Messages in this thread | | | Date | Fri, 21 Jul 2000 10:47:58 -0400 | From | john.cavan@sympatic ... | Subject | Re: TO HELL WITH IT THEN......(re: disk-destroyer.c) |
| |
Peter Svensson wrote: > The difference being that root _is_ allowed to crash the kernel. No, this > is more a question of providing a "cooked" interface or not. I generally > believe in cooked itnerfaces when they can abstract away differences in > lower levels. However, given the possible damage caused by an error I can > certainly understand if Linus chooses to include it.
Crashing the kernel and taking out the firmware are two entirely different things.
> The discussion has not been so much whether this patch is a good idea as > it has been about the claim that it is a security patch protecting from a > malicious root.
Not a whole lot is going to protect against a malicious root, but then again, protecting against a clueless root is a very good idea, the growth of Linux depends on that sort of basic protection.
From what I gather, the patch proposed by Andre essentially provides a layer of protection against general attack on the hardware. Correct me if I'm wrong, but this would defend against virii, trojans, and other remote exploit code. It may not defend against a root user with direct shell access, but a defense against remote attack is a good thing regardless. A sound security stance is to default *deny* and require an explicit allow from deliberate action by root. If someone needs to do firmware updates have them create a boot floppy with the necessary code and keep a less permissive kernel on the HD boot partition.
Not that my $0.02 counts much, but for what it's worth, I agree with Andre.
John
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |