Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > comp.os.linux.networking > #1226

Re: test protocol at client?

From Chris Davies <chris-usenet@roaima.co.uk>
Newsgroups comp.os.linux.networking
Subject Re: test protocol at client?
Date 2012-04-02 13:03 +0100
Organization Roaima. Harrogate, North Yorkshire, UK
Message-ID <75ho49xpg4.ln2@news.roaima.co.uk> (permalink)
References <4f76acb9$1@x-privat.org> <as4j49x081.ln2@news.roaima.co.uk> <4f7963cc$1@x-privat.org>

Show all headers | View raw


ghand <ghand45@hotmail.com> wrote:
> OK so I suppose it can't set itself up as a server let the client connect,
> and then detect what protocol is used at the client?

In many caess the server is expected to say "hello" before the client
is permitted to talk. (SMTP and FTP are examples of this.)


> Suppose I have a client and SOCKS server, and sniff the packets between
> them and capture them. (with tcpdump)

> I have done that with a SOCKS server I have. And i'd like something
> to detect whether the packets, are SOCKS4 or SOCKS5.

Personally I don't know the difference. But I see no reason why there
couldn't be something that looked at the packet contents and determined
whether to route them to a SOCKS4 or SOCKS5 server.

A quick google suggests that you might want to take a look at "redsocks";
looks like it already does this.

Chris

Back to comp.os.linux.networking | Previous | Next — Previous in thread | Find similar | Unroll thread


Thread

test protocol at client? "ghand" <ghand45@hotmail.com> - 2012-03-31 09:05 +0100
  Re: test protocol at client? Chris Davies <chris-usenet@roaima.co.uk> - 2012-03-31 12:03 +0100
    Re: test protocol at client? "ghand" <ghand45@hotmail.com> - 2012-04-02 10:31 +0100
      Re: test protocol at client? Chris Davies <chris-usenet@roaima.co.uk> - 2012-04-02 13:03 +0100

csiph-web