Linux sys_call_table变动检测

catalogue

0. 引言
1. 内核ko timer定时器，检测sys_call_table adress变动
2. 通过/dev/kmem获取IDT adress
3. 比较原始的系统调用地址和当前内核态中的系统调用地址发现是否有sys_call_table hook行为

 

0. 引言

内核rookit通常以系统调用为攻击目标,主要出于两个原因

1. 在内核态劫持系统调用能以较小的代价控制整个系统，不必修太多东西 
2. 应用层大多数函数是一个或多个系统调用不同形式的封装，更改系统调用意味着其上层所有的函数都会被欺骗

当前的系统调用地址保存在系统调用表中，位于操作系统为内核保留的内存空间(虚拟地址最高1GB)，系统调用入口地址的存放顺序同/usr/include/asm/unistd.h中的排列顺序，按系统调用号递增9

Relevant Link:

http://www.blackhat.com/presentations/bh-europe-09/Lineberry/BlackHat-Europe-2009-Lineberry-code-injection-via-dev-mem-slides.pdf

 

1. 内核ko timer定时器，检测sys_call_table adress变动

1. The module does a copy of the Syscall Table to save all syscalls pointers
2. After this first step, the module uses the kernel timer to check every X secondes the diff between the Syscall Table and the copy.
3. If a diff is found, the module creates a workqueue to execute the python script and restore the good syscall pointer. 

The python script is executed with root creds and the syscall number which is hooked, is passed as the first argument of script (sys.argv[1]).

0x1: hook_detection.c

/*
**  Copyright (C) 2013 - Jonathan Salwan - http://twitter.com/JonathanSalwan
** 
**  This program is free software: you can redistribute it and/or modify
**  it under the terms of the GNU General Public License as published by
**  the Free Software Foundation, either version 3 of the License, or
**  (at your option) any later version.
** 
**  This program is distributed in the hope that it will be useful,
**  but WITHOUT ANY WARRANTY; without even the implied warranty of
**  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
**  GNU General Public License for more details.
**
**  You should have received a copy of the GNU General Public License
**  along with this program.  If not, see <http://www.gnu.org/licenses/>.
**
**  For more information about this module, 
**  see : http://shell-storm.org/blog/Simple-Hook-detection-Linux-module/
**
*/

#include <asm/uaccess.h>
#include <linux/kernel.h>
#include <linux/kthread.h>
#include <linux/module.h>
#include <linux/slab.h>
#include <linux/syscalls.h>
#include <linux/timer.h>
#include <linux/workqueue.h>

#define PATH_PYTHON "/usr/bin/python2.7"
#define PATH_SCRIPT "/opt/scripts/hook_detected.py"

#define TIME_SLEEP 30000 /* in msec */

static struct timer_list timer_s;
static struct workqueue_struct *wq;
static unsigned int syscall_table_size;
static unsigned long *addr_syscall_table;
static unsigned long *dump_syscall_table;

static int exec_python_script(unsigned int sys_num)
{
  char s_num[32];
  char *argv[] = {PATH_PYTHON, PATH_SCRIPT, s_num, NULL};
  static char *envp[] = {"HOME=/", "TERM=linux", "PATH=/sbin:/bin:/usr/sbin:/usr/bin", NULL};
  struct subprocess_info *sub_info;

  sprintf(s_num, "%d", sys_num);
  sub_info = call_usermodehelper_setup(argv[0], argv, envp, GFP_ATOMIC);
  if (sub_info == NULL)
    return -ENOMEM;
  call_usermodehelper_exec(sub_info, UMH_WAIT_PROC);
  return 0;
}

static unsigned long *get_syscalls_table(void)
{
  unsigned long *start;

