Main Page...About Me, Jaxor, Notes...

Welcome to My Pages!

I am 26-year guy, currently living in Turku, Finland. I have been involved with software, computers and internet for many years, although I don’t do programming nor work in the IT-industry. I made these pages from a passion on computer security and privacy. I hope you and your friends and relatives have found them usefull. I personally have implemented all that is presented here to secure my computer and have gone even further. Please send any me comments or questions using email. If at all possible, use my PGPkey while communicating.

My Notes
- Just coincidence? After the cases "Kavkazcenter" and "Our parlaments computer insecurity" (see below) strange "happening" have been going on for me and few people that I have discussed these issues with. For example, one internet-address in north Finland (it is ofcourse coincidence that in that address are north Finland regions security police and central criminal police hq:s) has continously downloaded my pages, and after I discussed this issue over phone to "one person", shortly after that same address has began downloading he's homepages too. This is ofcourse just coincidence, I mean, nobody could not possible be listening to (one of mine) my cellurar phone conversations right? Also, its just coincidence that this same address is downloading mine and my friends pages every now and then, I mean, nobody couldnt be interested to know what Im publishing in my pages every now and then right? My "civilian" GSM has also cutted off the conversations with couple "friends of mine" to whom I have discussed about these issues. This is also, ofcourse, just coincidence, even I have never had such problems with other people and with other cellurar phones and I dont have such problems with my anonymous GSM:s at all. Nobody would be trying to eavesdrop or manipulate my GSM connection right? And its also coincidence that my GSM:s signal, that normally where very weak when I was calling from my home, suddently "jumped off" to maximum when I called couple "friends of mine" about these issues. I mean, nobody would be trying to conduct "false base station" attack against my GSM conversation right? Also, my GSM phone has reported "place sim-card to the phone" errors and its ofcourse pure coincidence that my russian friend has had similiar problems at the very same time right? It is also pure coincidence, ofcourse, that two of my friend who have know about "these issues" have received strange phone calls and/or SMS messages from ex-Soviet territory phone numbers about the same time. I mean, these issues are ofcourse totally unrelated to all other issues right? I mean, its nice to know that all of this is just coincidence, so that I dont have to laught about the incompetence and amateur-style espionage attempts by "certain people or organizations". ;)

- The computers used by members of parlament and their assistants in here Finland have severe security holes. These laptop computers dont have firewalls, file encryption and wiping tools, automatic update is not turned on, operating system (WindowsXP) is on its default settings for most, computers only support 802.11b WLAN which is insecure, etc. etc. As a bonus, they use TeliaSonera GSM:s which are totally insecure because they use COMP-128-1 and A5/1 for security. I contacted them months ago but they havent bothered to answer me, nor to reporters I have contacted later. Oh dear... Update! Now they are trying to sue me for publishing their computer security holes, read more here. Also, I suggest that your read comments and bashing to he's email from here and here by the other people who also write to the full disclosure -list. I bet my response to this our parlaments "security guru" is worth reading too. And btw. the person from our parlament it-security department doesnt even have latest updates to he's own email program, as you can read from here... Uuuhh...what kind of security expert would use old version programs that lack updates and patches? The situation is much worse than I thought, these people really dont know a drek about computer security!

- The fundamental insecurity of bluetooth and cellurar phones/PDA:s proven again. In the last few months, several very dangerous security holes have been found in bluetooth devices like cellurar phones, PDA:s etc. For example, this security hole/proof of concept allows attacker to turn cellurar phones to wiretaps to eavesdrop everything near them and calls made by them and downloading all messages/addresses/etc. from the cellurar phone over a mile away! Here is proof-of-concept-video about the bluetooth hack. Latest information about the issue in here. The bluetooth is enabled by default in almost all models and even turning it completely off doesnt always help to protect you! My solution: KISS. Keep It Simple Stupid. Dont get any hardware that even has support for these kinds "latest features" that you dont really need and just open more doors to hackers. Also, dont forget about this cellurar phone virus/worm either...

