> I recall from quite a while back that people were having problems
> with identd running wild and consuming large percentages of CPU. Was
> any progress ever made fixing this? I recently upgraded a sparc to
> a -current snapshot from Dec 3rd, with a kernel from Feb 14th, and 
> I'm experienceing it very frequently... 

Someone said it could be a design flaw, not a simple bug.

I was thinking of setting a CPU time limit on identd (just like getty
does) to kill the crazy copies after, say, 5 minutes of running.  This
doesn't fix the bug, but it saves me from killing the processes.  I hope
to do this sometime soon.

  -jm