Open Communication?

Users of AT&T or T-Mobile in the U.S. can be safely assured their calls will NOT be publicly wiretapped for AT LEAST a few months. Mainly because security researcher Karsten Nohl will be releasing an open-source, distributed computing project, designed to crack the encryption used by regular GSM-devices on" transmission (also known as "A5/1").

[The whole point of the venture is to spur tele-carriers to improve the security-flaws of their respective service protocols. The bug Nohl is talking about in the link at the end of this post, has been known for over 15 years! So.., covert cellular surveillance has probably been going on for years by now...]

"We're not creating a vulnerability, but publicizing a flaw that is alreay being exploited, very widely", Nohl said in a phone interview.

Utilizing 80 high-performance computers to distribute the workload, will use about 3 months to generate a key-table. But if about 160 people on the Internet were to offer computing-power, it would take half that time to complete. Which is Nohl's vision for the project, and the end-result of this major computational task, is to generate an encryption-key reference-table by use of which will grant anyone the ability to de-crypt cellular transmissions.

http://news.cnet.com/8301-27080_3-10316812-245.html?part=rss&subj=news&tag=2547-1_3-0-20

SSH, domains, crypto...

Finally, I've acquired a location to set up a desktop computer as a domain-gateway for my personal LAN/WAN/WiFi-connections! Sw33t! My earlier conclusion to carry an USB memory dongle was a good idea. I got so many notifications about brute-force attack-attempts running SSH-blockage rules on my firewall, so I decided to drop them all together.


Using private-key authentication with a passphrase, proved to be MUCH better. Nearly any excessive bandwidth-overhead, nor serious lagging (which was the main problems when running SSH firewall rules for brute-force attack tracking and blocking. It was basically too CPU intensive to be useful).

Regarding the key-authentication, both the generation of keys, and re-configuring of the SSH server to accept the keys was pretty straightforward. So now I'm enjoying fully secured SSHv2 sessions.

But again, I decided against utilizing some parts of my plans, like using 256bit AES cipher, instead of the intended 448bit Blowfish cipher, as it would be overkill with regards to the processing power available on the desktop machine I'm going to use ;P (Pentium-III 1st gen. 800Mhz). From what I know, Norwegian government-sections use 256bit AES, so it will more than suffice for my uses. Less is more.

I also set up a basic, free static hostname (with wildcards) to reach my public gateway-machine, and a secondary hostname for dynamic http-forwarding to the web-server hosted on the domain.

Good luck to the ones wanting to crack these streams! ;D