An using MWG 6.8.7 that uses native NTLM to authenticate to a single AD domain.
Recently users are been requested to re-authenticate at randon whilst connected to the internet via WW. This occurs during peak usage periods and normally a refresh or browser restart fixes the problem for end-user. There are 3 WW appliances in master-site instance arrangement, and some 30,000 users. I can not enable Trace as the IP Address affected is randon and I can not replicate fault in a controlled environment. The frequency of fault is low compared to total usage.
I have observed under User Management > Windows Domain Membership > that the Status for Domain Controllers changes from Green to Black momentarily.
I have changed the DC's ( using both VM and Physical DC ), have check the DC log files ( no related errors and no other reported problems on AD land), and implemented multi-process on to the WW appliances ( as suggested by McAfee Support ). Additionally there is nothing abnormal on the WW error log files.
I am aware that all domain controllers were upgraded to 2008R2 recently but authentication issue started way after the Windows server upgrade.
Interested if anyone out there has experienced similar problem.