Date: Tue, 04 Jul 2000 11:50:55 +0100 From: Tim Priebe <tim@polytechnic.edu.na> To: "Louis A. Mamakos" <louie@TransSys.COM> Cc: David Gilbert <dgilbert@velocet.ca>, Joerg Micheel <joerg@cs.waikato.ac.nz>, freebsd-stable@FreeBSD.ORG Subject: Re: Ethernet MTUs > 1500? Message-ID: <3961C18F.E8753501@polytechnic.edu.na> References: <14689.22689.894466.908666@trooper.velocet.net> <20000704153914.C60136@cs.waikato.ac.nz> <14689.23903.87264.511506@trooper.velocet.net> <200007040353.XAA03131@whizzo.transsys.com> <14689.27160.907313.347624@trooper.velocet.net> <200007040508.BAA59763@whizzo.transsys.com>
next in thread | previous in thread | raw e-mail | index | archive | help
"Louis A. Mamakos" wrote: > > > >>>>> "Louis" == Louis A Mamakos <louie@TransSys.COM> writes: > > > > Louis> There's some confusion here, because the MTU is typically > > Louis> associated with a protocol stack like IP and refers to the > > Louis> largest sized (IP in this case) packet that can be sent on the > > Louis> network interface. In the case of Ethernet interfaces that > > Louis> support and use VLAN tags, the MTU is still 1500 bytes, > > Louis> regardless of the fact that the frame size is a few bytes > > Louis> longer to accomodate the VLAN tag information. > > > > Louis> So, this has no effect on the Ethernet type field (or 802.3 > > Louis> length fields) since the higher level protocol packet size is > > Louis> unchanged. > > > > Why, then, are the vlan MTUs hardwired at 1496? > > You got me. Perhaps the code is busted. Perhaps the code is trying > to accomodate ethernet NICs that cannot send or receive larger > than "normal" ethernet frames. (e.g., those with VLAN tags or > priority labels). > > I just know that on networks that I've built and used with Ethernet > switches which use VLAN tagging, I've had no problems transporting > 1500 byte MTU IP packets inside of ethernet frames. If the NIC card > in the FreeBSD host can't send the larger frames to accomodate the > VLAN tags, then it's broken. From the days I was last getting this to work ~3.2-RELEASE, the real problem is not an MTU of 1496 (unless it is a router), but an MRU of 1496. From what I rembember fixing it to send the extra 4 bytes was not that difficult. I eventually fixed my firewall, to mostly function with a MTU and MRU of 1500 on the vlan interfaces, but at the cost that it would spontaniously reboot if you tried to establish a tcp connection. This has not been a real problem for me, but has been for some others that report the same behaviour, and wanted to use it in a web server. I think it is probably time to have a policy decision, that we are willing support the 802.1Q and 802.1p frames on equipment that is able to send and recieve the extra 4 bytes per frame. Then a framework can be developed to support this with out too much additional overhead in the drivers. I am willing to put some time in to this. For clarification, I am not suggesting that we need to look at prioritisation of packets, based on 802.1p, at this time, just stop discarding them if they are "too big". Tim. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3961C18F.E8753501>