From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from lists.gentoo.org ([140.105.134.102] helo=robin.gentoo.org) by nuthatch.gentoo.org with esmtp (Exim 4.62) (envelope-from ) id 1GuErf-00050S-IT for garchives@archives.gentoo.org; Tue, 12 Dec 2006 21:08:40 +0000 Received: from robin.gentoo.org (localhost [127.0.0.1]) by robin.gentoo.org (8.13.8/8.13.8) with SMTP id kBCL7R6g009784; Tue, 12 Dec 2006 21:07:27 GMT Received: from aa012msr.fastwebnet.it (aa012msr.fastwebnet.it [85.18.95.72]) by robin.gentoo.org (8.13.8/8.13.8) with ESMTP id kBCL2xHP024126 for ; Tue, 12 Dec 2006 21:02:59 GMT Received: from [23.252.116.250] (23.252.116.250) by aa012msr.fastwebnet.it (7.3.105.6) (authenticated as sergio.polini) id 4556F291016C1672 for gentoo-user@lists.gentoo.org; Tue, 12 Dec 2006 22:02:59 +0100 From: Sergio Polini To: gentoo-user@lists.gentoo.org Subject: Re: [gentoo-user] OT: wlan0 is sssloooow [100% SOLVED] Date: Tue, 12 Dec 2006 22:03:06 +0100 User-Agent: KMail/1.9.1 References: <200611261648.03512.sp_rm_it@yahoo.it> <200612112156.48512.sp_rm_it@yahoo.it> <200612120815.19070.michaelkintzios@gmail.com> In-Reply-To: <200612120815.19070.michaelkintzios@gmail.com> Precedence: bulk List-Post: List-Help: List-Unsubscribe: List-Subscribe: List-Id: Gentoo Linux mail X-BeenThere: gentoo-user@gentoo.org Reply-to: gentoo-user@lists.gentoo.org MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Content-Disposition: inline Message-Id: <200612122203.06541.sp_rm_it@yahoo.it> X-Archives-Salt: 25a7524a-4d49-4015-95b0-7babaa130cb3 X-Archives-Hash: b02eaf792e03ae3cd677d9c5fc54a778 Mick: > On Monday 11 December 2006 20:56, Sergio Polini wrote: > > Yes, it was OT and very few of you are interested ;-) > > However, somebody would like to know who the killer was. > > > > The original subject was "wlan0 is sssloooow [99% SOLVED]" > > because pinging the Belkin F5D7230-4 wireless router worked from > > Windows but not from Linux. > > The answer is: because Windows sends ICMP messages with 32 bytes > > of data, Linux sends them with 56 bytes of data. > > Moreover, Linux IP datagrams have the DF (don't fragment) bit > > set, Windows ones have not. > > On Linux, ping -s 15 192.168.2.2 works. > > How did you ever find this out?! I used two "tools": Douglas Comer, "Internetworking with TCP/IP", and Wireshark (former ethereal). I'm still installing my new laptop, f.i., I've not yet setup a firewall. My former firewall was based on the "Iptables Tutorial" by Oskar Andreasson, but I didn't understand all the details. So I've started studying, i.e. reading Comer and looking at frames, datagrams, etc. by Wireshark. I needed that. For example, I had used nmap to look for an echo port on the Belkin router, but ping, i.e. ICMP, doesn't know anything about ports! As to the ping problem, I started a Windows virtual machine by VMware Player, then Wireshark both in the Linux "real" machine and in the virtual one, then pinged the Belkin router. Looking at the output produced by the Linux Wireshark and by the virtual Windows one, there were just two differences: the DF bit, and the data field length in the ICMP messages. > I'll know that's another thing to test when an access point > is playing up. I wonder why belkin is set up this way. I think that a router should send ICMP messages such as "fragmentation needed and DF set" (Type 3, , Code 4) and "time to live exceeded in transit" (Type 11, , Code 0), but Belkin does not (traceroute, and tracert, 192.168.2.1 print stars). The Belkin 54G wireless router, F5D7130, had serious security holes: http://www.governmentsecurity.org/archive/t15618.html My model, F5D7230, is more secure. Eventually too much ;-) > Thanks for sharing. :-) Alan McKinnon: > I'm very glad you did post this update as here in the office we had > this very problem three weeks ago. One morning every non-Windows > host in our building suddenly could not see past the gateway, could > not ping it and was essentially off-air. We eventually tracked it > down to one of these Belkin wireless routers, but never figured out > why it was doing what it did. > > Now we do know, so thanks for the heads-up! I must thank all of you. Gentoo would not be such an attractive system without the continuous support by all of you. Sergio -- gentoo-user@gentoo.org mailing list