Restore an old Dir.edb - Will it backfill from other servers

Posted on 2003-03-20
Medium Priority
Last Modified: 2011-09-20

Exchange 5.5 SP4.  I need to restore my Directory Database from 3 months ago.  I am wondering if it will backfill itself from the other servers in the site?  The Disaster Recovery white papers seem to suggest that it will.  I would also like to know if there are any 'gotchyas' I should look out for?

Here is my specific problem.  Everytime Online IS Maintenance runs the dsamain process maxes out the CPU to 100%.  A Reboot fixes the problem until IS Maintenance is run again.  I ran eseutil integrity check on it and it reported bad pages.  So I am assuming this is my problem.  Unfortunately it is not effecting client performance so it has gone unnoticed for 3 months.  The only thing that alluded us to the problem was the backups started running into the day?  The Exchange Backups are working, but backing up this corrupt directory.  I would prefer to use another server and move all mailboxes onto it. But this would be expensive as the server is is a very remote location.

So... Can I restore a 3 month old directory database without any issues - Apart from the time it takes to replicate/backfill from the other servers?

Many thanks,

Stephen Monty.
Question by:Steppy
Welcome to Experts Exchange

Add your voice to the tech community where 5M+ people just like you are talking about what matters.

  • Help others & share knowledge
  • Earn cash & points
  • Learn & ask questions
LVL 24

Expert Comment

by:David Wilhoit
ID: 8178651
NOOOOO!!!! cannot restore a dir. more than 30 days out, you'll blow your whole org!

Try defragging the dir first, I've done it before to clear up just such an issue. And no, it isn't corrupt. Backup software that is exchange-aware will not backup a database with checksum errors. so it's still intact, may just need a little clean up. Try this first, and let me know how it goes.

Author Comment

ID: 8178733
Thanks, that's what I keep hearing, but everything I read suggests that you can restore a good backup of the Directory at anytime and it will backfill itself via replication from the other servers in the site.  A defrag might be worth a shot though.  You didn't have any problems defragging dir.edb?  I thought that it was 'against the rules' to run eseutil on the directory database - infact any utilities on the dir.edb?

Just so you know where I'm coming from, I have read in article Q184186 that you can actually take a copy of dir.edb after installing Exchange and use it at anytime to recover from a corrupt directory?

Something else I have read - This is from a Q&A section of the Technet article "Exchange Server Part 9 - Disaster Recovery":

"Q: When I am restoring a server in a site, if I do not have a backup of the directory (Dir.edb ) for the server, can I backfill the directory from a replica on another server in the site?

A: No. It is critical that you have a backup of the directory for each Exchange computer because the directory is unique for each computer. Even if you have only the original directory backup, you can restore this backup and then backfill changes from another server in the site."

I also read similar things in the Exchange Server Disaster Recovery White Pages.

Are you sure I will kill my Org if I restore an old directory.  My thought process was that I would restore it, it would have old data which would not replicate out to the other servers.  However, the other servers would replicate any data that the restored directory did not contain.  I think it is all based on UID numbers as to who has the new and old information.  But unless you use Authrest.exe then a restored database is always assumed to be older.

Thanks again.

Author Comment

ID: 8178817
Here is the errors in the output from the Integrity check I did:

ERROR: page 2487: bad leaf page links

ERROR: error -1206 at line 3466 of G:\ese97\src\ese\repair.cxx

ERROR: error -1206 at line 3092 of G:\ese97\src\ese\repair.cxx

ERROR: node [4104:17]: subtree check (page 2487) failed with err -1206

ERROR: error -1206 at line 3186 of G:\ese97\src\ese\repair.cxx

ERROR: node [2305:0]: subtree check (page 4104) failed with err -1206

ERROR: error -1206 at line 3186 of G:\ese97\src\ese\repair.cxx

ERROR: error -1206 at line 2968 of G:\ese97\src\ese\repair.cxx

ERROR: error -1206 at line 2702 of G:\ese97\src\ese\repair.cxx

ERROR: error -1206 at line 1634 of G:\ese97\src\ese\repair.cxx

               Data table will be rebuilt

               Long-Value table will be rebuilt

               Indexes will be rebuilt

               Long-Value refcounts will be rebuilt

     checking table "MSysDefrag1" (43)

          checking data

          checking index "TablesToDefrag" (44)

          rebuilding and comparing indexes


ERROR: error -1206 at line 951 of G:\ese97\src\ese\repair.cxx

Operation terminated with error -1206 (JET_errDatabaseCorrupted, non-db file or corrupted db) after 12.953 seconds.

Does Powershell have you tied up in knots?

