Re: Ubuntu 8.04 / 8.10

[Ray Wells <vk2tv@exemail.com.au>]

Dave Platt wrote:
> Ray Wells wrote:
>> Peter,
>>
>> This problem arose with fbb around kernel 2.6.20 (can't remember just 
>> which one now)
>>
>> The fix I apply is to compile fbb with  -fno-stack-protection added 
>> to its Makefile. I don't have a fix if you install fbb from a deb 
>> package.
>>
>> It seems that stack protection was included in kernels as a security 
>> plug for stack overflow.
>>
>> I sent quite a bit about this to the xfbb list around 18 months ago.
>
> The real question (to me at least) is whether the stack-smash abort
> is a false alarm, or whether it indicates a real problem in the
> FBB software.
>
> If there really is a way in which the FBB software is managing to
> smash its own stack (via e.g. a buffer that's too small, indexing
> out of bounds, etc.), then simply disabling the stack protection
> feature via recompilation is a bit like replacing a blown fuse
> with a larger one.  You may get away with it, or it may burn your
> house down :-(
>
> It'd probably be necessary to compile xfbb with -g and run it
> under GDB or a similar debugger, and investigate the state of the
> stack at the time of the abort, to figure out what's being
> over-written.
> -- 
> To unsubscribe from this list: send the line "unsubscribe linux-hams" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>
The logic I applied at the time was that since fbb had always been 
broken (but not previously detected) I wasn't going to break it any more 
by disabling stack protection. I've certainly not detected any problems 
with fbb since I started doing this. If somebody with c programming 
skills wants to fix the real problem, we will welcome you with open 
arms, but I don't have those skills and I do need to keep the bbs 
functioning. I have achieved that target.

Ray vk2tv