• Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 584
  • Last Modified:

Exchange reseed database error

We have a 2 server cluster running Exchange 2010.  Server 1 crashed today and this time, left the database copies in a "failed and suspended" state. (we have 9 databases - all are "failed and suspended").  Normally when one of the servers crashes, the databases fail over to the other server.  I have to admit however that I don't know which server was the active server when this crash took place - should have been server 2 but server 2 is the one with the "failed and suspended" copies of the databases.
I tried the reseed command (last time this happened, the reseed worked without issue) but get the error attached.
I also tried to do a database update via the EMC wizard (right click the database and choose "update database copy") - received the same error.
We have circular logging enabled.  I've tried to locate the log files so I can dismount the database(s), move the log files somewhere else and then remount the database(s) but I can't find the log files.  Same thing applies if I want to do a manually copy, I still need to copy the DB and the log files to a shared area in order to copy to my server 2.
I don't want to reboot server 2 with the databases in a failed state - unless someone out there can assure me that will resolve my issue.
Hoping to get a response soon - I'd like to fix this today if possible.
ReseedError.pdf
0
skbarnard
Asked:
skbarnard
1 Solution
 
dsnegi_25decCommented:
Can you try dismount the one database & check the database status if that shows clean shutdown then you can delete all the logs and mount the database again. If you see logs files are required then make sure logs file are clean state

http://msexchangeguru.com/2009/07/12/exchange-database-recovery-using-eseutil-commands/
0
 
James HIT DirectorCommented:
Do you have a witness server setup?
This would have kept your DB's up and running regardless of which server was primary.
0
 
MaheshArchitectCommented:
If you don't have necessary log files for soft recovery of exchange databases, then try hard repair of databases with eseutil /p <database path> and then try to mount the same, if mounted successfully, then reseed passive copy

http://blog.pluralsight.com/eseutil-exchange-2010

Mahesh
0
Industry Leaders: We Want Your Opinion!

We value your feedback.

Take our survey and automatically be enter to win anyone of the following:
Yeti Cooler, Amazon eGift Card, and Movie eGift Card!

 
Seth SimmonsSr. Systems AdministratorCommented:
Need to figure out where the database files actually are first.
According to the screenshot, it's not where it's expected to be ("The system cannot find the file specified").

You also won't find any log files (except the one being overwritten) since circular logging is enabled - which means you can only restore your database to the point of the last full backup.
0
 
skbarnardAuthor Commented:
Thanks for all the responses.  
Dsnegi_25Dec - it seems the link you provided is primarily for Exchange 2003 so I haven't tried anything from that article.
Mahesh - I tried the eseutil command that you provided and got essentially the same error - that the file can't be found (see attached)
Seth2740 - I agree with you that I need to find the files. I believe the folder is on the C: drive but not in Program Files, it's a folder called ExchangeDatabases and there are folders for each database - folders are called VOL1, VOL2 etc and in each of those folders is the associated DB (DB1.edb etc).  There is a sub folder called "logs" which has a bunch of files (text documents - all 1,024 KB in size)
We do have an additional server that I believe is supposed to be the witness server and it's where the DAG resides.  I haven't yet checked that server so I will do so - it could be that server is where the log files are kept and the "connection" between the servers has been lost.
That's why I like to ask questions here - you help me think things through.
I'll report back on that point but any other thoughts or suggestions you have will be greatly appreciated.
EseutilError.pdf
0
 
MaheshArchitectCommented:
Have you verified database path physically \ through adsiedit.msc

If You may change your database path to reflect to actual physical path

Check below article and use adsiedit.msc method
http://mistech.com/how-to-move-exchange-2010-mailbox-database-to-another-drivepartition/

Then need to check for Database state and then hopefully you can mount and then reseed passive copy

Mahesh
0
 
skbarnardAuthor Commented:
To give an update - it appears I may have a bad controller so before I do any more trying to reseed the database copies, I need to get the controller back online (the error states the controller is locked up)
Have a service call logged, currently waiting for a call back.
Will keep you posted as I discover more.
0
 
skbarnardAuthor Commented:
**** Final update - thanks to all who contributed ==== It was a reboot/controller issue

Since the node that had the "failed and suspended" databases was the passive node, I first verified with support that a failover WOULD NOT occur then I rebooted the 2nd node.
Once the server came back up, the controlller was no longer locked up and was back online.
I didn't need to do a reseed at that point, I was able to right click the database and choose the "resume database copy" option.
All databases came back online when I chose that option.
0
 
skbarnardAuthor Commented:
I just had to reboot the server to clear the controller lockup - that was the issue causing me to not be able to 'reseed' the databases.
0

Featured Post

Technology Partners: We Want Your Opinion!

We value your feedback.

Take our survey and automatically be enter to win anyone of the following:
Yeti Cooler, Amazon eGift Card, and Movie eGift Card!

Tackle projects and never again get stuck behind a technical roadblock.
Join Now