Thanks for reading this posting and please excuse my cross group posting as I am desperate for some guru help! I have a Win3k server running Exchange 2003 SP1. It's configured as a front end. Every second day or so I get the following error "Process INETINFO.EXE (PID=1416). Topology Discovery failed, error 0x80040a02." From here the server gets really screwed and needs to be rebooted to fix it. The error that immediately follows is: Process MAD.EXE (PID=320). All Domain Controller Servers in use are not responding: srv1.xxxxx.co.nz srv2.xxxxv.co.nz srv3.xxxxx.xxxx.co.nz srv4.xxxx.co.nz These servers all well connected and even reside on the same subnet. The four servers provide the required roles (GC, Config DC, and DC). I have enabled medium diagnostic logging on MSExchangeDSAccess. This shows regular and consistant topology discovery so far (I expect a crash within the next 24 hours and hope the logging will give some better indication). The server has only internal DNS addresses configured and appears to be registered with DNS correctly. The server has no WINS address configured. Can anyone shed some light on this? I have read my articles that range from DNS to Kerberos time outs. thanks, Bob.