My Linux system has been up for 6 days now. Never high loads maybe 1 or so
(My 50 users aren't busy users). It also does tcp/ip and ipx routing.
Suddenly today I have a 5 load average the whole day.
At first I saw nothing unusual running but then I noticed a few processes
like this
root 12440 0.0 0.0 768 0 ? DW 16:03 0:00 (in.ntalkd)
root 12450 0.0 0.0 768 0 ? DW 16:05 0:00 (in.ntalkd)
root 12460 0.0 0.0 768 0 ? DW 16:07 0:00 (in.ntalkd)
root 12467 0.0 0.0 768 0 ? DW 16:09 0:00 (in.ntalkd)
root 12491 0.0 0.0 768 0 ? DW 16:11 0:00 (in.ntalkd)
Now as you see they don't really take any CPU time also at first I thought
in.ntalkd was just buggy and it had nothing to do with the kernel.
However I haven't managed to kill them up to now. Not even with -9 (SIGTERM)
That's why I think it's needs to be seen by the kernel list because -9 is
supposed to NUKE it no matter if it's buggy or not.
Any ideas? Is the load average fake? (The system doesn't seem slower and
usually with a 2 load average i can feel it 486/dx4 100 16m mem 64 mb swap)
Thanks,
Andi Gutmans