cancel
Showing results for 
Search instead for 
Did you mean: 

Hardware crash - Restore Exchange 2003 Information Store

banan2
Level 4
Hello,

Recently a customer of ours had a critical hardware failure on their Exchange server. The whole RAID array was destroyed and we had to put up a new server.
The system is up and running, the email was cached to clients, so we didn't do any restore job on Exchange, until we found out that they missed a few public folders. It is not many users in the enviroment anyway.

Now I am trying to restore those public folders, but I can't seem to actually do it.

So far I have done this, in the following order:
  1. Installed a server with Server 2003, same name as the crashed server.
  2. Installed Server 2003 SP 2
  3. Joined the server to the domain
  4. Installed Backup Exec 12.5 with the Exchange Agent
  5. Installed Exchange Enterprise with the /disasterrecovery switch
  6. Installed Exchange SP2 with the /disasterrecovery switch
  7. Tried to run a restore in Backup Exec. Just a simple restore with only the Information Store selected. No redirection, no modifications in the options for the Exchange Agent.
  8. Failure.
The error message I get is this:

0xe000ff00 - Unable to restore some Microsoft Exchange components because one or more of the databases are currently mounted or because the Recovery Storage Group is not properly configured. Use the Exchange System Manager to check the configuration of the Recovery Storage Group and the mount status of the databases.

The database is not currently mounted, it's a clean /disasterrecovery installation. I have not done anything with the RSG as I have no experience with it and as far as I have read the RSG should not be used to restore public folders or complete Information Store Backups.

The crashed server had Exchange Standard, does it matter if I use Enterprise to restore?
Have I even done it the right way? As far as I have read of guides on the internet that is pretty much the way to do it.
Just ask if you need more information.

Thanks in advance.

26 REPLIES 26

banan2
Level 4
Anyone?

CraigV
Moderator
Moderator
Partner    VIP    Accredited
Hey dude,

Have you tried to stage to disk first? There are some issues when restoring from tape, and staging to disk first tends to help with that.
We had an Exchange server crash last year with the same issue...RAID failed when 1 drive failed, and the site rep pulled a second drive out the server!
We staged to disk first to get the *.stm and *.edb files, and from there mounted the Exchange IS.

https://www-secure.symantec.com/connect/articles/restoring-exchange-or-individual-mailboxesitems-using-backup-exec-howto

Laters!

Matt_Alexander
Level 5
Did staging to disk like CraigV said solve the issue you were having?

banan2
Level 4
Hello,

Sorry for the delay, did get a problem with cataloging my media aswell, but fixed that now.

No, I tried to duplicate the set to a disk and restore from there, but it caused the same error message.

CraigV
Moderator
Moderator
Partner    VIP    Accredited
Hi there,

Have you upgraded to the latest SP which is SP3 for BEWS?

Martin_Forster1
Level 5
Hi

some suggestions:

is the "logical" Path to the db the same ?  (Adminstrative Group, Database Group ...) Check the logfile.
     - there was (maybe is) a bug in the redirecting feature.

If you have created a empty database to be overwritten, is the check mark set to allow the overwrite ?

In the restore job( or was it in the agent) is box to define a temporary working directory which must be able to hold the whole DB.

Have restored the whold DB Group ? I had isssues in the past when i have not restored all the data in one restore job ( db and logs)

If this all hints does not help. Enable the the logging on the agent and the media server, and post it with the application log of your exchange server, please.

Regards

Martin forster

banan2
Level 4
Yes, I have the lastest SP.

banan2
Level 4
The logical path is the same, with pretty much the default Exchange config.

The check mark is set to allow overwrite, yes.

The temporary dir is working and have enough capacity to hold the DB.

I have tried to restore the whole storage group yes.


I will enable logging and check it out. Coming back to you :)

Martin_Forster1
Level 5
maybe you have the budget for this one

http://www.ontrackpowercontrols.com/solutions/exchange-email-recovery/
they can directly work with .bkf files.

Kind regards ...

banan2
Level 4
I have heard of Ontrack Powercontrols, and we don't have the budget for it unless we have to.

I enabled logging and looked through it and found some things, but not really sure what it means.
Backup Exec log is uploaded here: http://files.tocom.no/logfiles/SERVER2003-bengine01.txt
The only application log message that comes up at restore is this: http://files.tocom.no/logfiles/eventlog.txt

Martin_Forster1
Level 5
Hi banan2

The outcome of the logs - nearly none.

