08-24-2012 02:33 AM
Solved! Go to Solution.
08-31-2012 04:30 AM
Use CAS Array FQDN...as the clients will connect to the CAS array and not the CAS members.
08-24-2012 03:31 AM
Are all your mailbox archiving task set to run at the same time, synch at the same time? I remember you were asking about running synch for all of your task at the same time, is that right? I did say that it is not a good idea.
You need to schedule your task out evenly because you are running 1 EV server to host 8 active tasks. How many active users/mailbox have you got in your exchange environment?
Also go through the MAPI troubleshooting one example technote is here:
http://www.symantec.com/business/support/index?page=content&id=TECH49069
I would also look at clearing out your Outlook profile more frequently form the technote above.
08-24-2012 03:40 AM
Many Thanks LCT, yes after your suggestion we have changed sync to differrent hours throughout the day for differrent archiving tasks,
Number of EV enabled total users about 2000 at the moment spread across 4 active servers in Exchange 2010 DAG, 4 servers are passive.
Will change the MAPI profile expiry time to 1 and see what happens.
Do I need to look at setting registry key "NSPI max sessions per user" in Domain controller?
Thanks again
08-24-2012 03:44 AM
if you are running Windows 2008 R2 domain controllers yes then I would advise you to.
08-24-2012 03:45 AM
Can I also manually remove All profiles, would that have any impact?
08-24-2012 03:59 AM
yes you can delete them manually...when the archiving task runs they will get created...hence you need to remove regularly using the above.
Also I would advise you to schedule for EV services restart on a daily basis this will help clearing out cached connection etc.
08-30-2012 02:19 AM
We went through all troubleshooting steps, MAPI stguffs, we have set profile expiry to 1
Also Manually crealed all profiles
We have NSPI max sessions per user to max 0xffffffff ....
Exchange trottling policy is set
Mailbox Sync and Schedule for each tasks now been scheduled during differrent time slots rather than running all at the same time.
Still some archiving tasks failing randomly, sometimes restarting task controller service fixes the issue, but sometimes some tasks still keeps failing. We found that from EV server if we create a profile for that specific system mailbox for failing tasks and open outlook using that profile, then the task starts fine again.
Any clue please? What can we check to further troubleshoot this issue please?
Many Thanks
08-30-2012 03:09 AM
Ok, I have set the following two reg keys, all archiving task seems to be started fine, let see what happens in next few days
HKEY_CURRENT_USER\Software\Microsoft\Exchange\Exchange Provider
Value name: Closest GC
Data type: REG_DWORD
Radix: Hexadecimal
Value data: 0x00000001
HKEY_CURRENT_USER\Software\Microsoft\Exchange\Exchange Provider
Value name: DS Server
Data type: REG_SZ (string)
Value data: FQDN of the global catalog server
08-30-2012 03:22 AM
If that still gives you problems then I would log a case with Tech Support to do indepth tracing of the problem.
08-30-2012 04:12 AM
08-31-2012 03:38 AM
If we have a CAS Array do we specify the CAS array FQDN or just any specific CAS Member?
08-31-2012 03:39 AM
If we have a CAS Array do we specify the CAS array FQDN or just any specific CAS Member?
08-31-2012 04:30 AM
Use CAS Array FQDN...as the clients will connect to the CAS array and not the CAS members.