Raymond Brooks
asked on
Exchange domain request caching - Wont use other DC's when the primary goes down
Hi folks, So in my clients environment they have multiple Domain Controllers, but whenever a DC goes down Exchange keeps sending requests to that DC instead of using the other available ones, and this process doesnt time out for at least 12 hours, after said 12 hours it decides to use to another DC.
Is there any way to speed up this process or force Exchange to use the other DC?
Thanks in advance.
Is there any way to speed up this process or force Exchange to use the other DC?
Thanks in advance.
SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
By Default, Exchange checks for the available DC’s every 15 mins. In case any DC is rebooted/down, next time when exchange does the query, it identifies that DC which is not reachable and will mark it as down. After which, Exchange will try using another available DC.
If Exchange doesn’t gets a response from a DC, it will not switch to another DC immediately. Only when it thinks that the DC is not functional, it switches to another DC. Till that time, it will continue to send the requests to the DC which isn’t responding also.
Check few things:
1. Primary and Alternative DNS should be there to respond on Exchange server.
2. All DCs should have its IP as primary DNS, even it can create such issue.
However, we have same issue on few Exchange servers in our environment. I have already logged a case in MS and working with them. In last reboot activity of DC we have capture Pref logs, Net Mon logs and Event viewer logs those are given to MS for analysis. but still waiting for reply.
If Exchange doesn’t gets a response from a DC, it will not switch to another DC immediately. Only when it thinks that the DC is not functional, it switches to another DC. Till that time, it will continue to send the requests to the DC which isn’t responding also.
Check few things:
1. Primary and Alternative DNS should be there to respond on Exchange server.
2. All DCs should have its IP as primary DNS, even it can create such issue.
However, we have same issue on few Exchange servers in our environment. I have already logged a case in MS and working with them. In last reboot activity of DC we have capture Pref logs, Net Mon logs and Event viewer logs those are given to MS for analysis. but still waiting for reply.
ASKER CERTIFIED SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
ASKER
Thanks, just yesterday i was going over some documentation on DAC and saw the Preferred AD command.
On your exchange server you need the following...
- Primary and Secondary DNS entires (both DNS servers need to be in the same AD Site, where Exchange resides)
- Both domain controllers need to be DC/DNS/GC
Make sure that you check your Active Directory Health and Replication using the following commands..
repadmin /replsum
repadmin /showrepl
repadmin /bridgeheads
DCDiag /v
If everything is correct there has been a known issue with Microsoft DNS called DNS sticky Keys. This is a syndrom when DNS will not release from the primary. This is not normal behavior and it should fail over to secondary just after 1 second of not being available.
Will.