utmp/wtmp problem!

[nestea <nestea@myrealbox.com>]

hi folks,

my debian box was newly installed, upgraded to unstable a week ago.  i compiled 2.4.9 kernel yesterday and it works just fine.  this morning i login as root and fire the 'w' command, i got weird result ;

 14:37:30 up 1 day, 20:30,  2 users,  load average: 0.00, 0.00, 0.00
USER     TTY      FROM              LOGIN@   IDLE   JCPU   PCPU  WHAT
root     pts/0    xxx.xxx.xxx.xxx 09:46    0.00s  0.31s  0.01s  w 
debian:/etc/init.d#

i wonder why it shows that there are currently 2 users, then i do a 'ps axf';

sunday:/etc/init.d# ps axf
  PID TTY      STAT   TIME COMMAND
    7 ?        SW     0:00 [kupdated]
    6 ?        SW     0:00 [bdflush]
    5 ?        SW     0:00 [kreclaimd]
    4 ?        SW     0:06 [kswapd]
    3 ?        SWN    0:00 [ksoftirqd_CPU0]
    1 ?        S      0:03 init [2]  
    2 ?        SW     0:00 [keventd]
    8 ?        SW     0:05 [kjournald]
  127 ?        S      0:00 /sbin/syslogd
  130 ?        S      0:00 /sbin/klogd
  141 ?        S      0:00 /usr/sbin/inetd
  154 ?        S      0:04 /usr/sbin/sshd
 1852 ?        S      0:01  \_ /usr/sbin/sshd
 1853 pts/0    S      0:00      \_ -bash
 2427 pts/0    R      0:00          \_ ps axf
  171 ?        S      0:00 /usr/sbin/atd
  174 ?        S      0:00 /usr/sbin/cron
  182 tty1     S      0:00 /sbin/getty 38400 tty1
  254 ?        S      0:00 /usr/sbin/squid -D -sYC
  257 ?        R      1:24  \_ (squid) -D -sYC
 2240 ?        S      0:00      \_ (unlinkd)
sunday:/etc/init.d#

i still have not idea ... i know i can
'cat /dev/null > /var/run/utmp' and
'cat /dev/null > /var/log/wtmp'

to resolve this, but i really want to know why.

any idea?