Buffer Overflows

A buffer is a holding area for data.To speed processing, many software programs use a memory buffer to store changes to data, then the information in the buffer is copied to the disk. When more information is put into the buffer than it is able to handle, a buffer overflow occurs. Overflows can be caused deliberately by hackers and then exploited to run malicious code.

There are two types of overflows: stack and heap.The stack and the heap are two areas of the memory structure that are allocated when a program is run. Function calls are stored in the stack, and dynamically allocated variables are stored in the heap. A particular amount of memory is allocated to the buffer. Static variable storage (variables defined within a function) is referred to as stack, because they are actually stored on the stack in memory. Heap data is the memory that is dynamically allocated at runtime, such as by C's malloc() function.This data is not actually stored on the stack, but somewhere amidst a giant "heap" of temporary, disposable memory used specifically for this purpose. Actually exploiting a heap buffer overflow is a lot more involved, because there are no convenient frame pointers (as are on the stack) to overwrite.

Attackers can use buffer overflows in the heap to overwrite a password, a filename, or other data. If the filename is overwritten, a different file will be opened. If this is an executable file, code will be run that was not intended to be run. On UNIX systems, the substituted program code is usually the command interpreter, which allows the attacker to execute commands with the privileges of the process's owner, which (if the setuid bit is set and the program has ownership of the root) could result in the attacker having Superuser privileges. On Windows systems, the overflow code could be sent using an HTTP requests to download malicious code of the attacker's choice. In either case, under the right circumstances, the result could be devastating.

Buffer overflows are based on the way the C or C++ programming languages work. Many function calls do not check to ensure that the buffer will be big enough to hold the data copied to it. Programmers can use calls that do this check to prevent overflows, but many do not.

Creating a buffer overflow attack requires that the hacker understand assembly language as well as technical details about the OS to be able to write the replacement code to the stack. However, the code for these attacks is often published so that others, who have less technical knowledge, can use it. Some types of firewalls, called stateful inspection firewalls, allow buffer overflow attacks through, whereas application gateways (if properly configured) can filter out most overflow attacks.

Buffer overflows constitute one of the top flaws for exploitation on the Internet today. A buffer overflow occurs when a particular operation/function writes more data into a variable (which is actually just a place in memory) than the variable was designed to hold.The result is that the data starts overwriting other memory locations without the computer knowing those locations have been tampered with.To make matters worse, most hardware architectures (such as Intel and Sparc) use the stack (a place in memory for variable storage) to store function return addresses.Thus, the problem is that a buffer overflow will overwrite these return addresses, and the computer-not knowing any better-will still attempt to use them. If the attacker is skilled enough to precisely control what values are used to overwrite the return pointers, the attacker can control the computer's next operation(s).