[pptp-server] a Cosmetic problem

[Joe Polcari]

Jerry,

Does this mean that since I am masqing anything coming in on my eth0 interface
and going out any interface except eth0, that I should set eth0 MTU
to 1452 or the PPP interface MTU to 1452?

Joe

Jerry Vonau wrote:

> Joe:
>
> Well I'll try to explain..
>
> The web server on the lan see the pptp server's lan card has
> a mtu of 1500,
> but if it sends the max of 1500 then it will not fit into
> the ppp frame.
> There is overhead involved with the encapsulation, as a
> result it has to
> fragment the frame. If it has the "don't fragment bit" set
> then it can't
> go any farther down the pipe....
>
> Found this at :
> http://feenix.burgiss.net/ldp/adsl/configure.html
>
> ----quote------
>
> Note: PPPoE adds 8 bytes of extra overhead to the ethernet
> frames
> and the correct maximum setting for the ppp0 interface MTU
> is
> 1492. If the MTU is set too high, it may cause failure of
> some web
> pages to load properly, and possibly other annoying problems
> related
> to Path MTU Discovery. You may need to also set the MTU for
> interfaces
> on any masqueraded LAN connections MTU to 1452. This does
> not apply to
> PPPoA, or bridged configurations, just PPPoE!
> ----------------
>
> Since both PPPoE and PPTP run on PPPD this may be the same
> type of problem.
>
> Just a shot in the dark, but it fits what your are
> describing
>
> Jerry
>
> Joe Polcari wrote:
> >
> > No, I haven't. Can you explain why that might make a difference
> > based on the address of the web page???
> >
> > Jerry Vonau wrote:
> >
> > > Joe:
> > >
> > > Have you tried playing with the mtu settings on either the ppp link or the
> > > ethernet interface for the lan? Sounds like a fragmentation problem to me, similar
> > > to some PPPoE problems loading web pages.
> > >
> > > Jerry Vonau
> > >
> > > Joe Polcari wrote:
> > >
> > > > Nope. That's not it.
> > > > I can go to http://www.vibrationresearch.com or
> > > > http://www.vibrationresearch.com/
> > > > but not to
> > > > http://www.vibrationresearch.com/pptpd or
> > > > http://www.vibrationresearch.com/pptpd/
> > > >

To see more info, follow the thread - Joe