[NBLUG/talk] A new colo box (with interesting ideas)

ME dugan at passwall.com
Tue Apr 22 13:06:01 PDT 2003


>> Recent rumblings about how good AES is. Some claim to be on the edge of
>> breaking it. Not first hand knowledge, info from Security Course at the
>> uni.
>
> I am interested in your ideas and I wish to subscribe to your news
> letter!
>
> Seriously, what's the deal with this?

Well, we have not heard more about it, but in our cs340 course, our
professor let us know that there was an announcement that AES ewas broken,
but peer review showed that the original announcement was incorrect.
However, others stated that possible weaknesses were exposed, and then a
greater offert was put into trying to break AES. Now, we are left hanging
in class. Our professor was going to speak to us about AES, but has put it
off for another week or so to see what kind of results come from the
reviews being conducted. (He does not want to teach us something that is
going to be considered "broken" in the not so near future.

>> What about loop-encrypt with anoyther cipher?...  I head double-XOR with
>> same key is very good! ;-)    (joke)
>
> I hear rot-26 is better.

:-D

>> Also, consider vmware using a lo-encrypted volume. Actually have a
>> virtual
>> machine that is also running Linux on Linux. Can configure multiple IP
>> on
>> the main box and do pass-through for real IP on each instance of vmware.
>
> I don't see any reason to do that over UML.
> Why do you do that?

Familiarity. One of those thing, "going with what I know about" instead of
"trying the unknown." For me, me server is still colocated in socal, and
the risk of downtime in trying something new on a production machines
makes me feel a bit "off."

-ME




More information about the talk mailing list