  /* hack :/ */
  for (start = (unsigned long *)0xc0000000; start < (unsigned long *)0xffffffff; start++)
    if (start[__NR_close] == (unsigned long)sys_close){
      return start;
    }
  return NULL;
}

static unsigned int get_size_syscalls_table(void)
{
  unsigned int size = 0;

  while (addr_syscall_table[size++]);
  return size * sizeof(unsigned long *);
}

static void check_diff_handler(struct work_struct *w)
{
  unsigned int sys_num = 0;

  while (addr_syscall_table[sys_num]){
    if (addr_syscall_table[sys_num] != dump_syscall_table[sys_num]){
      printk(KERN_INFO "hook_detection: Hook detected ! (syscall %d)\n", sys_num);
      write_cr0(read_cr0() & (~0x10000));
      addr_syscall_table[sys_num] = dump_syscall_table[sys_num];
      write_cr0(read_cr0() | 0x10000);
      exec_python_script(sys_num);
      printk(KERN_INFO "hook_detection: syscall %d is restored\n", sys_num);
    }
    sys_num++;
  }
}
static DECLARE_DELAYED_WORK(check_diff, check_diff_handler);

static void timer_handler(unsigned long data)
{
  unsigned long onesec;

  onesec = msecs_to_jiffies(1000);
  queue_delayed_work(wq, &check_diff, onesec);
  if (mod_timer(&timer_s, jiffies + msecs_to_jiffies(TIME_SLEEP)))
    printk(KERN_INFO "hook_detection: Failed to set timer\n");
}

static int __init hook_detection_init(void)
{
  addr_syscall_table = get_syscalls_table();
  if (!addr_syscall_table){
    printk(KERN_INFO "hook_detection: Failed - Address of syscalls table not found\n");
    return -ECANCELED;
  }

  syscall_table_size = get_size_syscalls_table();
  dump_syscall_table = kmalloc(syscall_table_size, GFP_KERNEL);
  if (!dump_syscall_table){
    printk(KERN_INFO "hook_detection: Failed - Not enough memory\n");
    return -ENOMEM;
  }
  memcpy(dump_syscall_table, addr_syscall_table, syscall_table_size);

  wq = create_singlethread_workqueue("hook_detection_wq");

  setup_timer(&timer_s, timer_handler, 0);
  if (mod_timer(&timer_s, jiffies + msecs_to_jiffies(TIME_SLEEP))){
    printk(KERN_INFO "hook_detection: Failed to set timer\n");
    return -ECANCELED;
  }

  printk(KERN_INFO "hook_detection: Init OK\n");
  return 0;
}

static void __exit hook_detection_exit(void)
{
  if (wq)
    destroy_workqueue(wq);
  kfree(dump_syscall_table);
  del_timer(&timer_s);
  printk(KERN_INFO "hook_detection: Exit\n");
}

module_init(hook_detection_init);
module_exit(hook_detection_exit);

MODULE_AUTHOR("Jonathan Salwan");
MODULE_DESCRIPTION("Hook Detection");
MODULE_LICENSE("GPL");

Relevant Link:

http://shell-storm.org/blog/Simple-Hook-detection-Linux-module/hook_detection.c
http://shell-storm.org/blog/Simple-Hook-detection-Linux-module/

 

2. 通过/dev/kmem获取IDT adress

核心逻辑是通过idt获取80中断的地址，获取了sys_call_table之后，进行一次拷贝，之后就可以进行diff对比

#include < stdio.h >
#include < sys/types.h >
#include < fcntl.h >
#include < stdlib.h >

int kfd;

 struct
 {
      unsigned short limit;
      unsigned int base;
 } __attribute__ ((packed)) idtr;

 struct
 {
      unsigned short off1;
      unsigned short sel;
      unsigned char none, flags;
       unsigned short off2;
 } __attribute__ ((packed)) idt;


 int readkmem (unsigned char *mem, 
              unsigned off, 
              int bytes)
 {
      if (lseek64 (kfd, (unsigned long long) off, 
                                    SEEK_SET) != off)
      {
                 return -1;
      }

      if (read (kfd, mem, bytes) != bytes) 
     {
             return -1;
     }

 }
 
 int main (void)
 {
      unsigned long sct_off;
      unsigned long sct;
       unsigned char *p, code[255];
      int i;


