Fucking RMT... ChatMon update REQUIRED soon

[Aikar]

So apparently the RMT are now accessing chatmons spamfilter.

I just added "Xmas Bonus" to the filter list 1hr ago, and they are already changing the text on that.

So within the next few days I'm going to have to encrypt the spamfilter file, and that means all the older chatmons arn't going to be able to read it.


Geezus this is getting frustrating.

[Khajit]

EXTERMINATE!
http://www.clivebanks.co.uk/Doctor%2.../NewDalek1.jpg

[Spekkio]

That would actually be a rather cute application of a hash function. Hash each word and if it exists in the hash table, reject the tell. No way to know what words correspond to each hash beyond a brute force dictionary attack.

[TheMysteriousX]

Encrypting it won't do anything if they're determined; they'll just pull the encryption keys out of the plugin once it's in memory.

IDK how you've structured the data store for it, but could you convert it to use a hash table easily?
Salt the phrases differently with each release so that if they want to know what is filtered, they have to brute force the table every time it is updated, which will take significantly longer than reading the encryption key.

[psubond]

thx for the heads up and i apprecate all the work you guys put into this stuff.

[Spekkio]

Encrypting it won't do anything if they're determined; they'll just pull the encryption keys out of the plugin once it's in memory.

IDK how you've structured the data store for it, but could you convert it to use a hash table easily?
Salt the phrases differently with each release so that if they want to know what is filtered, they have to brute force the table every time it is updated, which will take significantly longer than reading the encryption key.

yeah, that's exactly what i was suggesting. i kinda assumed the random salt change every update was rather a given to make updates even more time consuming. the table could even be updated much more regularly with random strings added as pad to further obfuscate the contents and require more time be applied on their end to try to circumvent the system.

[Aselin]

Isn't it scary that the RMT spammers know how to do this?

A lot of us have seen the news specials and articles about gold farmers in China, and the images about them show them to be poorly educated, living in cramped and poor conditions. They give the appearance they have enough computer knowledge to operate a computer and an MMO, but that's it.

It is highly possible these RMT companies have hired hackers or someone with that level of knowledge to embed exploits in Javascript and Flash, and now to get past the spam filters and Chatmon itself. Then, whatever they find, they pass it along to the ones spamming us.

[rog]

Isn't it scary that the RMT spammers know how to do this?

A lot of us have seen the news specials and articles about gold farmers in China, and the images about them show them to be poorly educated, living in cramped and poor conditions. They give the appearance they have enough computer knowledge to operate a computer and an MMO, but that's it.

It is highly possible these RMT companies have hired hackers or someone with that level of knowledge to embed exploits in Javascript and Flash, and now to get past the spam filters and Chatmon itself. Then, whatever they find, they pass it along to the ones spamming us.

It's scary that they know how to open a .txt file, check which words at being filtered, and not use them? A 4 year old could figure that out.

[Destinye]

It's scary that they know how to open a .txt file, check which words at being filtered, and not use them? A 4 year old could figure that out.

It's just people generally assume RMT are these mindless drones. Though in the end, no matter what is done they will get around it if they're determined enough, as proven by them getting around both filters.

Hell most hackings come from China/North Korea, so it's not out of the realm of possibility that they have hackers on standby ready to get around the next Chatmon update. Especially if it has anything to do with FF's memory.

[Aikar]

dont think a hash based approach would work since were using patterns and regular expressions to check.

But as for encryption, it wont be that easy. for one i dont put passwords in easy to find strings now if they want to guess every combination of raw bytes in quadrillions of possibile combinations then let them.

Then theres also the fact the DLL's are packed with a very good commercial packer thats not easily depackable, with modules loaded that detect attempts to reverse engineer.

I don't play around

[fishstix]

Does the custom filter work? Maybe I'm doing it wrong, but things still get past like if I add:
*cheap*
to the spamfilter_custom.txt file

[xopher]

One RMT loads his computer with chatmon. Another sends him tells til they go through.

Encryption defeated.

[Mafai]

is it possible to filter based on name? id like to be able to do

If name = Ifr* ignore....

seems all the RMT tells i get start with the 1st 3 letters of my server.

[Tankjr]

http://images.usatoday.com/money/_ph...-bud-radio.jpg

Here's to you Mr. Windower Guy. Oh Sultan of C#.

When life gives you lemons...

you crush them mercilessly.

Keep up the great work you Knight of .NET

[shepardG]

87% of Tankjr's posts cause my face to smile

just saying

[Dramafinderftw]

http://images.usatoday.com/money/_ph...-bud-radio.jpg

Here's to you Mr. Windower Guy. Oh Sultan of C#.

When life gives you lemons...

you crush them mercilessly.

Keep up the great work you Knight of .NET

you funny'

Spoiler: show

congrats on amano fat fuck

[The Blackrose]

One RMT loads his computer with chatmon. Another sends him tells til they go through.

Encryption defeated.

Yes Mr. Armchair Cryptologist, that's certainly within the realm of feasibility.

[Pikarya]

Yes Mr. Armchair Cryptologist, that's certainly within the realm of feasibility.

China can do anything, mang.

[rog]

is it possible to filter based on name? id like to be able to do

If name = Ifr* ignore....

seems all the RMT tells i get start with the 1st 3 letters of my server.

http://forums.windower.net/topic/111...5&#entry191965

[xopher]

Yes Mr. Armchair Cryptologist, that's certainly within the realm of feasibility.

just pointing out that no matter how much incredible encryption you put into it, even the simplest methods can defeat it. ad infinitum.