- Funny thing happened to me a while back. Kavkazcenter www-pages where pulled down from the internet here Finland and I got really mad about it. I told publicly on Sunday night that Im going to host that site in my own www-pages, since Im not affraid of Suojelupoliisi (Finnish security police) or Russians who want the site off the net. Well, on Monday morning I couldnt get into my pages nor into my email. But when I got to my workplace, I could access them both without any problems. Strange. I went back to home, check settings on my computer, resetted my ADSL modem, etc. etc. but I still couldnt get access to those pages. I notified to my ISP about the issue but they couldnt figure out what was the problem. This went on several days, until suddently I was again able to get to my pages and my email from my home computer too. Now, call me paranoid, but I dont believe this is just "bad luck". I personally think that FAPSI or GRU where behind this "information blockage" of mine, to make sure that Kavkazcenter would not be reopened too soon. Ofcourse, its possible that some Finnish element was involved here, like for example Viestintävirasto (center of communnication service) or Suojelupoliisi (Finnish security police). I really dont know. But I dont believe it was just "bad luck".

- There has been lot of discussion about hardware trojans. Trojans, that are installed onto the hardware of computers etc. or ruin the hardware on your computer/etc. Some cases have been published and the theory is correct: Not only the software on your computer can get infected or damaged from a malware, but also your hardware. BIOS of many components are very easy targets, CPU:s can be reprogrammed too. The old Chernobyl virus also affects hardware component. This gets scary. If you cant even trust the hardware you are using, the game is over. Permanently.

- TETRA might be backdoored for NSA sayes EU. TETRA is used for example in Finland for police, border guard etc. as "VIRVE" to increase security and flexibility over the old radio systems. However, the security of TETRA is not that good at all. Not only there is severe danger of backdoors for the NSA, but the encryption used in TETRA is poor. Its not open source crypto that is developed with time and effort and openly discussed and analyzed by professionals. Its the same kind of snake-oil crypto from the same people (ETSI) that have created other insecure encryption products like the ones used in cellular phones. Not only the entire algorithms might have severe weaknesses, they only give 64bit workload to the hacker at best. The encryption algoritms used in TETRA (TEA1-4) can be "replaced" if wanted to (in end-to-end communications) with other algoritms, but this isnt usually done. Actually, even the TEA algoritms are not usually used in TETRA and data is usually send in plaintext. Why dont they just get it? Why do they spend tens of millions of euros in Finland for example to build secure comm system for goverment and then use insecure ciphers on it or dont use encryption at all? I dont know whether to laught or to cry.

- Cellurar phones security. In short, there isnt much. For example, TeliaSonera in Finland uses COMP-128-1 for security, which means that its cellular phones can be cloned and its secret key (Ki) retrieved over-the-air. This means that calls made with that user can be eavesdropped pretty easily. Also, TeliaSonera uses A5/1 for encryption, that can be broken in less than a second using a laptop computer. Still, they have guts to say publically that their cellurar phone traffic cannot be eavesdropped in any "reasonable way without massive amounts of supercomputers". Some service providers (like Radiolinja here Finland) use COMP-128-2 or even COMP-128-3, which provide much greater security and, as far as it is known today, cannot be "broken" in any available means. To counter the A5/1 poor encryption, some (like Radiolinja in Finland) are also starting to use A5/3 (Kasumi) for encryption (Kasumi will be used in UMTS), which provides very good security if combined with COMP-128-2 or COMP-128-3. But, you should remember, that by using false base stations, you GSM phone can be forced to turn crypto completely off, and most phones dont even alert you when crypto is turned off... :( Some links related to this issues here and here.

- Oh, and you think SMS messages sent using GSM are secure? Guess again, they can be spoofed aswell, even without breaking the crypto on your GSM or anything like that. The senders number can be manipulated so that not even the telecommunication authorities could tell where did the message actually originate. This is pretty scary, since this means that by using this spoof, you can identify yourself as someone else using SMS messages, pay shoppings with that, or subscribe to services etc. by simply spoofin the sending number on the message. For this reason, SMS messages should not be used for *anything* but unfortunally they are, and there is no way users of GSM can do anything about it. Oh dear... Remember that you should not trust SMS messages any more than you trust email: Their content and sender information can easily be spoofed!

