[ale] Memory leak (hacked?)

[David Corbin]

I have had my public server develop a sudden memory leak.   (Honest - it is 
sudden - I've done nothing to that system recently).  I keep ending up with 
the following kernel message: __alloc_pages: 0-order allocation failed 
(gfp=0xf0/0)  repeatedly.

When I reboot this morning, I watch top, and sure enough, you could see the 
memory steadily creeping upwards.  This happens even when I boot into single 
user mode, and have ridiculously free processes:

  PID TTY      STAT   TIME COMMAND
    1 ?        S      0:03 init [S]
    2 ?        SW     0:00 [keventd]
    3 ?        SWN    0:00 [ksoftirqd_CPU0]
    4 ?        SW     0:00 [kswapd]
    5 ?        SW     0:00 [bdflush]
    6 ?        SW     0:00 [kupdated]
    9 ?        SW     0:00 [khubd]
   13 ?        SW     0:00 [kjournald]
   96 ?        SW     0:00 [kjournald]
   97 ?        SW     0:00 [kreiserfsd]
  125 ?        S      0:00 /sbin/portmap
  174 tty1     S      0:00 init [S]
  175 tty1     S      0:00 bash
  179 tty1     R      0:00 ps ax

I'm running 2.4.23

I can only think to (a) upgrade the kernel, or (b) reinstall (ugh).  Anyone 
have other ideas?

Fixing this is very important to me, as it's my email server...
-- 
David Corbin <dcorbin at machturtle.com>