|
Anyway, if you are looking at this now, it is probably for Dr Chow's CS591 class. Considering the meager content so far, you can check out the Java blog I post to.
The project idea
The history of modern day worms dates back to originally trying to use a worm to map the internet. This is a dangerous propostition. The ** worm attempted to do this, but grew out of hand. However, the idea is not without merit, but mapping the internet is a neverending task that is probably best ignored for our purposesl What is more practical is to develop a "worm" that could determine bandwidth analysis. Let me elaborate.
Common bandwidth measurement, packet tracing tools, seek to "fool" routers and nodes along the path by shortining the TTL value in a UDP packet. Tracert, for example, does this by slowly increasing the TTL and tracking progress while it occurs. You can find more in depth analysis in my paper: on bandwidth analysis. For this to work, a node must respect the RFCs that correspond to TTL values... but <sarcasm>suprise!</sarcasm>
What I seek to investigate is the use of a worm (probably some type of buffer overflow) IN A CONTROLLED ENVIRONMENT to acheive the same results. I seek to find determine:
- Is such an exploit feasible?
- Is there a way to limit growth/transport?
- Could this be (analog to the last item) used in a controlled setting? An uncontrolled setting?
Of course, the question really boils down to, "Is it possisble to write a non-malicious and still useful exploit?
|
