Problems with AT&T

> > If someone can identify what you are actually seeing, I'll check into
> > it.
> > If you are experiencing drops or slow traces, only through the core,
> > there is an issue with excessive de-prioritization of ICMP control
> > message with a particular router type (vendcor) in the core. End to end
> > data flow has not seemed to be affected but trace and ping core
> > latencies are looking very wierd. I've been asking customers to use
> > trace only for path detail and to use end to end ping for any
> > performance data.=20
> >
> > Yes, the core is MPLS enabled. Diffserv acted on only at the edges
> > though.=20
> >
> > Michelle
> It could certainly be customers who have broken themselves. I've heard
> lots of stories about people who do PMTUD but simultaneously filter
> ICMP Can't Frag messages.
> As soon as the Path MTU drops below whatever their local box is (usually
> 1500) they "break" although due to their own screwed up config.
> Since MPLS adds additional overhead, dropping the MTU, I'ld seriously
> consider this as a possible reason.

Speaking very generally and not about any one specific network, this
is likely to not be the issue. MPLS leads to problems on Ethernet,
but I've seen no problems in anything other than Eth/FE. GigE and POS
haven't had the same issue; for one, default POS MTU is ~4k, which is
more than enough to hold packets from hosts that assume 576 or 1500,
and PMTU over an MPLS network takes the MPLS label stack size into
account when doing discovery.

Also, some implementations have framers that can accept a packet
that's actually MTU+(N*4), where N is typically no more than 4, and
more likely 2.

And I think I can say without breaking any confidentially agreements
that AT&T's backbone Probably Isn't (nudge nudge wink wink) made up of
scads and scads of 10/100Mb links everywhere. :slight_smile:

All you need is 1 1500-byte MTU L2 network to have a customer with
a broken PMTU-D setup experience a problem. Even if you have lots of
GigE links, you often have some old gear connected to said switch
which doesn't do JumboFrames, requiring that you preserve the LCD.
As someone else mentioned earlier, if there was a per-neighbor MTU
this would resolve a huge part of the problem of transitioning to
jumboframes, since it would permit staged upgrades rather than forklift
all-or-nothing upgrades (always something to be avoided if possible).

I can say from experience that these broken customers _are_ out there
and generally refuse to admit/accept that doing PMTU-D while
simultaneously filtering all ICMP is why they are broken.

It isn't an MPLS problem, any encapsulation which reduces the effective
MTU to below 1500 would tickle the customer's config bug. These days,
however, few network operators are using additional encapsulations
other than MPLS.

Also, it isn't an ethernet problem, since the point at which the customer
breaks is solely determined by their local MTU. As such, it just happens
that most customers use ethernet as their lan media, and MTUs of less
than 1500 are extremely rare (hiding their config problem).

The biggest problem you can have with MPLS is if you have customers
who are connected at 4k or 9k or what have you, and who don't do
PMTUD; I've not seen this come up as a real operational issue.

I fail to see how that would be a problem, except for the fragmentation
issue. While fragmentation is a performance issue, it doesn't lead to
lack of reachability. Whereas doing PMTU-D and simultaneously filtering
the bits that actually make it work, does lead to reachability problems.