ECC prevents parity error crashes through error correction. So, if you have a 
bad RAM and your IOS device has no ECC, it will crash and the error handling 
code will print a sig 20, sig 22, PMPE, etc. depending on where the parity 
errors were found. Not a SegV. 

Furthermore, IOS has processes that constantly validate the integrity of its 
memory data structures eg checkheaps.

Quite a few people read these threads for support. Therefore, It will be 
irresponsible to encourage them to schedule maint windows and start ripping out 
their DRAMs for SegV exceptions that are *clearly* caused by software bugs in 
IOS.

I understand some folks enjoy commenting, but it should always be to add value 
rather than throwing out unfounded speculations without factual data to back it 
up.

/eninja

ex-Technical Leader - OS & big fat (and skinny) routers.




On Nov 12, 2010, at 3:17 AM, Gert Doering <[email protected]> wrote:

> Hi,
> 
> On Fri, Nov 12, 2010 at 12:55:23AM -0800, Eninja wrote:
>> Either way, I'd hate to bust your bubble and this is the last time I'll say 
>> this, SegV exceptions in IOS are always caused by software bugs. 
> 
> You don't seem to understand anything about hardware failure modes.
> 
> If you have a bad RAM that will change a pointer value for you *without*
> triggering a RAM parity error (because it's on a platform that has no
> ECC/parity, like 'early 7200 NPEs'), and the software uses that pointer,
> and accesses a bad memory location *because the RAM corrupted the pointer*,
> how can the resulting SegV be "always a software bug"?
> 
> gert
> -- 
> USENET is *not* the non-clickable part of WWW!
>                                                          //www.muc.de/~gert/
> Gert Doering - Munich, Germany                             [email protected]
> fax: +49-89-35655025                        [email protected]


_______________________________________________
cisco-nsp mailing list  [email protected]
https://puck.nether.net/mailman/listinfo/cisco-nsp
archive at http://puck.nether.net/pipermail/cisco-nsp/

Reply via email to