- VoIP vulnerable to spoofing is reality. VoIP is becoming even more popular way communicating as an alternative to regular telephones or cellurar phones. However, identity theft and spoofing is also not only possible, but very real threat according to this Securityfocus article. This is what happens, when, once again, new techs are produced and implemented without anyone thinkin about security.

- AES. Is it really that secure? Well, atleast NSA didnt design it, but they did choose it in practise. Now, this might sound paranoid, but...do you really think they would choose an algorithm to be used in almost everything you can possible imagine, if they couldnt break it? Would they choose an algorithm that, once people and companies start to use it, would effectively blind NSA and Echelon? That doesnt sound very sensible thing to do. Remember that they designed DES with short key lenght so they could easily crack it on demand. If NSA is the best codecracker there is, they might have found some ways to break algorithmX and bet, that nobody else has and will not discover it in the near future. Paranoid? Or just being realistic here? Rijndael was not chosen as AES because it offered the best security as they pointed out, but because it was fast and could be easily implemented. Many cryptographers agree that Twofish or Serpent where far superior in terms of security than Rijndael. After what I have read about it, I agree. I dont trust NSA. I have no reason to trust them. Neither do you. UPDATE! New attack methods might break AES! 128bit AES only 100bit secure? Even worse? Read more from here. Also, concider the fact that there is only 3-rounds of security margin to the 128bit and 5-rounds security margin to 256bit AES, since best known attacks currently breaks 7/10 and 9/14 rounds of AES. For example, the best attack only breaks 8/16 rounds of Twofish and 10/32 rounds of Serpent ciphers.

- MD5, SHA and SHA-1 hash functions are apparently compromised. SHA-256/512 might also be vulnerable. Since hash functions are commonly used in all kinds of encryption tools, this attack might be very dangerous, or then maybe nothing practical. We have to wait and see. And be affraid. Very affraid. Some links related to this issue here and here for example. UPDATE! New attacks breaks SHA-1 for real! Read more about this new attack from here. UPDATE! Digital signatures can be forged easily using laptop computer! Read more here.

- There are currently SEVERAL unpatched, well-documented and reported security vulnerabilities in Internet Explorer. Several of these vulnerabilities may put your computer in great danger even if you have secured your Internet Explorer settings and have installed all the patches to your Windows/Internet Explorer! This clearly shows that Microsoft does not care about their customers safety, please read the list here. Please notice that this list is not updated yet, some of these vulnerabilities are already patched, but most of them are not. Also remember, that it took Microsoft half year to patch critical, remote-exploitable hole in their Windows just couple months ago. Not good.

- SSH. Why do people think it offers that good security by default? Its like "Hey, they have / I use SSH so this connection is secure!". So far I havent seen a many sites or places that offers SSH that was somehow protected against man-in-the-middle-attacks. Now this is pretty serious if you ask me. People implementing and using SSH dont seem to understand, that sure they might be offering 2048bit RSA key to someone connecting to their server...but does the recipient get THEIR key or someone elses (hackers) key? The client cant know is the key he is receiving the "real" key from the "real" server or not. Its just a key. Anyone can name the key as "www.markusjansson.net" so that doesnt proof anything. In SSL, they use certificate authorities (CA) to avoid this. CA digitally signes the keys and therefore the recipient can be sure that he has the right key by checking the signature by using the publickey of the CA (that he has on hes computer by default since it cames with installation packages of the browsers). CA:s dont work well with SSH, but there is one thing SSH servers could do. They could simply post the fingerprint of the key they use to www-page or give it to their customers so that anyone using SSH could check that they have gotten the right key. How can this be so difficult to do and to understand? Who would be stupid enought to try to break even 512bit RSA key while all he needs to do is to commit man-in-the-middle-attack against the particular SSH server?!? Beats me. Still most security "professionals" dont get it. They just dont get it. Sure they say, SSH saves the publickey used and warns if it changes on next connection attempt...but how can you ever get the first key and therefore the rest of the connections for sure? You cant. Or how can you know that the server hasnt just created new keys? You cant. Unless you can receive the publickey or the fingerprint of the publickey used in some secure manner. Come on folks, its just 128 or 160bits of data! A small piece of paper will be just fine for it or some tiny corner in the www-page. Cant be that hard to get it!