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

LVL 24

Expert Comment

by:David Wilhoit
ID: 8178860
I thought that it was 'against the rules' to run eseutil on the directory database - infact any utilities on the dir.edb?

-not true. Just never run a /p on the database,totally a bad idea. but a defarg won't hurt it. Make a copy first, incas esomething goes bad. The output I see tells me that you have problems, defag may not succeed. But don't worry about that until, you try the defrag. if that fails, then we have issues.

you may also want to check and see if you have dirrep issues, since the directory is looking somewhat mangled.

LVL 24

Expert Comment

by:David Wilhoit
ID: 8178871
Thanks, that's what I keep hearing, but everything I read suggests that you can restore a good backup of the Directory at anytime and it will backfill itself via replication from the other servers in the site"

-Not true again. It will back fill, but anything over 30 days will screw up the org. If you have one inside 30 days, you're cool. If you don't believe me, call MS support, they'll tell you the same thing.


Author Comment

ID: 8178889
Thanks again.  I'll try the defrag option and see what happens?


Author Comment

ID: 8178956
It's not that I don't believe you, but this is directly out of Microsofts Technet - What's your view of it?

"XADM: Recovering Exchange from a Corrupted DirectoryPSS ID Number: 184186

Article Last Modified on 4/9/1999

The information in this article applies to:

Microsoft Exchange Server 4.0
Microsoft Exchange Server 5.0
Microsoft Exchange Server 5.5


This article was previously published under Q184186
When you install Microsoft Exchange Server, you should create a backup directory in the directory that contains the Dir.edb file. A default installation without optimization will have the Dir.edb file in the Exchsrvr\Dsadata directory. After replication has been completed with other servers in the site, the directory service should be stopped, and a copy of the Dir.edb file should be placed in the backup directory. Then you can restart the services.
This suggestion is made as a precautionary measure against future problems. As a standard course of action, you should have a solid backup plan with tape rotations and periodic testing of the backups. However, in a worst case scenario, this backup copy can be used to bring the server back on line if the directory has become unusable. Unlike the Priv.edb and Pub.edb, repair utilities should not be run against the Dir.edb file because it stays synchronized with the other directories in the site.

Also, unlike the Priv.edb and Pub.edb files, the Dir.edb file is machine name specific and the system cannot regenerate a new one if the old one is missing. If a directory becomes unusable and there are no viable backups, you can stop the directory service on the computer, and move all of the files out of the Dsadata directory on all drives that the Exchange Optimizer has placed them in. The original Dir.edb can then be copied from the backup directory and you can restart the directory. Through the process of backfill, the directory will bring itself up to date with the other directories in the site.

Without this original Dir.edb file or a valid backup, the only way to get a new directory is through full disaster recovery. This involves removing Microsoft Exchange Server from the computer and reinstalling it.
Keywords: kbhowto KB184186
Technology: kbExchange400 kbExchange500 kbExchange550 kbExchangeSearch kbZNotKeyword2"
LVL 20

Expert Comment

ID: 10896035
No comment has been added to this question in more than 21 days, so it is now classified as abandoned.

I will leave the following recommendation for this question in the Cleanup topic area:
   PAQ/  Refund

Any objections should be posted here in the next 4 days. After that time, the question will be closed.

EE Cleanup Volunteer

Accepted Solution

PashaMod earned 0 total points
ID: 10920265
PAQed, with points refunded (75)

Community Support Moderator

Featured Post

Efficient way to get backups off site to Azure

This user guide provides instructions on how to deploy and configure both a StoneFly Scale Out NAS Enterprise Cloud Drive virtual machine and Veeam Cloud Connect in the Microsoft Azure Cloud.

Question has a verified solution.

If you are experiencing a similar issue, please ask a related question

This article aims to explain the working of CircularLogArchiver. This tool was designed to solve the buildup of log file in cases where systems do not support circular logging or where circular logging is not enabled
If you troubleshoot Outlook for clients, you may want to know a bit more about the OST file before doing your next job. IMAP can cause a lot of drama if removed in the accounts without backing up.
The video tutorial explains the basics of the Exchange server Database Availability groups. The components of this video include: 1. Automatic Failover 2. Failover Clustering 3. Active Manager
There are cases when e.g. an IT administrator wants to have full access and view into selected mailboxes on Exchange server, directly from his own email account in Outlook or Outlook Web Access. This proves useful when for example administrator want…
Suggested Courses
Course of the Month14 days, 9 hours left to enroll

771 members asked questions and received personalized solutions in the past 7 days.

Join the community of 500,000 technology professionals and ask your questions.

Join & Ask a Question