[c-nsp] full duplex mismatch speed - dynamips

[Heath Jones]

Its good to get your experience Gert - Almost all of my experience to date
has been with control of both ends of the link - service provider, managed
hosting etc - so believe it or not, I haven't actually run into these issues
of forcing only one side of a link. I'll definately keep it in mind for when
that happens though!

Cheers!!


On 20 August 2010 10:12, Gert Doering <gert at greenie.muc.de> wrote:

> Hi,
>
> On Fri, Aug 20, 2010 at 10:03:12AM +0100, Heath Jones wrote:
> > You response appreciated. One fatal assumption though is me only forcing
> one
> > end of the link - where did that come from? Read back over my post,
> keeping
> > in mind that I force both ends to 100/full.
>
> If you can ensure(!) that both ends are always force-config'ed, then
> there is nothing wrong with forcing 100/full (except for those devices
> where that doesn't really work, unfortunately, there's a number of them).
>
> The problem is that networks change.  A server/router breaks down, gets
> exchanged in the middle of the night.  Configuration can not be
> copy-paste'd
> (because it's different hardware), so "duplex full" gets lost.
>
> These things happen, and in my experience, if a network runs for 5 years,
> you have lots and lots of duplex mismatches creep into your network in
> various places - we took over a larger hosting business some years ago,
> and they had been religiously nailing "manual forced full-duplex!!"
> everywhere.  I found at least 10 customer connections that had
> developed duplex mismatches over time...
>
> [..]
> > Do you have a view on what causes an end (or both ends) of a link to all
> of
> > a sudden change state and think the other end is not capable of 100/full?
> > That is what I am trying to understand. I am also referring to a link
> that
> > is not erroring like mad and the occurrences might be as low as once per
> > week.
>
> Never seen that.  But that could be cabling that "just barely so" works,
> and if something changes (humidity, ...) tolerance is exceeded...
>
> gert
> --
> USENET is *not* the non-clickable part of WWW!
>                                                           //
> www.muc.de/~gert/
> Gert Doering - Munich, Germany
> gert at greenie.muc.de
> fax: +49-89-35655025
> gert at net.informatik.tu-muenchen.de
>