Re: Bash Blues.

From: Kurt Seifried (kurtat_private)
Date: Thu Feb 13 2003 - 21:31:46 PST

  • Next message: Dack: "Re: Bash Blues."

    > uk2sec /bin/bash Advisory
    >
    > By sending a perl request on the GNU bash terminal we can cause a
    > Segmentation Fault.
    >
    > Work done was based on:
    > GNU bash, version 2.05a.0(1)-release (i686-pc-linux-gnu)
    > (Redhat 7.3)
    >
    
    Interesting. Logged in via ssh to a red hat 7.3 and an 8.0 system (both are
    completely up to date) doing that command immediately logs me out (bash
    falls down badly). Other then that the system is fine, no weird load/etc.
    For a quick moment bash spikes, but 2.5% cpu usage on a 600 mhz cyrix
    processor is not exactly scary ditto for memory, 1.2% out of 248 megs (256 -
    8 for the built in video) is not worrying. No resource limits are placed on
    bash via ulimit or the session via pam limits so it's not booting me out
    because of that.
    
    CPU states:  2.5% user,  2.3% system,  0.0% nice, 95.0% idle
    Mem:   247516K av,  239088K used,    8428K free,       0K shrd,   61180K
    buff
    Swap:  262072K av,   14456K used,  247616K free                  109576K
    cached
      PID USER     PRI  NI  SIZE  RSS SHARE STAT %CPU %MEM   TIME COMMAND
      370 seifried  16   0  3112 3112  1096 R     2.5  1.2   0:00 bash
      769 postfix   15   0  1256 1104  1000 S     0.7  0.4  13:52 qmgr
      366 seifried  15   0  1064 1064   820 R     0.7  0.4   0:00 top
      606 root      15   0  1356 1272  1188 S     0.1  0.5   0:47 sshd
    
    On Solaris with bash from sunfreeware (I think):
    
    $ /usr/local/bin/bash
    bash-2.05$ /usr/local/bin/bash --version
    GNU bash, version 2.05.0(1)-release (sparc-sun-solaris2.8)
    Copyright 2000 Free Software Foundation, Inc.
    bash-2.05$ uname -a
    SunOS sparkplug 5.8 Generic_108528-15 sun4u sparc SUNW,Ultra-1
    bash-2.05$ `perl -e 'print "*/*" x 2338'`
    Segmentation Fault - core dumped
    
    takes a few seconds but then it seg faults. Who knows, maybe it is
    exploitable.
    
    Kurt Seifried, kurtat_private
    A15B BEE5 B391 B9AD B0EF
    AEB0 AD63 0B4E AD56 E574
    http://seifried.org/security/
    



    This archive was generated by hypermail 2b30 : Fri Feb 14 2003 - 08:41:46 PST