[parisc-linux] nslookup segmentation fault

[parisc-linux] nslookup segmentation fault

[FAUCHER,DENNIS (HP-USA,ex1)]

Hello,

Is there a problem with nslookup?  Mine returns a segmentation fault.  Thank
you.

Sincerely,
Dennis Faucher
Technical Consultant
Hewlett-Packard Company
https://www.ecardfile.com/id/dennis_faucher (Contact)

Re: [parisc-linux] nslookup segmentation fault

[Randolph Chung]

> Is there a problem with nslookup?  Mine returns a segmentation fault.  Thank
> you.

paer[8:38] ~% uname -a               
Linux paer 2.4.16-64-smp #1 SMP Sat Dec 29 02:33:39 MST 2001 parisc64 unknown

paer[8:38] ~% nslookup -silent www.debian.org
Server:         127.0.0.1
Address:        127.0.0.1#53

Non-authoritative answer:
Name:   www.debian.org
Address: 198.186.203.20

what version of kernel/glibc? did you change anything in
/etc/nsswitch.conf?

the last dozen lines or so of a strace might be useful.

randolph
-- 
   @..@                                         http://www.TauSq.org/
  (----)
 ( >__< )
 ^^ ~~ ^^

Re: [parisc-linux] nslookup segmentation fault

[Randolph Chung]

Neat....

Here's part of Dennis' strace of nslookup:
> m4450dmf:/etc# strace nslookup athp.hp.com
[...]
> rt_sigprocmask(SIG_BLOCK, [HUP INT TERM], NULL, 8) = 0
> getpid()                                = 22574
> socket(PF_INET, SOCK_STREAM, IPPROTO_IP) = 3
> close(3)                                = 0
> socket(PF_INET6, SOCK_STREAM, 0)        = -1 EAFNOSUPPORT (Address family
> not su
> pported by protocol)
> brk(0)                                  = 0x2f000
> brk(0x2f030)                            = 0x2f030
> brk(0x30000)                            = 0x30000
> brk(0x33000)                            = 0x33000
> --- SIGSEGV (Segmentation fault) ---
> +++ killed by SIGSEGV +++

why is it trying to do ipv6?

randolph
-- 
   @..@                                         http://www.TauSq.org/
  (----)
 ( >__< )
 ^^ ~~ ^^

Re: [parisc-linux] nslookup segmentation fault

[Carlos O'Donell Jr.]

> Neat....
> 
> Here's part of Dennis' strace of nslookup:
> > m4450dmf:/etc# strace nslookup athp.hp.com
> [...]
> > rt_sigprocmask(SIG_BLOCK, [HUP INT TERM], NULL, 8) = 0
> > getpid()                                = 22574
> > socket(PF_INET, SOCK_STREAM, IPPROTO_IP) = 3
> > close(3)                                = 0
> > socket(PF_INET6, SOCK_STREAM, 0)        = -1 EAFNOSUPPORT (Address family
> > not su
> > pported by protocol)
> > brk(0)                                  = 0x2f000
> > brk(0x2f030)                            = 0x2f030
> > brk(0x30000)                            = 0x30000
> > brk(0x33000)                            = 0x33000
> > --- SIGSEGV (Segmentation fault) ---
> > +++ killed by SIGSEGV +++
> 
> why is it trying to do ipv6?
> 
> randolph
> -- 


I'm feel worried about what brk() is trying to do.

PF_INET==2 and PF_INET6==10 are only a single bit appart :}
Cosmic radiation maybe?

I'm not sure what is going on. Though I tried to build
Bind 8.3.1 on x86 and if failed (need to look into bits/types).

eventlib.c: In function `__evGetNext':
eventlib.c:298: structure has no member named `fds_bits'
eventlib.c:299: structure has no member named `fds_bits'
eventlib.c:300: structure has no member named `fds_bits'

c.

Re: [parisc-linux] nslookup segmentation fault

[Jochen Friedrich]

Hi Carlos,

> I'm feel worried about what brk() is trying to do.

brk() is used to increase the core size. Nothing unusual.

> PF_INET==2 and PF_INET6==10 are only a single bit appart :}
> Cosmic radiation maybe?

Nope. The resolver supports both IPv4 and IPv6. The following just seems 
to be a probe if both AFs are available.

getpid()                                = 2880
socket(PF_INET, SOCK_STREAM, IPPROTO_IP) = 3
close(3)                                = 0
socket(PF_INET6, SOCK_STREAM, 0)        = 3
getsockname(3, {sin_family=AF_INET6, sin6_port=htons(0),  
inet_pton(AF_INET6, "::", &sin6_addr), sin6_flowinfo=0, sin6_scope_id=0}, [28]) = 0
close(3)                                = 0
brk(0)                                  = 0x2f000
brk(0x2f030)                            = 0x2f030
brk(0x30000)                            = 0x30000
brk(0x33000)                            = 0x33000
--- SIGSEGV (Segmentation fault) ---
+++ killed by SIGSEGV +++

gdb is even worse:

(gdb) run
The program being debugged has been started already.
Start it from the beginning? (y or n) y

Starting program: /usr/bin/nslookup www.kame.net
(no debugging symbols found)...(no debugging symbols found)...
(no debugging symbols found)...(no debugging symbols found)...
(no debugging symbols found)...(no debugging symbols found)...
(no debugging symbols found)...(no debugging symbols found)...
Program received signal SIGSEGV, Segmentation fault.
Cannot access memory at address 0x0
(gdb) bt
Cannot access memory at address 0x0

Looks like yet another glibc related problem. In particular since nslookup 
seems to use pthread (which *does* have issues with the latest glibc):

open("/lib/libpthread.so.0", O_RDONLY)  = 3
read(3, "\177ELF\1\2\1\3\0\0\0\0\0\0\0\0\0\3\0\17\0\0\0\1\0\0@`"..., 1024) 
= 1024

Cheers,
--jochen