Need full modem intercept control, not dedicated.

[SpenceQ]

I'm going to learn a web programming language that will allow me complete control of the modem client side. That was not my main purpose, but it is the emergency right now, so I will need to focus my tutoring on medem activity first. I was hoping Perl with CPAN would work but they too can't. I want to intercept traffic and drop packets that I don't solicit. I want to create a database from a log file and search valid IP's and domains as communication is in progress.

Most languages allow you control as a user. I want admin right control from a program even before the OP system handles the data.

Working with pFirewall.log, would XP's Netsh shell help in any way?. What I'm looking for is the equivalent of Linux's Iptables controls.(http://www.cse.msu.edu/~minutsil/iptables.html). Also I need the firewall file's attributes be to be read enables WHILE the system is writing to it.

help!. Being invaded on a regular bases and is getting to the point I can't even read any tutoring without my disconnecting when I get high unsolicited data transfers. (good thing i don't do banking on my desktop. )

I'm familiar with 16bit Delphi,VBasic,Clipper and old stuff.

Thanks.

Spence

 

[jaeusm]

I'm going to learn a web programming language that will allow me complete control of the modem client side.

When I see the phrase "web programming language", I tend to think of languages like JavaScript, PHP, etc. Those languages really don't allow direct access to hardware.

I want to intercept traffic and drop packets that I don't solicit.

You didn't specify, but I'm assuming you're wanting to write this program for Windows. What you're wanting to do would be a fun challenge, but unless you are very familiar with Windows internals and data structures, the Win32 api, and very adept at C/C++, then you're not going to get very far.

I want admin right control from a program even before the OP system handles the data.

The OS is always going to handle the data first, unless you write a low level driver.

Why not just invest in a good firewall?

 

[SpenceQ]

When I see the phrase "web programming language", I tend to think of languages like JavaScript, PHP, etc. Those languages really don't allow direct access to hardware.


You didn't specify, but I'm assuming you're wanting to write this program for Windows. What you're wanting to do would be a fun challenge, but unless you are very familiar with Windows internals and data structures, the Win32 api, and very adept at C/C++, then you're not going to get very far.


The OS is always going to handle the data first, unless you write a low level driver.

Why not just invest in a good firewall?

Setting up the firewall goes through the same process I would program. I would continually be adding to the filter as the hackers find new proxies. If I can create a database from pfirewall.log, that would give me the intruders. (I could maintain the flatfile, but MYSQL would be quicker.) Now strangely enough, XP in it's security control panel also allows you to type in the IP's you don't want. I would even bet there's a limit. I mean it doesn't take much thinking to connect the dots and allow a pick list access from a file instead of hard copy. There would be a slight slow down in page response, but more prominent using a flatfile. Just reading a packet header isn't much overhead. A failed communication would bring up the standard browser alert saying cannot access. If all goes well, I could improve the firewall to a state type but I'd need to read the packet and make decisions.

Now with a database list, eventually a provider would expend all his available proxy domains to this one thief, and they would have to tell him to take a hike making him suspect to the provider even. A few a few hundred changes maybe the provider will get suspicious.? The first few months would see high activity at the desktop PC, but it would slowly dwindle as the intruder expends available addresses.

[As an aside, the real reason for 90% of firewall trouble is that MS wants advertizers to get through. So rather than fully giving it's 100% effort to allow simple fixes to the problem, it would rather do the go between, and not make it too easy to fix, but at the same time don't make it too easy for the advertizers. Why do you think they don't implement an "inclusion" list as well as their filter list for their mail service?. How many older people never have more than 12 or so people on their mail list, and would simply just enter those names and reject the rest? Right, solicitation. The sales industry would protest.]

I was thinking that maybe I can use the old Hayes modem commands and simply drop a communication link rather than a disconnect(as I do now manually). I would of course need to know the header information for the packet so requires a read, which could be encrypted. (an intruder just started a download now as I'm typing. getting fed up!)

So I'll finish later, need to disconnect and reconnect.

OK, back up. If anyone knows how to drop a AOPen FM56 modem commlink without disconnecting, that would save me a lot of time in the meantime as I try to figure out a fix. I can tell when their hacking most of the time.

I haven't done API shell calls since my 16bit Delphi days, but it isn't really that difficult as long as the Win documentation is clear. There's also Netsh shell provision in windows that perhaps could be of some use. I'm looking into it. A Dos level program called SLIC allowed you to do the modem communication to the pre/internet terminals back then. (CompuServe days.) It had a command set, .....receive,send,etc

Thanks for the help.

Spence