TMG server stops responding after every restart
Posted on 2012-03-27
We have strange trouble with TMG server:
It is installed on esx 4.1, OS under TMG is Server 2008 R2 SP1, actual TMG is 2010 SP2, but behaviour was the same in TMG SP1.
We restart the OS, and server hangs during next morning between 8:30 and 10:30 a.m.
Server stops responding to ping from external and internal networks. Login in RDP is not available - it must be done via esx console with server local admin account.
We must stop wspsrv - process manually (service Microsoft forefront firewall hangs in stopping when we attempt to restart the service) and then start Firewall service manually again.
In the event logs are no errors and service state in msc console looks running.
After we restart the process, the server gets to similar state a few times during cca 2-3 hours, we must repeat this 2-3 times and then suddenly everything works correctly and is working until next server restart (patch management etc.)
This behaviour happens again EVERYTIME we restart the server.
Strange is - we restart server at lunch - then all afternoon is the network under load from users and all is OK, in the next morning - happens again.
We installed fresh new Server 2008 R2 with TMG to the same hardware box under esx - import configuration - same behaviour.
No antivirus, backup solution or third party software on new server aplied. Only roles the server hosts is TMG and NPS (only as Radius server).
Service is set as automatic - delayed start.
We did not find any scheduled task that could be related.
Any help would be appreciated.