When I wrote the decoder for Chevin Software we attempted to be as
"networkingly" honest as possible as
to what we saw on the line. I, as a result had endless traces sent to me
from clients who claimed that our
software was buggy when we depicted the jam code 0xAA/0x55 in all its glory
when compared to A.N. OTHER
analysers that did'nt capture this at all and was therefore "bug free". I
think that due to these types of support calls
most analyser developers don't show the truth of the matter.
Regards,
Phil.
----- Original Message -----
From: "Priscilla Oppenheimer"
To:
Sent: Saturday, June 02, 2001 11:05 PM
Subject: Re: ethernet jam signal ? [7:5796]
> At 10:07 AM 6/2/01, E Joseph wrote:
> >Priscilla,
> > What would a the resulting jam look like on a
> >sniffer trace??
>
> If the jam signal is all ones, it would look like 0xFFs on a Sniffer. The
> jam doesn't have to be all ones, though. On old bus coax networks, it was
> actually rare to see the jam because by the time the jam happened the
> clocking was so messed up that the Sniffer had already stopped capturing
> and just reported a runt, CRC error, collision.
>
> On a network with repeaters (hubs), when the repeater detects a collision
> it sends a 96-bit jam composed of alternating ones and zeros, which would
> look like 0xAAs or 0x55s. That's easier to see.
>
> A lot of Sniffers don't capture bad frames or runts, so you wouldn't see
> any of these jams in those cases.
>
> Someone else may have some additional information. It's always been a
> question of mine also whether you can really see jams or not.
>
> Of course in networks where full-duplex switch ports have replaced hub
> ports, this is no longer relevant.
>
> Please send messages to the group, not to me. Thanks,
>
> Priscilla
>
>
>
> > Thank You,
> > Ed
> >
> >
> >
> >--- Priscilla Oppenheimer wrote:
> > > When a transmitter detects a collision, the
> > > transmitter continues to send
> > > the preamble, (if the preamble has not completed),
> > > and also sends 32
> > > additional bits, which are called a jam signal. The
> > > jam signal extends the
> > > duration of the collision event to ensure that all
> > > stations hear the
> > > collision. The contents of the jam can be any
> > > pattern that is not
> > > intentionally designed to be the 32-bit CRC value
> > > corresponding to the
> > > (partial) frame already transmitted. Most
> > > implementations send all ones.
> > >
> > > Completely sending the preamble and transmitting a
> > > jam signal guarantees
> > > that a signal stays on the media long enough for all
> > > transmitting stations
> > > involved in the collision to recognize the collision
> > > and react accordingly.
> > >
> > > Priscilla
> > >
> > >
> > > >On Thu, 24 May 2001, [EMAIL PROTECTED] wrote:
> > > >
> > > > > What is an "ethernet a jam signal"
> > >
> > >
> > > ________________________
> > >
> > > Priscilla Oppenheimer
> > > http://www.priscilla.com
> >[EMAIL PROTECTED]
> >
> >
> >__________________________________________________
> >Do You Yahoo!?
> >Get personalized email addresses from Yahoo! Mail - only $35
> >a year! http://personal.mail.yahoo.com/
>
>
> ________________________
>
> Priscilla Oppenheimer
> http://www.priscilla.com
Message Posted at:
http://www.groupstudy.com/form/read.php?f=7&i=6980&t=5796
--------------------------------------------------
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
