Cisco HDLC Flags (was PPP vs HDLC) [7:64779]

2003-03-07 Thread s vermill
The question was whether or not Cisco used the "standard" 0x7E as a flag in
their HDLC implementation.  The only WAN protocol analzer I could dig up
predates Cisco HDLC by a few decades.  So I did rely on an o'scope as
planned.  Between keepalives, a constant binary
10011001100110 can be observed.  This is the classic 7E7E
scope trace.  Seems to me that most HDLC implementations that I can remember
do two 0x7Es and then all zeros for the remainder of the frame header (no
"payload zeros").  Cisco apparently just keeps 'em a comin'.  Of course,
once keepalives or real data hit the line, there's no way to distinguish the
7E flag between frames.  But given that 7E7E is used as an idle pattern, it
isn't too much of stretch to assume they serve as frame delimiters as well. 
I'll try to verify that with a Cisco HDLC-capable WAN analyzer if I ever get
my hands on one and can get it latched on to a Cisco serial connection.

 




Message Posted at:
http://www.groupstudy.com/form/read.php?f=7&i=64779&t=64779
--
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]


RE: Cisco HDLC Flags (was PPP vs HDLC) [7:64779]

2003-03-07 Thread Priscilla Oppenheimer
Thanks Scott! To synthesize:

The original question was whether efficiency would be improved if Cisco HDLC
were used instead of PPP. Most of us said "no," which is correct, (taking
efficiency to mean header overhead.)

A few of us added the caveat that a Cisco HDLC header/trailer maybe has a
couple bytes less than a PPP header/trailer. That's not true.

RFC 1331 specifies the PPP header like this: 

Flag 1 byte (0110) 
Address 1 byte 
Control 1 byte 
Protocol 2 bytes 
info (variable) 
FCS 2 bytes 
Flag 1 byte (0110) 

With the new info from Scott that says Cisco HDLC uses flags also, one could
say that the above description of PPP matches Cisco HDLC also.

(The exact behavior of the flags for either one, from a frame-format and
efficiency point of view, is really not too relevant. In fact the newer RFC
for PPP (1661) doesn't mention them and the document on Cisco HDLC doesn't
mention them either.)

The main point is that Cisco HDLC and PPP are equivalent as far as
efficiency, (if by efficiency we mean header overhead.) QED. :-)

___

Priscilla Oppenheimer
www.troubleshootingnetworks.com
www.priscilla.com


s vermill wrote:
> 
> The question was whether or not Cisco used the "standard" 0x7E
> as a flag in their HDLC implementation.  The only WAN protocol
> analzer I could dig up predates Cisco HDLC by a few decades. 
> So I did rely on an o'scope as planned.  Between keepalives, a
> constant binary 10011001100110 can be
> observed.  This is the classic 7E7E scope trace.  Seems to me
> that most HDLC implementations that I can remember do two 0x7Es
> and then all zeros for the remainder of the frame header (no
> "payload zeros").  Cisco apparently just keeps 'em a comin'. 
> Of course, once keepalives or real data hit the line, there's
> no way to distinguish the 7E flag between frames.  But given
> that 7E7E is used as an idle pattern, it isn't too much of
> stretch to assume they serve as frame delimiters as well.  I'll
> try to verify that with a Cisco HDLC-capable WAN analyzer if I
> ever get my hands on one and can get it latched on to a Cisco
> serial connection.
> 
>  
> 
> 




Message Posted at:
http://www.groupstudy.com/form/read.php?f=7&i=64798&t=64779
--
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]