All of lore.kernel.org
 help / color / mirror / Atom feed
From: Alistair Tonner <Alistair@nerdnet.ca>
To: trainier@kalsec.com, netfilter@lists.netfilter.org
Subject: Re: irc
Date: Mon, 17 Nov 2003 19:00:21 -0500	[thread overview]
Message-ID: <200311171900.21550.Alistair@nerdnet.ca> (raw)
In-Reply-To: <OF3489540F.204A1A44-ON85256DE1.0071CE7C-85256DE1.0072EC94@kal.kalsec.com>

On November 17, 2003 03:55 pm, trainier@kalsec.com wrote:
> The ident thing is partially true.  EFnet (the network I use), does claim
> to require ident.  If you actually look though, it initiates the
> connection, before it sends out it's ident request.

	At least two of the UnderNet servers deny connections without
	ident.... and several misconfigured private servers for ISPs do the 
	same ... *sigh* (leastways far as I've been able to track my 
	*cough* households usage....)

>
> I am not concerned at all with dcc.  dcc was a huge mistake and should've
> never been implemented into irc.
> I am, however, interested in the nat irc handlers.  Where do I get them
> and how do I use them? What're they for?
>
> Regards,
>
> Tim

	They facilitate the DCC connections... 
	I've never had problems with basic connections to IRC servers
	without the NAT handler .. .the iptables NATting and 
	ESTABLISHED,RELATED rules handle server connections 
	just fine.  DCC initiates  a *new* connection inside packets across
	 the initial connection: iptables irc modules can peek inside these
	packets and setup the connection to the other end of the DCC request
	 (thus Direct Client to Client), this connection is NOT to the server and 
	has to be tracked separately from the initial server connection.
	the ip_nat_irc and ip_conntrack_irc modules in iptables manage this
	rather well.... And apparently we've fixed the problem that was caused
	by some clients trying to outsmart NAT by using the outside ip at the 
	client in the most recent iptables release and POM.  (way to go developers)

	(if you had set your client to use the ip address that the IRC server saw,
	ip_nat_irc and ip_conntrack_irc would label DCC connections as 
	'forged dcc requests' since they already had the natted address in the
	packet ... instead of the unnatted address)

>
>
>
>
> Antony Stone <Antony@Soft-Solutions.co.uk>
> Sent by: netfilter-admin@lists.netfilter.org
> 11/17/2003 03:05 PM
>
>
>         To:     netfilter@lists.netfilter.org
>         cc:
>         Subject:        Re: irc
>
> On Monday 17 November 2003 7:56 pm, Alistair Tonner wrote:
> >                If you've several systems that want to connect and do DCC
>
> you
>
> >                will want to make sure you load the conntrack and nat irc
> >                handlers from iptables ... they aren't needed for plain
>
> connections,
>
> >                but are for DCC sends/recieves.
>
> Indeed, however I assumed that anyone interested in the security of having
> a
> firewall wouldn't be using insecure things like DCC.   However, your
> reminder
> that there is a conntrack helper for this protocol is a good one.
>
> >                As a rule these days a LOT of irc servers want an identd
>
> reply ...
>
> >                identd is a horrible security problem, but you can use
>
> several
>
> >                alternatives ... I've a python script that acts as a
>
> chrooted identd
>
> >                server -- works a charm replying with random numbers ....
>
> You mean they actually require an identd response before allowing a
> connection (rather than just making it take a bit longer than usual)?
>
> What's the point in that?   It adds nothing to security, adds very little
> to
> logging opportunities, and only interferes with people trying to keep
> their
> networks to themselves.
>
> Ho Hum; it's a strange world on the Internet....
>
	Indeeeeeeeed it is..... *cough* wintendo *cough* skiddies included.
	(sorry ... i'm sitebanning a bunch of morons from a mush right now.)


> Antony.

-- 

	Alistair Tonner
	nerdnet.ca
	Senior Systems Analyst - RSS
	
     Any sufficiently advanced technology will have the appearance of magic.
	Lets get magical!


  reply	other threads:[~2003-11-18  0:00 UTC|newest]

Thread overview: 36+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2003-11-17 20:55 irc trainier
2003-11-18  0:00 ` Alistair Tonner [this message]
  -- strict thread matches above, loose matches on Subject: below --
2025-03-23 17:36 IRC Julia Lawall
2021-05-20 17:09 IRC Nishanth Menon
2021-05-20 20:43 ` IRC Luca Ceresoli
2009-03-05  8:51 IRC Harald Hoyer
2005-06-01 15:06 IRC Charles Coffing
2005-06-01 15:24 ` IRC Mark Williamson
2005-03-30 20:08 irc Jody Belka
2005-03-30 20:57 ` irc Rik van Riel
2004-03-12 18:15 irc Yoshinori K. Okuji
2004-03-13  0:33 ` irc Sergey Matveychuk
2003-11-17 18:42 irc trainier
2003-11-17 19:29 ` irc Antony Stone
2003-11-17 19:56   ` irc Alistair Tonner
2003-11-17 20:05     ` irc Antony Stone
2003-11-17 18:38 irc trainier
2003-11-17 17:58 irc trainier
2003-11-17 17:43 irc trainier
2003-11-17 18:11 ` irc Antony Stone
2003-11-17 16:28 irc bmcdowell
2003-11-17 16:38 ` irc Antony Stone
2003-11-17 15:30 irc trainier
2003-11-17 15:39 ` irc Antony Stone
2003-11-17 16:10   ` irc SBlaze
2003-11-17 16:18     ` irc Antony Stone
2003-11-17 16:35       ` irc Jeffrey Laramie
2003-11-17 17:33         ` irc Antony Stone
2003-11-17 18:04           ` irc Jeffrey Laramie
2003-11-17 15:41 ` irc dan radom
2003-11-17 15:50   ` irc Antony Stone
2003-11-17 15:15 irc trainier
2003-11-17 15:20 ` irc Antony Stone
2003-11-17 15:33   ` irc Jeffrey Laramie
2003-11-17 15:39     ` irc Antony Stone
2003-11-17 15:22 ` irc dan radom

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=200311171900.21550.Alistair@nerdnet.ca \
    --to=alistair@nerdnet.ca \
    --cc=netfilter@lists.netfilter.org \
    --cc=trainier@kalsec.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.