[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [RRG] MTU, jumboframes, ITR & ETR placement, ITR function in hosts

On 26 nov 2007, at 6:36, Robin Whittle wrote:


because of broken PMTUD behavior in
way too many hosts connected to the internet.



I assume you mean that the host doesn't respond to RFC 1191 "Packet
Too Big" (actually "Datagram Too Big") messages by resending the
message in smaller packets according to the Next-Hop MTU value in
that message.  I assume you are not referring to filters which block
these messages, either in the network of the sending host or in the
network nearer the destination host.
Actually I am. Or rather, I'm assuming this is the issue because I don't have access to the internals of broken PMTUD deployments. I would hope that the IP stacks aren't so broken that if the ICMP messages make it back in one piece it still doesn't work, some two decades after the RFC was published. 


If these messages are dropped by filters in the sending host's
network, or anywhere between the sending host and the nearest ITR
(perhaps outside that network's ISP) then there is something wrong
with the filtering system and I think it should be fixed.


You and me both.


But if there are only occasional applications or operating systems
which don't do RFC 1191 PMTUD correctly, then wouldn't they have
trouble today
What you see is that if you are behind a link with a < 1500-byte MTU (but not directly attached to that link because then the TCP MSS option avoids trouble) that you can't reach certain destinations on the internet: sessions establish but then stall. In most cases, you, the client, will request something and the request is small but the reply is large. So this triggers PMTUD on the part of the remote website etc but they never adjust their packet sizes so nothing happens. If this happens with one site you could go after them and hopefully they'll fix it, but that doesn't work when it's tens or hundreds of thousands. The only option you have as a customer behind a PPPoE link or some such is for your CPE to adjust the TCP MSS so others never send you large packets in the first place. And because everyone does that, the people responsible for the problem continue their brokenness. 


2 - "Only" insist on a gigabit upgrade to all ISPs who want to run
   ITRs (we are intending they all do it pretty soon) and then
   generally accept that ITRs have to be in ISP networks and not
   in parts of end-user networks which don't have gigabit links
   to the ISP.


:-)


I suppose there is the prospect of re-engineering 100Mbps gear to
have a higher MTU, but I guess that means new silicon, with very
little profit from higher prices - so I assume this is not much of
an option.


Right, "100 Mbps gear with new silicon" would be 1000 Mpbs gear.


Also, so many things have their Ethernet interface built
in, that there would need to be complete upgrades to many servers
and devices, not "just" plugging in faster or jumboframe-compatible
Ethernet cards into routers.
If it has gigabit, in most cases it will do jumboframes. My Dell FreeBSD machine: 

# ifconfig em1 mtu 16115
ifconfig: ioctl (set mtu): Invalid argument
# ifconfig em1 mtu 16114
# ifconfig em1
em1: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> mtu 16114

My Apple MacBook Pro laptop:

# ifconfig en0 mtu 9001
ifconfig: ioctl (set mtu): Invalid argument
# ifconfig en0 mtu 9000
# ifconfig en0
en0: flags=8863<UP,BROADCAST,SMART,RUNNING,SIMPLEX,MULTICAST> mtu 9000
(Although my 4-year-old Powerbook also has gigabit ethernet but can only handle 1504.)
I'm not sure if there are any cable/DSL CPEs out there that support larger packets on both ends, though. My Cisco 826 can do 4470 on the DSL (ATM) side but it only has 10 Mbps ether and: 

(config)#in e0
(config-if)#mtu 1504
% Interface Ethernet0 does not support user settable mtu.


What about routers themselves?  Wouldn't the frame size be a
fundamental design decision?
Sure. But again, if it can do gigabit ethernet, it can probably do larger packets.
Don't forget that in the days before ethernet wiped out pretty much all other link technologies, there was lots of other stuff out there and it all supported more than 1500 bytes: packet over SONET, ATM, FDDI... So unlike switches, routers tend to be designed for some flexibility in this area. 

--
to unsubscribe send a message to rrg-request@psg.com with the
word 'unsubscribe' in a single line as the message text body.
archive: <http://psg.com/lists/rrg/> & ftp://psg.com/pub/lists/rrg