try not to laugh but we still have a number of nt4 servers in our company and recently (over the last 3 weeks or so) we've been experiencing some major issues with all of them.
NB: these servers were all physical and now they have been virtualised (P2V) onto ESX servers.
these servers run an application called emslink and for some bizarre reason the server will run perfectly for about a week and then suddenly the CPU hits 100% and the application that the servers are running fails. we checked task manager when the application died and we've managed to work out that the CPU hits 100% because the emslink.exe process is using all of the CPU (and then the server runs out of resources and the application fails).
this problem with the emslink.exe process using 100% CPU has been seen before and normally to resolve the problem we would just connect to the server using PCAnywhere and just reboot it, but now we are experiencing a second issue and that is that when you try to reboot the server it gets stuck at "writing unsaved data to the disk" or something like that and won't shutdown / restart. The only way we can restart the server is to connect to the hypervisor (esx 4.1 / vsphere client) and reset the virtual machine.
We've not had these issues before and I can't work out what the problem is or why it's happening now.
All servers run SP6a and windows nt4 server.
Richard S Powell