- Breakthrought in cryptanalysis. Funny, that even after the Bernsteins paper still many people dont realize what has happened. Basicly speaking all DSS/DH/RSA keys of 1024bits are totally insecure and only 4000bits give some marginal security. You need atleast 12000bit asymmetric key for some "real" security today. RSA commented and claimed that Bernsteins paper was not a "breakthrought", but I doubt. Perhaps its not a breakthrought in practise but it certantly points out that you need larger security margin for asymmetrical keys. Now this is not the thing I personally am worried about that much, I mean, sure SSL and other similiar are easily broken by NSA, CIA, GRU and such. What I am worried about is the fact that intelligence agencies and codebreakers have known this for...how long? 10years? 20years? Again, remember the fact that all crypto that the civilian academy or world armies have had, has always been broken in relatively short time. Viganere, ADFGVX, Enigma, etc. ...all of them where broken long before civilian academy and the people using them had any idea about it! Also remember that NSA was aware of differental cryptoanalysis in 1970:s when designing DES, while civilian academy discovered it in 1990:s! As such advances in cryptanalysis can happen, what algorithms and keysizes are REALLY secure today? 16000bit RSA? 64000bit RSA? ECC systems? Or should we all just stick to One-Time-Pads? ;)

- Mozilla, Netscape & Konqueror. People have been asking why I dont use them since they are much more secure than IE. Well, perhaps they are, perhaps not, but thats not the point since Im using Opera right now, which is, btw. propably even more secure than Mozilla and Netscape. I contacted Netscape couple years ago for one serious bug/annoyance in their browser: It doesnt show the RSA keysize at all! Same goes with Mozilla and Konqueror! Now this is pretty serious if you ask me. You can surf to any https site and see that the (symmetric) encryption is 128bits. Good, but what is the asymmetric encryption there? 384? 512? 1024? 2048bit RSA? You cant tell unless you are surfing with Opera or Internet Explorer or manually count the bytes and calculate it. Now, using 128bit symmetric and 512bit asymmetric keys is like being a total idiot and lacking even the basic knowledge of cryptography since 512bits RSA is only as secure as 50bits of symmetric encryption! So you might as well use 56bit DES for symmetric cipher if you use 512bit RSA. If you want to see one example about this ignorance, check out Merita banks (largest bank in Scandinavia) internet bank in here. Again, if you are using Netscape, Mozilla or Konqueror, you wont see anything being wrong... Oh well. (BTW. I also contacted Merita bank over two years ago and only now they have bothered to but strong crypto on their main site and keep this weak crypto in their alternative address.) Solutions? Dont use Mozilla, Netscape or Konqueror on https connections since you have no practical way of knowing how (in)secure that communication is since you dont know the RSA strenght. And keep bombing their feedback/bug forms about this issue, maybe after two more years they will bother to fix it. If they know even basics of cryptography and SSL/TLS that is.

- IE and SSL. Now you might think that IE is secure and good when surfing to https sites since it shows RSA too...wrong. Microsoft STILL hasnt bothered to fix this serious security vulnerability that allows man-in-the-middle-attacks to be conducted against https connections! Way to go Micro$oft, you have effectively ruined the purpose of SSL/TLS! SSL/TLS is not secure when using Internet Explorer unless you manually check and verify each and every sertificate you get on https connection.

