Mailing List ArchiveSupport open source code!
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]Re: IPChains rules
- To: <tlug@example.com>
- Subject: Re: IPChains rules
- From: "Stephen J. Turnbull" <turnbull@example.com>
- Date: Mon, 5 Mar 2001 12:10:15 +0900
- Content-Transfer-Encoding: 7bit
- Content-Type: text/plain; charset=us-ascii
- In-Reply-To: <Pine.LNX.4.30.0103031439560.8903-100000@example.com>
- References: <15008.33424.983448.180679@example.com><Pine.LNX.4.30.0103031439560.8903-100000@example.com>
- Reply-To: tlug@example.com
- Resent-From: tlug@example.com
- Resent-Message-ID: <s_QuxC.A.oZE.yTwo6@example.com>
- Resent-Sender: tlug-request@example.com
>>>>> "Tobias" == Tobias Diedrich <ranma@example.com> writes: Tobias> Now I'm not sure what you mean with that ^^;; ipchains -y flag IIRC. Filter on SYN flag. Tobias> Works for ping, traceroute, >> Wrong. Ping is ICMP, traceroute is UDP. No state ... sorry. Tobias> Well it's true that these are "stateless" connections, but Tobias> the packet filter treats them as statefull connections Tobias> anyway, which is possible and works quite good. Well, you should be more careful. The connections aren't stateful, the filter is. That is useful, it is more efficient to open windows for ports from the TCP/IP stack than to do it with user space routines. But they shouldn't have polluted the namespace by using TCP/IP terminology for states which do not correspond accurately to TCP states. Tobias> It will recognize the corresponding ICMP-ECHO-REPLY packet Tobias> from that host and set the state to ESTABLISHED and will Tobias> let this packet through (with my rules that is) and after Tobias> that the connection is finished. A second ICMP-ECHO-REPLY This is very fragile. It requires that a lot of information about the various protocols and their common _incorrect_ usages be built into the TCP/IP stack. That's not a good idea (requires rebuilding the kernel and reboot to change, I suppose?) I guess if it's done as a module it's not so bad, but it's not robust to specification error. Tobias> packet would not be let through because I have not sent a Tobias> ICMP-ECHO-REQUEST packet first. Man-in-the-middle by definition means the attacker gets in first. This is probably better than nothing (although it throws away the potentially useful information that multiple packets came back), but it doesn't create state (or a connection!) where there was none. Tobias> I'm pretty sure that traceroute is ICMP too btw. On the LBL implementation, only if you use the -I flag. -- University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN Institute of Policy and Planning Sciences Tel/fax: +81 (298) 53-5091 _________________ _________________ _________________ _________________ What are those straight lines for? "XEmacs rules."
- References:
- Re: IPChains rules
- From: "Stephen J. Turnbull" <turnbull@example.com>
- Re: IPChains rules
- From: Tobias Diedrich <ranma@example.com>
Home | Main Index | Thread Index
- Prev by Date: Re: linux only with linux?
- Next by Date: Re: IPChains rules
- Prev by thread: Re: IPChains rules
- Next by thread: Re: IPChains rules
- Index(es):
Home Page Mailing List Linux and Japan TLUG Members Links