Please check if the Service Accounts of Backup and Exchange have write Permissions, and disable the virus scanner.
Permissions are needed in the temp directory and in the Exchange Server directories.
Does the orginal drive letters exist, where the database was before ?


Do you know this guide : http://technet.microsoft.com/en-us/library/aa996770%28EXCHG.65%29.aspx

Please provide me a copy of the Backup Agent Log.
And from the Application log im interested of the soruce MsexchangeIS and ESE.
And provide me the restore.env file mentioned in the ms article ( probably located in the temp directory.

How to enable the agent logging
http://seer.entsupport.symantec.com/docs/254212.htm



banan2
Level 4
Hi,

Thanks for the help so far.

I am using the Administrator account as my service account, and it does have permission to the temp folder and the Exchange server folders.
The original drive letters are being used, at least on the disk with the database.

I have read that guide and tried everything in it, unfortunately with no luck.

I have uploaded the Agent log here: http://files.tocom.no/logfiles/SERVER2003-beremote01.txt


In between the MSExchangeSA errors that comes in application log, appears two informationlogs from MSExchangeIS which contains only "Exchange VSS Snapshot metadata document has been collected". ID 9604. There is no ESE entries.

There is no .env file in the temp folder.

Martin_Forster1
Level 5
 Hi banan2

according to your agent logs there seems be a issue with vss.
It tries to communicate with the vss and get
0x80010119, which results in some follow up error messages.

can verfiy the vss on that server ?
additional Informatino is available here
http://msexchangeteam.com/archive/2008/08/25/449684.aspx

command vssadmin list writers

it must contain something a Exchange Writer.


Maybe its time for a workaround, install 2nd exchange server with a RSG, and redirect the restore.
Or make a granular restore of the objects into a empty database.

Regards

banan2
Level 4
I have a Microsoft Exchange Writer, and it is stable.

Install yet another Exchange server? Wouldn't that pretty much cause the same error? 

GRT backup was not enabled on the crashed server, won't be able to do a granular restore.

Martin_Forster1
Level 5
Hi banan2,

no i dont think that the failure will occur again, because the vss returns an error after it gets called by the agents.

As next step of guessing you could enable vss tracing, this is described here:
http://blogs.technet.com/asksbs/archive/2008/04/23/troubleshooting-vss-and-backup.aspx

and post that log.
and check the performance of the machine and its disks, vss problems often are performance problems.

regards

CraigV
Moderator
Moderator
Partner    VIP    Accredited

If you suspect your VSS writer, open up a command prompt and run:

vssadmin list writers

This will show an output like:

Writer name: 'System Writer'
   Writer Id: {e8132975-6f93-4464-a53e-1050253ae220}
   Writer Instance Id: {70e35b00-9921-434c-b886-013b6ea4d958}
   State: [1] Stable
   Last error: No error

Writer name: 'SqlServerWriter'
   Writer Id: {a65faa63-5ea8-4ebc-9dbd-a0c4db26912a}
   Writer Instance Id: {6894931d-9c22-4578-8f4b-f408dfd63767}
   State: [1] Stable
   Last error: No error

Writer name: 'MSDEWriter'
   Writer Id: {f8544ac1-0611-4fa5-b04b-f7ee00b03277}
   Writer Instance Id: {850e1b49-3546-433a-8dde-536320cd250c}
   State: [1] Stable
   Last error: No error

Writer name: 'WMI Writer'
   Writer Id: {a6ad56c2-b509-4e6c-bb19-49d8f43532f0}
   Writer Instance Id: {23222a01-3df0-492f-9e38-4b5cc67ad1e7}
   State: [1] Stable
   Last error: No error

Writer name: 'BITS Writer'
   Writer Id: {4969d978-be47-48b0-b100-f328f07ac1e0}
   Writer Instance Id: {b2ec3364-2069-4daf-96f8-40893e04666e}
   State: [1] Stable
   Last error: No error

Writer name: 'Event Log Writer'
   Writer Id: {eee8c692-67ed-4250-8d86-390603070d00}
   Writer Instance Id: {02ed8110-65a6-4a62-8517-704bf1d41f9d}
   State: [1] Stable
   Last error: No error

Writer name: 'COM+ REGDB Writer'
   Writer Id: {542da469-d3e1-473c-9f4f-7847f01fc64f}
   Writer Instance Id: {a5869bcd-67c3-43d5-bc65-9cb66d259d59}
   State: [1] Stable
   Last error: No error

Writer name: 'Registry Writer'
   Writer Id: {afbab4a2-367d-4d15-a586-71dbb18f8485}
   Writer Instance Id: {f2ac38e9-1344-47a3-8335-761223e1c090}
   State: [1] Stable
   Last error: No error


If there ARE any errors, it's going to state that. If so, give the tech article below a read on how to sort out VSS errors:

http://seer.entsupport.symantec.com/docs/261993.htm

banan2
Level 4
Sorry for not replying yet, I haven't had much spare time the last week.

