Date: Mon, 27 Mar 2006 21:03:50 +0100 From: Brian Candler <B.Candler@pobox.com> To: Tyrone@telecity.se Cc: freebsd-isp@freebsd.org Subject: Re: udp packet problem Message-ID: <20060327200350.GA843@uk.tiscali.com> In-Reply-To: <D3BBF0C6F2FC0448BFCA2F965F2192630F6F12@sto1.tcy.prv> References: <D3BBF0C6F2FC0448BFCA2F965F2192630F6F12@sto1.tcy.prv>
next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, Mar 27, 2006 at 10:56:30AM +0200, Tyrone@telecity.se wrote: > We have a Freebsd 5.4 router doing traffic shaping on a vlan interface. > > Problem is we can't receive UDP packets larger than 2K is this a > limitation or a setting within IPFW? * Show your exact setup: what version of FreeBSD, what types of NIC you have, how you have configured them, how you've configured ipfw etc. * Show what makes you think that UDP packets larger than 2K are being dropped. (For example, show a tcpdump of packets on the incoming interface and on the outgoing interface. Show what program you're using to send UDP packets) * Preferably, boil this down to a simple test which demonstrates the problem in the simplest possible way. For example, if you turn off ipfw completely and can still demonstrate the problem, then this shows that ipfw is not at fault. > How can I fragment UDP packets that are larger than the interface MTU ? This happens automatically when sending packets. Inbound packets which are larger than the configured MTU but successfully received, are accepted and not modified. But outbound packets which are larger than the MTU of the interface they are about to be sent out from, are fragmented. > kernel option or FW ruleset ? No, it just happens (unless you can demonstrate otherwise). Once you have a more accurate description of your problem, freebsd-net is probably a better place to post it. HTH, Brian.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20060327200350.GA843>