Jaxor encryption tool
Here is a nice little cryptoprogram I designed (the beta version is coming...some day...). The program is called Jaxor. It is based SHA-256/384/512 hash functions and XOR. In short, SHA-256/384/512 is used to create a pseudorandom key from the users passphrase. The pseudorandom key is then XOR:ed to the first 256/384/512bits of plaintext. Then a new pseudorandom key is created using the just created block of ciphertext and passphrase by hashing them. Then the pseudorandom key is XOR:ed to next 256/384/512bits of plaintext. Then new pseudorandom key is created using the just created block of ciphertext and passphrase by hashing them. And again, the pseudorandom key is XOR:ed to next 256/384/512bits of plaintext...And so on and so on until the entire message is encrypted.

Basicly what we are talking here is using the one-way hash function to create pseudorandom keys which are then XOR:ed to datastream. The "master key" (passphrase) is salted using the just created ciphertext to make new keys. You can input ASCII text (cleartext) from clipboard or type it in the window. You can also decide whether you want the output (ciphertext) to be placed to window or directly to clipboard. Then all you need to do is to press "Crypt" and you will be prompt for passphrase. The decryption works the same way: you input the (cipher)text and press "Crypt" and you will be prompt for passphrase.

This method of encryption is pretty strong. It is as strong as you selected passphrase is or as strong as the SHA-256/384/512 hash function is. There are no known shortcuts to break this type of crypto. "Known plaintext" and "Chosen plaintext" attacks are effective to determine the key used for encryption, but they only discover the proper 256/384/512bits used for encrypt the specific segment, they do not compromise the actual encryption key which is the selected passphrase.

However, you must be very carefull not to use same passphrase on two messages! Since the hash of the passphrase is used to create encryption key, using the same passphrase on two encryptions will result you having identical encryption key for the FIRST 256/384/512bits of plaintext! After that, the key is ofcourse not the same anymore (if you dont use the same plain english message on both cases), since the passphrase is salted and new key is created. IF YOU WANT to use use same passphrase to all messages (like you are encrypting to your friend and have agreed upon one passphrase in secure manner), you can do it easily if you know how. The trick is as follows:
- Create a pseudorandom set of marks/characters (just press keys in somewhat random way, more you get the better it is ofcourse, perhaps 64marks) to your notepad etc.
- Type your message.
- When a passphrase is asked by Jaxor, type the agreed passphrase AND add (cut/paste) that pseudorandom set of marks/characters you just created to it. For example, if your passphrase was (without quotas) "This is clever passphrase" and pseudorandom marks/characters you typed where (without quotas) "hwei239vcmawerkrthafjhgfjqweq3f"....then you type at the passphrase box (without quotas) "This is clever passphrasehwei239vcmawerkrthafjhgfjqweq3f".
- Then you get the encrypted message and you can put it in email, whatever...and somewhere in that email/message you say (in plain english) something like "Hi Mike! I have used (without quotas) "hwei239vcmawerkrthafjhgfjqweq3f" for salting the passphrase. The encrypted message is below."
- Send the message as usual.
- When the recipient gets the message he/she will be able to decrypt it easily. He/she knows that the passphrase you always use is (without quotas) "This is clever passphrase" and since you tell him to salt it with (without quotas) "hwei239vcmawerkrthafjhgfjqweq3f".....he types in the passphrase box (without quotas) "This is clever passphrasehwei239vcmawerkrthafjhgfjqweq3f". VOILA!

=> Please notice again notice that it is safe to tell the salt in plain english. Attacker trying to decrypt your messages does not gain anything from knowing the salt used. Therefore salt can and should be sent in plain english...if you encrypt the salt used for encrypting how in the earth can the recipient ever decrypt the message!?! He cant. So just remember "Its safe to let attacker know the salt. He gains absolutely nothing from it because the salt and the passphrase are hashed using one-way hash function (SHA-256/384/512)."

Im currently working with version that has PRNG that would create the salt for each message and clean it out of the plaintext without any user interaction. However, such program would make the .exe file even larger, so for now on... :)