One of our servers has in the past few weeks started having high utilization issues (100%) and has become unusable.

It's NW65SP5 and hadn't had any patches loaded prior to this problem popping up. It runs NPS and DNS. I can't put a newer service patch on as it kills NPS.

Stopping NPS (Apache & Tomcat) and DNS does nothing, or at least it doesn't for long.

I've tried a newer TCPIP and setting the file cache maximum size in the autoexec and startup to flush the nss cache (hey i'll try anything)

When I look at Profile/Debug in the NRM, it's always the same -

Thread Name: Server XX
Thread ID: xxxxxxx
Processor Assignment: 0
Total Execution Time: xxxxxxx
MP Status: In classin NetWare
Stack Bottom: xxxxxxx
Current Stack Pointer: Not available for running threads
Stack Top: xxxxxx
Stack Size: 32,640 Bytes
Suspend Reason: Yielded CPU

Active Work To Do Information
Work To Do ID: xxxxxxxx
Work Location: xxxxxxxx(TCPIP.NLM|RestoreLBPolicy+2FFC)
Work To Do Description: snmp_UDP_Scheduler_Process
Work Owner: TCPIP.NLM

Any help from minds immeasurable superior to mine would be greatly appreciated.