I have yet not tried to restore from a second exchange server, but I will try next week. I will post the results, most probably on monday.

banan2
Level 4
Hello,

I did not work to restore from a second server, same error on that one.

I did turn on VSS logging and got this out of the logfile (trace.txt on C:\):
[0000032812,0x0004c0:0x05f0:0xb42b5fc4] modules\sec\security.cxx(2200): CVssSidCollection::AddSid: Adding SID: SERVER2003\Administrator
[0000032812,0x0004c0:0x05f0:0xb42b5fc4] EXIT  {CVssSidCollection::AddSid}: hr: 0x00000000
[0000032812,0x0004c0:0x05f0:0xb42b5fc4] ENTER {CVssSidCollection::AddSid}:
[0000032812,0x0004c0:0x05f0:0xb42b5fc4] modules\sec\security.cxx(2141): CVssSidCollection::AddSid: Adding SID S-1-5-21-4143271690-2742500164-1435389183-512 ...
[0000035375,0x000340:0x05c8:0xa862afd4] modules\sec\security.cxx(2162): CVssSidCollection::AddSid: Unexpected error while mapping SID to a real account. [1789] Ignoring SID...
[0000035375,0x000340:0x05c8:0xa862afd4] EXIT  {CVssSidCollection::AddSid}: hr: 0x00000000
[0000035375,0x0004c0:0x05f0:0xb42b5fc4] modules\sec\security.cxx(2162): CVssSidCollection::AddSid: Unexpected error while mapping SID to a real account. [1789] Ignoring SID...
[0000035390,0x0004c0:0x05f0:0xb42b5fc4] EXIT  {CVssSidCollection::AddSid}: hr: 0x00000000
[0000035390,0x000340:0x05c8:0xa862afd4] ENTER {CAutoSid::CreateBasicSid}:
[0000035390,0x000340:0x0488:0xa862afd4] modules\sec\security.cxx(2162): CVssSidCollection::AddSid: Unexpected error while mapping SID to a real account. [1789] Ignoring SID...
[0000035390,0x000340:0x0488:0xa862afd4] EXIT  {CVssSidCollection::AddSid}: hr: 0x00000000
[0000035390,0x0004c0:0x05f0:0xb42b5fc4] ENTER {CAutoSid::CreateBasicSid}:
[0000035390,0x000340:0x05c8:0xa862afd4] EXIT  {CAutoSid::CreateBasicSid}: hr: 0x00000000
[0000035390,0x0004c0:0x05f0:0xb42b5fc4] EXIT  {CAutoSid::CreateBasicSid}: hr: 0x00000000
[0000035390,0x000340:0x05c8:0xa862afd4] ENTER {CAutoSid::CreateBasicSid}:
[0000035390,0x0004c0:0x05f0:0xb42b5fc4] ENTER {CAutoSid::CreateBasicSid}:
[0000035390,0x000340:0x05c8:0xa862afd4] EXIT  {CAutoSid::CreateBasicSid}: hr: 0x00000000
[0000035390,0x0004c0:0x05f0:0xb42b5fc4] EXIT  {CAutoSid::CreateBasicSid}: hr: 0x00000000
[0000035390,0x000340:0x05c8:0xa862afd4] ENTER {CVssRegistryKey::Close}:
[0000035390,0x0004c0:0x05f0:0xb42b5fc4] ENTER {CVssRegistryKey::Close}:
[0000035390,0x000340:0x05c8:0xa862afd4] EXIT  {CVssRegistryKey::Close}: hr: 0x00000000
[0000035390,0x0004c0:0x05f0:0xb42b5fc4] EXIT  {CVssRegistryKey::Close}: hr: 0x00000000


It seems like it has some problems with the SID on the Administrator user. I have no clue about why or what.
The pretty much same error shows up at the bottom of the logfile, you can read the whole thing here -> http://tocom.no/Files/logfiles/trace.txt

CraigV
Moderator
Moderator
Partner    VIP    Accredited
Mmm...OK, create another account with the same credentials and then use that for the restore. What you need to do though is add that user into Backup Exec as well.