Answered by:
Exchange backup with Symantec issue

Question
-
I had issue related to backup. When restarting the admin password one of my backup hanged and I need to stop it.
Later any backup on exchange was having error of writer status, which I blogged and find it was because of cancel pending status remains as it is and sometime it may take hours to clear the state.
So I found 2 otherways, one is to dismount/mount database or restart the Information store service.
I did the second one and backup started without any issue and first it goes to consistency check and it took almost a day to finish the consistency check and started the backup process. But in the middle of the back it again goes to consistency check and for 2 days it wa running as it stuck in that state.
Is there any way to sort this issue. I am sure its related to the hanging and cancel of exchange backup. Rest all my other backup with Symantec Backupexec 2010 is working fine.
Only problem is with the Exchange server backup. Is it any way to solve this consistency check issue.
MYy EDB database is 246 GB and exchnage is a cluster of 3 server (Hub/Client Acces, Mailbox1, Mailbox2)
The current log files in the database folder is 115004 files with a total size of 112 GB.Please let me know is this cause the consistency check time. If this is the problem how I can clear this issue.
3 servers in exchange cluster. One server with HUB and Client Access, second with Mailbox1 and 3rd with Mailbox2.
Waiting for your comments at the earliest
Saturday, December 10, 2011 4:44 AM
Answers
-
Ehlo, Yes, The large amount of log files cause that delay in checking, Instead of deleting committed files manually,
I suggest enable circular logging and let the Exchange do the correct job for you, below are the suggested steps:
1- Dismount the database(s).
2- Confirm Database header using Eseutil /mh command For Example, eseutil /mh "D:\DB\Database.edb" The log required value should be 0 and State value should be : Clean Shutdown
3- Mount the database
4- Enable Circular logging.
5- Dismount/Remount the Database, or restart the information store.
6- Wait about 15 Minutes till the logs are cleared.
7- Disable Circular logging.
8- Dismount/Remount the Database, or restart the information store.
9- Confirm log files are being created.
10- Run a Full Backup job. Hope that helps.
Ibrahim Al Masry http://www.ibra.me/- Edited by Ibrahim Soliman Monday, December 12, 2011 12:58 PM
- Marked as answer by Rowen-Xu Friday, December 16, 2011 8:03 AM
Monday, December 12, 2011 12:55 PM -
Ibrahim
How can I enable circular logging.
Gulab | MCITP: Exchange 2010-2007 | Lync Server 2010 | Windows Server 2008 | Skype: Exchange.Ranger | Blog: www.ExchangeRanger.Blogspot.com- Proposed as answer by Jonas Andersson [MCITP] Wednesday, December 14, 2011 1:39 PM
- Marked as answer by Rowen-Xu Friday, December 16, 2011 8:03 AM
Tuesday, December 13, 2011 6:43 AM -
Dear ALL
Thanks for all your support. Instead of doing circular logging I was searching for the real cause. I moved all my committed logs to another area but I couldnt able to do it on 3 log files which were locked by some other process. It is from this timw onwards logs are not getting cleared from the server.
I done all testing with backup exec service and all other service, no way to remove these 3 logs. Finally with process explorer it was found that System process with ID 4 is the one locking these logs. I still doesnt why these logs locked by system procecss.
So before restarting the server I moved the cluster and all services to my passive node. Then after successful moving of cluster to the passive mailbox server, I tried to clear these 3 logs. Hurraaay.. It was successfully gone from there.
I started my full backup again with backup exec .. it went successfully.. and all logs cleared....
I really appreciate all comments from you people which got soem idea for me to play with.
- Marked as answer by Sabir_rm Monday, December 19, 2011 6:48 AM
Monday, December 19, 2011 6:48 AM
All replies
-
My exchange is so far working fine with around 140 GB of disk space left for exchange and logs.
Now the problem is my Symantec backup is going to consistency check as the trasaction is around 115000 files with 122 GB size. I bilieve this might be causing such a delay for consistency check.
I need to know how I can know the commited logs and what all things I need to do to move the committed logs from mailbox folder to other area which will help to complete the backup.
thnx
Saturday, December 10, 2011 6:45 AM -
Here is the command to check which log file is committed into the database.
eseutil /mk "C:\Program Files\Microsoft\Exchange Server\E00.chk"
Gulab | MCITP: Exchange 2010-2007 | Lync Server 2010 | Windows Server 2008 | Skype: Exchange.Ranger | Blog: www.ExchangeRanger.Blogspot.comSaturday, December 10, 2011 9:12 AM -
Hi Sabir,
If convenient, please try Windows Server backup instead.
Check event log whether there's any error or warning related.
The issue could occur due to large number of logs of the databases.
