Wednesday, 27 August 2008

Pretty lights

More on DNS I know. May as well be another person beating a dead horse. But I give you pretty: http://www.doxpara.com/?p=1206
It is a video of the patched and not patched world wide. It intrigues me that there is a blinking light on the map of Australia about 3 hours north of Adelaide, I doubt it is Alice Springs, to south, maybe Coober Peadie if my geography serves.
Onto some more supposition by me (mainly in reply to Dan [the guy who discovered the Researched the DNS flaw] here);
I agree with what has been said, that we need more security on an inherintly in-secure network. But some (percieved) anonymity and some plain text is good, and what the internet is all about.
Could you imagine every site moving to https, for starters what is the point, who needs to read my blog through an encrypted channel? Really why, I don't really have any direct post functionality, and only a handful of readers, it is not like I am directing them to blindly do anything either.
Onto DNS, I was thinking the other day of another way to fix the issue. Deploy a port knocking technique on the reply based on the query, so that ports would have to be knocked in the correct order on the DNS server pre accepting back the lookup. Similar to the way a person gets into a safe, knowing the numbers isn't good enough you need to know the sequence. This would stop NAT being an issue as the DNS server can make the request out on all ports getting an auto map back on these ports. And would be more secure as the attacker would have to guess the right ports to knock on the way back, or read the request and then generate the reply and reply back, but if they can do that they are already in the middle and its game is over.
What do you think?
Peace out all, especially Dan, good job.

Thursday, 14 August 2008

DNS woes continue... sorta

So as I said, and the original discoverer Dan said, it was just a patch. Not a fix, not a be-all and end all solution. A temporary patch. We already know some nat devices break the patch's fix. But from the looks here and here it can be broken. The first link even details how, but there is a caveat. It is not easy, and a lot of bandwidth with low latency is required.
The first article explains how they did it over Gige in 10 hours. So most DNS servers that are doing resolves for clients, are probably not even on 20mbs of bandwidth, and latency 10+ times that of ethernet, not including the clients themselves causing some load. So you could say it would take 10+ times longer to do this over the internet, so 100hours. Someone will hopefully notice at around hour 20… But it isn't that simple, what if some baddie hits a server with a mere 100 clients... (Most botnets are 10 times this size). Chaos again. We need a better fix. I mentioned before some kind of signed DNS, I am the first to admit I have gaps in my knowledge as I have never heard of DNSSEC, now I that have listened to the Blackhat talk I have heard about it. I had a quick look at wikipedia and the official site and it is interesting. Of course windows servers only support it as a secondary, also the glaring-hole of non NSEC3 servers allowing enumeration of sites is just plain silly. Seriously just hash The users request domain "Not Found" and add it to the RFC, done.
I think it should include the option for encrypting replies, may as well, could be useful for higher secure organisations.
This is a very real and very now threat, there are at least two pieces of software out there to attack it, one being the very good, but very newbie friendly metasploit.
Well I am pretty much just re-iterating and expanding on my comments on darknet but there you go.