你不知道的文件上传漏洞php代码分析

漏洞描述

开发中文件上传功能很常见，作为开发者，在完成功能的基础上我们一般也要做好安全防护。
文件处理一般包含两项功能，用户上传和展示文件，如上传头像。

文件上传攻击示例

upload.php

<?php
$uploaddir = 'uploads/'; 
$uploadfile = $uploaddir . basename($_files['userfile']['name']);
if (move_uploaded_file($_files['userfile']['tmp_name'], $uploadfile)){
 echo "file is valid, and was successfully uploaded.\n";
} 
else {
 echo "file uploading failed.\n";
}
?>

upload.html

<form name="upload" action="upload1.php" method="post" enctype="multipart/formdata">
select the file to upload: <input type="file" name="userfile">
<input type="submit" name="upload" value="upload">
</form>

上述代码未经过任何验证，恶意用户可以上传php文件，代码如下

<?php eval($_get['command']);?>

恶意用户可以通过访问 如http://server/uploads/shell.php?command=phpinfo(); 来执行远程命令

content-type验证

upload.php

<?php
if($_files['userfile']['type'] != "image/gif") {//获取http请求头信息中contenttype
 echo "sorry, we only allow uploading gif images";
 exit;
}
$uploaddir = 'uploads/';
$uploadfile = $uploaddir.basename($_files['userfile']['name']);
if (move_uploaded_file($_files['userfile']['tmp_name'], $uploadfile)){
 echo "file is valid, and was successfully uploaded.\n";
} else {
 echo "file uploading failed.\n";
}
?>

该方式是通过http请求头信息进行验证，可通过修改content-type ==> image/jpg绕过验证，可以通过脚本或burpsuite、fiddle修改
如下
content-disposition: form-data; name="userfile"; filename="shell.php"
content-type: image/gif

图片类型验证

该方法通过读取文件头中文件类型信息，获取文件类型

备注：如jpeg/jpg文件头标识为ffd8

upload.php

<?php
$imageinfo = getimagesize($_files['userfile']['tmp_name']);
if($imageinfo['mime'] != 'image/gif' && $imageinfo['mime'] != 'image/jpeg') {
 echo "sorry, we only accept gif and jpeg images\n";
 exit;
}
$uploaddir = 'uploads/';
$uploadfile = $uploaddir . basename($_files['userfile']['name']);
if (move_uploaded_file($_files['userfile']['tmp_name'], $uploadfile)){
 echo "file is valid, and was successfully uploaded.\n";
} else {
 echo "file uploading failed.\n";
}
?>

可以通过图片添加注释来绕过此验证。
如添加注释<?php phpinfo(); ?>,保存图片后将其扩展名改为php，则可成功上传。
上传成功后访问该文件则可看到如下显示

文件扩展名验证

 通过黑名单或白名单对文件扩展名进行过滤，如下代码

upload.php

<?php
$blacklist = array(".php", ".phtml", ".php3", ".php4");
foreach ($blacklist as $item) {
if(preg_match("/$item\$/i", $_files['userfile']['name'])) {
 echo "we do not allow uploading php files\n";
 exit;
}
}
$uploaddir = 'uploads/';
$uploadfile = $uploaddir . basename($_files['userfile']['name']);
if (move_uploaded_file($_files['userfile']['tmp_name'], $uploadfile)){
 echo "file is valid, and was successfully uploaded.\n";
} else {
 echo "file uploading failed.\n";
}
?>

当黑名单不全，构造特殊文件名可以绕过扩展名验证

直接访问上传的文件

将上传文件保存在非web root下其他文件夹下，可以防止用户通过路径直接访问到文件。
upload.php

<?php
$uploaddir = 'd:/uploads/';
$uploadfile = $uploaddir . basename($_files['userfile']['name']);
if (move_uploaded_file($_files['userfile']['tmp_name'], $uploadfile)) {
　　echo "file is valid, and was successfully uploaded.\n";
} else {
　　echo "file uploading failed.\n";
}
?>

用户不可以直接通过http://localhost/uploads/ 来访问文件，必须通过view.php来访问
view.php

<?php
$uploaddir = 'd:/uploads/';
$name = $_get['name'];
readfile($uploaddir.$name);
?>

查看文件代码未验证文件名，用户可以通过例如http://localhost/view.php?name=..//php/upload.php，查看指定的文件

解决漏洞示例

upload.php

<?php
require_once 'db.php';
$uploaddir = 'd:/uploads/'; 
$uploadfile = tempnam($uploaddir, "upload_");
if (move_uploaded_file($_files['userfile']['tmp_name'], $uploadfile)) {
　　$db =& db::connect("mysql://username:password@localhost/database");
　　if(pear::iserror($db)) {
　　　　unlink($uploadfile);
　　　　die "error connecting to the database";
　　}
　　$res = $db->query("insert into uploads set name=?, original_name=?,mime_type=?",
　　　　　　array(basename($uploadfile,basename($_files['userfile']['name']),$_files['userfile']['type']));
　　if(pear::iserror($res)) {
　　　　unlink($uploadfile);
　　　　die "error saving data to the database. the file was not uploaded";
　　}
　　$id = $db->getone('select last_insert_id() from uploads');
　　echo "file is valid, and was successfully uploaded. you can view it <a href=\"view.php?id=$id\">here</a>\n";
} else {
　　echo "file uploading failed.\n";
}
?>

view.php

<?php
require_once 'db.php';
$uploaddir = 'd:/uploads/';
$id = $_get['id'];
if(!is_numeric($id)) {
　　die("file id must be numeric");
}
$db =& db::connect("mysql://root@localhost/db");
if(pear::iserror($db)) {
　　die("error connecting to the database");
}
$file = $db->getrow('select name, mime_type from uploads where id=?',array($id), db_fetchmode_assoc);
if(pear::iserror($file)) {
　　die("error fetching data from the database");
}
if(is_null($file) || count($file)==0) {
　　die("file not found");
}
header("content-type: " . $file['mime_type']);
readfile($uploaddir.$file['name']);
?>

上述代码文件名随机更改，文件被存储在web root之外，用户通过id在数据库中查询文件名，读取文件，可以有效的阻止上述漏洞发生

总结

通过以上示例分析，可总结一下几点

1.文件名修改，不使用用户上传的文件名

2.用户不可以通过上传路径直接访问文件

3.文件查看采用数据库获取文件名，从而在相应文件服务器读取文件

4.文件上传限制文件大小，个人上传数量等

 以上就是本文的全部内容，希望对大家的学习有所帮助，也希望大家多多支持。