>> [MSS clamping is] the Wrong Fix, but since it appears you're willing
>> to paper over the problem instead of fixing it....
> I understand the "right fix" IYO has to do with the fixing of a
> router somewhere between me and cvsup servers on two different
> continents.  Perhaps even within my not particularly competent ISP.

Yes.  You can see why taking the superficially easy way out is popular.

> Is there some tool other than traceroute for finding the perpetrator,

I don't know of any tool - including traceroute - for finding the
perpetrator, short of sniffing traffic at various points between the
offending endpoint and the MTU-limiting gateway and seeing when the
"unreachable - need to frag" ICMPs start vanishing.

In the few cases where I've been able to get some idea where the
problem is, it's always been close to the problematic endpoint, the
host that tries to send bulk data and fail because of the PMTU-D black
hole.

> and what's the magic phrase to use to describe the problem in a way
> that the Microsoft-quality technical staff might understand?

If "RFC 2923 section 2.1" doesn't do it, there probably isn't much that
will, short of an interactive explanation in person.

/~\ The ASCII				der Mouse
\ / Ribbon Campaign
 X  Against HTML	       mouse@rodents.montreal.qc.ca
/ \ Email!	     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B