At present, most of the ** and application systems have upload functions, including the upload of documents, **avatars, ** and other content. There is no problem with the file upload function itself, but when the system does not verify the content, suffix, and file type of the uploaded file, or the processing logic is not secure enough, malicious users can bypass the file type requirement to upload "arbitrary" files to the system or even an executable file backdoor, which will eventually lead to serious consequences.
The dangers of "any" file uploads
The harm of "arbitrary" file upload is closely related to the type of file uploaded by malicious users, which can be divided into the following categories:
Upload a malicious script language, and the web container of the system server will interpret and execute the corresponding malicious script.
Upload the flash policy file crossdomainxml, which controls the behavior of flash in that domain;
Upload virus and Trojan files to trick users or administrators into executing;
upload phishing or contain scripts for phishing and spoofing;
Upload a webshell to remotely control the system or cause the system to be disabled.
The protection policy for any file upload
The directory of the file upload is set to be unexecutable;
The whitelist method is used to check the file type by MIME type, suffix, file header, etc.
rewrite file names and file paths with random numbers;
Set the domain name of the file server separately;
Deploy security protection devices to detect malicious file uploads in real time.
Originated: security and confidentiality.