Exchange database, and fixing it. First, use backup on the exchange server to try and back it up. Since NT Backup won't let you back up a corrupt database, it will serve as a good gauge as to whether it really is the database or something else that is going on that is associated with the power outage. Hopefully, nothing is wrong with the Exchange database and your backup will go OK. If you have room on some drive, I would try to do this locally. If not, run it to tape (if possible). You could even create a share and run it to a workstation. Whatever the case, I would do this as a first step.

If it really is the exchange database, then there are a few options. The first is to restore from tape backup. This should probably be your last resort, but hopefully, you have a backup.

Another option is to down the server and run the exchange database repair utility (ESEUTIL). This is a good tutorial on using ESEUTIL to repair a database:

http://www.msexchange.org/tutorials/Exchange-ISINTEG-ESEUTIL.html

That should get you started. Proceed with caution - I know you're probably stressed but doing something rash will just make things worse.

0
 
CCNPwanabeAuthor Commented:
I have a back up that ran from last night, it wont be a big deal if I loose anyting,  I m fairly confident that back up will fail, it always does when there are cache checksum errors.

Do you think that im experiancing here is related in some way or in whole to a corrupt data base???
Thats really the question.  I dont want to waist any more time by running eseutill and ISInteg if its isnt gong to fix the 4 problems listed above.
0
Problems using Powershell and Active Directory?

Managing Active Directory does not always have to be complicated.  If you are spending more time trying instead of doing, then it's time to look at something else. For nearly 20 years, AD admins around the world have used one tool for day-to-day AD management: Hyena. Discover why

 
SembeeCommented:
Personally faced with that, I would only do one thing.
Go to Microsoft.com, find the number for support and pay the fee. The quicker you get Microsoft involved the better your chance of success. This site is not really suitable for this type of disaster recovery.

Simon.

--
If your question has been answered, pleased remember to accept the answer and close the question.
0
 
Beachdude67Commented:
No way I would pay the support fee - I have dealt with M$ $upport 3 times in my career and they are 0-3 in fixing problems. If you have a backup and you don't mind losing 1 day of mail I'd do it. The fact that the checksum failed when you checked the database indicates that the database is indeed corrupt.
0
 
SembeeCommented:
If you can afford to be down for a day without email. Most of my clients cannot and the fee is very small in comparison. Microsoft support also depends on who you get. If the service you are getting is poor, get it escalated. Otherwise your data is at risk.

Simon.

--
If your question has been answered, pleased remember to accept the answer and close the question.
0
 
redseatechnologiesCommented:
...and remember, for every hero that saves the world, thousands more die without a mention.  A disaster recovery situation IS NOT the time to have a crack at it yourself - If you don't want to call MS, call a professional Exchange Consultant.

I know, I've been there.
0
 
CCNPwanabeAuthor Commented:
Thanks Guys for the optomistic remarks.  
The back up ran last night and failed of course after copying the entire database(s).  And as one can assume because of such, the transaction logs didnt commit.
HOWEVER... as I stated in my first post, in the original question, one of the drives in the array had failed, and was in the process of restoring parity and data from the other drives in the array.
Ass soon  as the drive was finished restoring, I sidmounted the private message store ( on a hunch, ormaybe an educated guess) and when I remounted it, VIOLA ! No more wierd beheivior.  Iwas instantly able to delete email, move items from folder to folder. All the really wierd stuff stopped instantly.  I hope this helps others in my situation.. But it seems like for some reason ( and I thought this was part of the purpose of the transaction logs was ) when an array volume is re-building, exchange isnt able to access the data base, or do too much with the trasaction files... For what reason I have no idea.  My question I guess is this: What in the database would have caused all those problems I listed above, when every one of the issues is directly corolated to additions to the transaction logs...?????????????????????????
Maybe some one can explain what Im missing in my vague understanding of the way the data bases ( pub/priv edb&stm ) coorolate withthe trasnaction logs while exchange is in an operational state, as aposed to the way it beheives when its being backed up....  

0
 
redseatechnologiesCommented:
>>the transaction logs didnt commit.

Transaction logs don't commit with a backup - they flush.  They are already comitted, their use is when you are restoring an old backup, and then you can replay the transaction logs.

Deleting, moving, changing, all creates a transaction log, so you could well be correct there.

Whenever anything changes, new mail comes in, old mail is edited or deleted, whatever, a transaction log is written, and the change is committed to the database.

-red
0

Featured Post

Free learning courses: Active Directory Deep Dive

Get a firm grasp on your IT environment when you learn Active Directory best practices with Veeam! Watch all, or choose any amount, of this three-part webinar series to improve your skills. From the basics to virtualization and backup, we got you covered.

  • 3
  • 2
  • 2
  • +1
Tackle projects and never again get stuck behind a technical roadblock.
Join Now