/* request IDT and fill struct */



    asm ("sidt %0":"=m" (idtr));

    if ((kfd = open ("/dev/kmem", O_RDONLY)) == -1)
    {
                perror("open");
            exit(-1);
    }
    
    if (readkmem ((unsigned char *)&idt, 
               idtr.base + 8 * 0x80, sizeof (idt)) == -1)
    {
            printf("Failed to read from /dev/kmem\n");
            exit(-1);
    }
    
    sct_off = (idt.off2 < < 16) | idt.off1;

    if (readkmem (code, sct_off, 0x100) == -1)
    {
            printf("Failed to read from /dev/kmem\n");
            exit(-1);
    }

/* find the code sequence that calls SCT */


 sct = 0;
    for (i = 0; i < 255; i++)
    {
            if (code[i] == 0xff && code[i+1] == 0x14 && 
                                        code[i+2] == 0x85)
                    sct = code[i+3] + (code[i+4] < < 8) + 
                        (code[i+5] < < 16) + (code[i+6] < < 24);
    }
    if (sct)
            printf ("sys_call_table: 0x%x\n", sct);
    close (kfd);
 }

Relevant Link:

http://www.rootkitanalytics.com/kernelland/IDT-dev-kmem-method.php
http://www.phpweblog.net/GaRY/archive/2007/06/17/get_sys_call_table_address.html

 

3. 比较原始的系统调用地址和当前内核态中的系统调用地址发现是否有sys_call_table hook行为 

原始的系统调用地址在内核编译阶段被指定，不会更改，通过比较原始的系统调用地址和当前内核态中的系统调用地址我们就可以发现系统调用有没有被更改。原始的系统调用地址在编译阶段被写入两个文件

1. System.map: 该文件包含所有的符号地址，系统调用也包含在内
2. vmlinux-2.4.x: 系统初始化时首先被读入内存的内核映像文件

vmlinux-2.4.x文件通常以压缩的格式存放在/boot目录下，所以在比较之前必须解压这个文件，另一个问题是: 我们的比较的前提是假设system.map及vmlinuz image都没有被入侵者更改，所以更安全的做法是在系统干净时已经创建这两个文件的可信任的拷贝，并创建文件的md5 hash

在大多数被装载内核后门情况中，内核在系统初始化之后才被更改，更改发生在加载了rootkit的module或者被植入直接读写/dev/kmem的on-the-fly kernel patch之后。而通常情况下rootkit并不更改vmlinuz和system.map这两个文件，所以打印这两个文件中的符号地址就可以知道系统原始的系统调用地址，系统当前运行中的系统调用地址(可能被更改)可以同过/proc下的kcore文件得到，比较两者就知道结果

0x1: 获取原始内核系统调用函数地址

/boot/System.map-3.10.0-327.el7.x86_64

0x2: 获取当前内核系统调用地址

cat /proc/kallsyms 

0x3: 对比区别

从里面枚举sys_call_table的function point地址
/boot/System.map-3.10.0-327.el7.x86_64

和
cat /proc/kallsyms | grep sys_fork
进行diff对比 

cat System.map-4.4.0-53-generic | grep sys_fork

这里遇到一个问题，ubuntu对sys_socket相关的系统调用会进行内核地址重定位，因此需要对检测结果进行一个误报过滤，看是否所有的函数的gap都相同，如果相同，则说明是系统自己的function address relocation行为

def hex2dec(string_num):
    return str(int(string_num.upper(), 16))

def check_rookit(diff_func_table):
    last_func_address_gap = 0
    cur_func_address_gap = 0
    for item in diff_func_table:
        cur_func_address_gap = abs(int(hex2dec(item['cur_funcaddress'])) - int(hex2dec(item['ori_funcaddress'])))
        if last_func_address_gap != 0 and cur_func_address_gap != last_func_address_gap:
            return True
        else:
            last_func_address_gap = cur_func_address_gap
    return False

Relevant Link:

http://www.xfocus.net/articles/200411/754.html
http://www.magicsite.cn/blog/Linux-Unix/Linux/Linux41576.html
http://blog.csdn.net/tommy_wxie/article/details/8039695

Copyright (c) 2017 LittleHann All rights reserved