Virus jumping from linux drive to W7 drive

tehmoonrulz

Distinguished
Aug 27, 2009
33
0
18,530
Hello,

I am in the midst of getting ready for some computer and had a quick question regarding linux/windows drives.

I would like to boot into Linux by default and only boot into windows when I want to play a game. I will be using 2 drives for this and may use a third for just data. What are the feasible methods that a virus could go from my linux drive to the w7 drive?
 
In theory nothing is impossible. Somebody could write a virus that ran on a Linux machine and was capable of detecting Windows partitions and installing itself there. But, as far as we know, it has never been done. It's such a minority scenario that no-one would bother to do this (the same way that very few of the script kiddies bother with OS X - not because it's more secure than Windows but because it so much smaller a market, and hence target).

So, although it's not strictly impossible for a virus to transfer itself in this way, I would say that there is a greater chance that a meteorite will strike your PC and that you will win the lottery. Some things just aren't worth worrying about.
 

bmouring

Distinguished
May 6, 2006
1,215
0
19,360
Honestly, you'd more likely see someone write a virus that caused your drive to damage itself, although both are extremely unlikely.

These days, malware production/management is actually a surprisingly big business for the less-than-legitamate underground. In order for a virus to be successful, it has to have a good infection rate, allow discreet control over the infected machine, and continue to attempt spreading the virus while keeping the user unaware of the underlying issue.

There are hardly any Linux/BSD/etc. viruses out there for a few crucial reasons: they simply don't have the same market share as Windows and they are indeed designed with security in mind (and on by default, something Microsoft is awkwardly trying to move towards to this day).

Also note that a virus that originated on a Linux disk (surrounded by binary files written to the glibc/stdc++/etc. system API) would need to generate code that would hook into the completely-different Windows system-level API (win32, WFP, etc.) This alone would be tricky, require in-depth knowledge of vulnerabilities in both systems, and even then would only work in cases where the machine has dual-boot between Linux and Windows setup (we keep making that "effective target" sliver shrink more and more!)

So yes, like Pyroflea and ijack alluded to, it could be done, but there are many, many reasons it would never be done