[linux] strace netstat -anp

[roman f.v.]

> -> vedel by mi niekto poradit v com by mohol byt problem?
> -> 
> -> po vykonani prikazu netstat -anp dojde k segmentation fault.
> -> 
> -> ked spustim prikaz cez strace vypise :
> -> 
> -> write(1, "Active UNIX domain sockets (serv"..., 53Active UNIX domain
> -> sockets (servers and established)
> -> ) = 53
> -> write(1, "Proto RefCnt Flags       Type   "..., 82Proto RefCnt Flags 
> ->      Type       State         I-Node PID/Program name    Path
> -> ) = 82
> -> open("/proc/net/unix", O_RDONLY)        = 3
> -> fstat64(3, {st_mode=S_IFREG|0444, st_size=0, ...}) = 0
> -> old_mmap(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS,
> -> -1, 0) = 0x128000
> -> read(3, "Num       RefCount Protocol Flag"..., 4096) = 2688
> -> --- SIGSEGV (Segmentation fault) ---
> -> +++ killed by SIGSEGV +++
> -> 
> -> 
> -> aky by mohol byt dovod SIGSEGV ?
> 
> chyba v binarke (dosahana), chyba v programe (chybne napisany),
chyba cpu
> (overclock, ventilator), chyba ram, nestabilna verzia OS (napr.
debian SID)
> 

hm,
binarka je isto dobra md5sum mi ukazuje ten isty ako ma netstat
na stroji kde prikaz netstat -anp vykona v poriadku.
na vsetkych strojoch bezi rovnaky linux ....

2 stroj ktory robi seg. fault po netstat -anp to robi na tomto mieste:

open("/proc/net/tcp6", O_RDONLY)        = -1 ENOENT (No such file or
directory)
open("/proc/net/udp", O_RDONLY)         = 3
fstat64(3, {st_mode=S_IFREG|0444, st_size=0, ...}) = 0
old_mmap(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS,
-1, 0) = 0x128000
read(3, "  sl  local_address rem_address "..., 4096) = 768
--- SIGSEGV (Segmentation fault) ---
+++ killed by SIGSEGV +++

na ostatnych strojoch sa ten prikaz vykona v pohode....

aby ideas ?

thx... 
____________________________________
P. S. V. P. U.
http://www.pobox.sk/