EV 9 & Exchange 2010: There was a problem accessing an Exchange 2010 or later server

[RenaudT_]

Hi all,

I'm implementing an EV 9 SP3 solution with Exchange 2010 SP1 on a MS Cluster 2008 R2.

I added my MBX servers with my system mailbox account. The task in running fine. But after adding my first test user and do some manual store from Outlook I have those errors poping every minutes:

Warning 3419

There was a problem accessing an Exchange 2010 or later server.  The dispenser will re-queue the current item and sleep for 1 minute(s).

Task: Exchange Mailbox Archiving Task for serverXX

Information 2217

Message dispenser will resume processing

Task: Exchange Mailbox Archiving Task for serverXX

I can see my MSMQueues growing with mails, but no mails are really archived in EV.

I have executed the script for throttling against the EV service account. This account as send as permissions for my EV system mailbox.

So does any one have this behavior before?

Thank you very much!

Renaud

[JesusWept3]

Try using the DS Server to point at the CAS Server/Array that EV is targeting
http://support.microsoft.com/kb/319206

Make sure you are logged in as the EV Admin user when you do this and that you add it to HKEY_CURRENT_USER and not HKEY_LOCAL_MACHINE

It is also worth running the deployment scanner again to ensure that all the send as/receive as permissions check out correctly, that you can log on to the EV System mailbox you specifed as the EVAdmin user *without* being prompted for a username/password (if you are prompted, then you do not have the send as/receive as permissions set correctly)

Also check the event logs of the CAS Server itself to make sure that you aren't breaching any of the Throttling policies that Exchange 2010 uses which severely restricts connections (for most users its fine, but for EV it will end up with all sorts of connection issues otherwise)

[RenaudT_]

Jesus,

I applied the DS Server to point to my primary GC. Reboot both nodes of my cluster, and no changes.

The deployment scanner is all green, and I don't have any pop up for credentials when I launch Outlook.

On CAS servers there are no errors specific to EV or Throttling in the Windows Logs. Is there somewhere else to check?

Any other idea?

Many thanks!

Renaud

[JesusWept3]

Ok so the system mailbox you are targeting, that is definitely on the correct exchange server or on a server that is part of the DAG right? Also the CAS server that you put in DS server is that part of a software or hardware array? Another thing is if you have Blackberry Enterprise Server in the environment targeting Exchange 2010 then try having a word with the BES Admin people as they may have had similar issues also Last things are 1. Try changing DS Server to target a GC instead of a CAS server 2. If that doesn't work, try the Closest GC reg key from the link above 3. Ensure you have the performance hotfix installed for outlook 2007 Also with the DS/GC regkeys you dont need to reboot the machine, so as long as you're sure they went to HKEY CURRENT USER when logged in to the evadmin accept, assume it didn't work

[RenaudT_]

System Mailbox is on a database that is monted on the correct MBX server that is part of a DAG. And I put, in the DS entry, one of my two CAS. I've tested with GC, CAS,... doesn't change anything :(

The peformance hotfix was already installed.

It's really strange because no mails at all are archived, but MSMQ are growing when I manually store in vault. 

[JesusWept3]

Hmmm, any chance of getting a DTrace of Archiving Task?

1. Stop the Archiving Task for your Exch 2010 server
2. Open a Command Prompt and CD to your \Program Files (x86)\Enterprise Vault\ directory
3. Type "DTrace" and press Enter
4. Type "set ArchiveTask v" and press Enter
5. Type "log C:\Temp\ArchiveTaskFail.txt" and press Enter
6. Minimize the command prompt
7. Start the Archive Task back up
8. Let the task run for a couple of minutes and generate those same events
9. Go back to the command prompt and type "Exit" and it will then close the DTrace and stop logging
10. Upload the Log file here or PM the log file

I've seen it in my environment where it looks like purely a connection issue but there is something else in the background causing it to fail (in my case it was trying to sync a user that was deleted from the environment and it couldnt be found in the GAL)

 

[RenaudT_]

Here is the Dtrace.

Warning appears every minutes, at around the 29th second

Thank you for your help Jesus :)

[JesusWept3 · Accepted solution]

OK So!
Reading through the dtrace it is not an issue with connectivity to Exchange Server or with the GC server or with permissions or with the system mailbox, they all check out fine

It appears that it is failing attempting to archive two mailboxes and says that another thread is currently archiving that mailbox, puts it back in the queue and then picks it back up again and loops until it fails

Can you check your A2 queue for messages from Renaud Touillet and LGonzalez?
You may need to purge them, as there is either something wrong with the messages or corrupt in their mailbox?

Typically they should be the top messages listed within that queue, you may even want to connect to their mailbox and see if you can find the items and open them and see if they will save out to an MSG file

[JesusWept3]

Also there are a number of MAPI_E_FAILONEPROVIDER (0x8004011D) when attempting to open the mailbox's root folder for these two guys

[JesusWept3]

Yet one more thing, can you check the CAS Server (chgva-exclb.mercuria.net) for any mention of the EVAdmin user in the Application logs, specifically any errors that look like this

Process Microsoft.Exchange.RpcClientAccess.Service.exe (PID=3788). User 'Sid~YOURDOMAIN\EVAdmin~RCA~false' has gone over budget '124' times for component 'RCA' within a one minute period. Info: 'Policy:EVPolicy, Parts:CAS:123;'. Threshold value: '100'.