Some (most?) processes don't cleanup their pid files when

they exit. Sometimes they simply can't (core-dump, for example).
So, when searching for a running program send standard error
output from ps(1) to never-never land.

	# This should quite those
	# ps: kvm_getprocs: No such user
	# errors. Since ports use a duplicate rc.subr(8) you might
	# still see this error from ports startup scripts.

etc/rc.subr

@@ -267,7 +267,7 @@ _find_processes()
 	fi
 
 	_proccheck='
-		ps -o "pid,command" '"$_psargs"' |
+		ps 2>/dev/null -o "pid,command" '"$_psargs"' |
 		while read _npid '"$_fp_args"'; do
 			case "$_npid" in
 			    PID)