If convenient, you could try to dismount database and delete those logs. (This may cause some rist you can try it after you try everything else)
Thanks.
Rowen
TechNet Community Support
Monday, December 12, 2011 3:17 AM -
Ehlo, Yes, The large amount of log files cause that delay in checking, Instead of deleting committed files manually,
I suggest enable circular logging and let the Exchange do the correct job for you, below are the suggested steps:
1- Dismount the database(s).
2- Confirm Database header using Eseutil /mh command For Example, eseutil /mh "D:\DB\Database.edb" The log required value should be 0 and State value should be : Clean Shutdown
3- Mount the database
4- Enable Circular logging.
5- Dismount/Remount the Database, or restart the information store.
6- Wait about 15 Minutes till the logs are cleared.
7- Disable Circular logging.
8- Dismount/Remount the Database, or restart the information store.
9- Confirm log files are being created.
10- Run a Full Backup job. Hope that helps.
Ibrahim Al Masry http://www.ibra.me/- Edited by Ibrahim Soliman Monday, December 12, 2011 12:58 PM
- Marked as answer by Rowen-Xu Friday, December 16, 2011 8:03 AM
Monday, December 12, 2011 12:55 PM -
And if it doesn't help, raise a support ticket at Symantec, their support are helpful and good
Jonas Andersson | Microsoft Community Contributor Award 2011 | MCITP: EMA 2007/2010 | Blog: http://www.testlabs.se/blog | Follow me on twitter: jonand82Monday, December 12, 2011 9:58 PM -
Ibrahim
How can I enable circular logging.
Tuesday, December 13, 2011 5:58 AM -
Gulab
The command output is as follows:
Checkpoint file: D:\Program Files\Microsoft\Exchange Server\Mailbox\First
Storage Group\E00.chkLastFullBackupCheckpoint: (0x0,0,0)
Checkpoint: (0x3780A2,8,0)
FullBackup: (0x3771A5,8,16)
FullBackup time: 12/12/2011 11:38:11
IncBackup: (0x0,0,0)
IncBackup time: 00/00/1900 00:00:00
Signature: Create time:02/21/2008 12:29:24 Rand:174026654 Computer:
Env (CircLog,Session,Opentbl,VerPage,Cursors,LogBufs,LogFile,Buffers)
( off, 852, 42600, 16384, 42600, 2048, 2048,2000000000)Operation completed successfully in 0.63 seconds.
Tuesday, December 13, 2011 6:00 AM -
Ibrahim
How can I enable circular logging.
Gulab | MCITP: Exchange 2010-2007 | Lync Server 2010 | Windows Server 2008 | Skype: Exchange.Ranger | Blog: www.ExchangeRanger.Blogspot.com- Proposed as answer by Jonas Andersson [MCITP] Wednesday, December 14, 2011 1:39 PM
- Marked as answer by Rowen-Xu Friday, December 16, 2011 8:03 AM
Tuesday, December 13, 2011 6:43 AM -
Gulab
Does this means ( FullBackup: (0x3771A5,8,16) ) , I can delete all logs before E00003771A5 .
Tuesday, December 13, 2011 10:08 AM -
Yes, you can..
Jonas Andersson | Microsoft Community Contributor Award 2011 | MCITP: EMA 2007/2010 | Blog: http://www.testlabs.se/blog | Follow me on twitter: jonand82Wednesday, December 14, 2011 1:39 PM -
Dear ALL
Thanks for all your support. Instead of doing circular logging I was searching for the real cause. I moved all my committed logs to another area but I couldnt able to do it on 3 log files which were locked by some other process. It is from this timw onwards logs are not getting cleared from the server.
I done all testing with backup exec service and all other service, no way to remove these 3 logs. Finally with process explorer it was found that System process with ID 4 is the one locking these logs. I still doesnt why these logs locked by system procecss.
So before restarting the server I moved the cluster and all services to my passive node. Then after successful moving of cluster to the passive mailbox server, I tried to clear these 3 logs. Hurraaay.. It was successfully gone from there.
I started my full backup again with backup exec .. it went successfully.. and all logs cleared....
I really appreciate all comments from you people which got soem idea for me to play with.
- Marked as answer by Sabir_rm Monday, December 19, 2011 6:48 AM
Monday, December 19, 2011 6:48 AM -
I faced the similar problem with my Exchange Server database, but could not fix the issue by following the steps provided here.
I use Windows backup to backup and restore the database. It is a severe corruption and i get numerous errors while restoring the database.
Then my Admin suggested me to find out some third-party application, and i got Exchange Backup Recovery tool on the third-part tools forum.
I thought it could help other users so i am posting it.
--
Anax
Tuesday, January 24, 2012 8:12 AM