Re: [DNA] Revised text for link-information Ethernet section forcomment

[Greg Daley <greg.daley@eng.monash.edu.au>]

Hi,

I've tried to address some of your concerns below.

Please tell me if it's still off-track.

Tom Petch wrote:
> I agree with Dan and would add:
> 
> Properly the technology is IEEE 802.3 CSMA/CD (and even that leaves out quite a
> lot of the
> title) and I would like you to introduce it as such, adding something like
> 'commonly referred to as Ethernet'
> 
> If you are going to use Ethernet (as opposed to say CSMA/CD or IEEE 802.3) in
> the rest of the document, then you should add a sentence to that effect (the
> term Ethernet scarcely appears in IEEE 802.3).

That sounds fair enough.

How about the following introduction:

"Section title:   IEEE 802.3 CSMA/CD
------------------------------------

IEEE 802.3 CSMA/CD (commonly referred to as Ethernet) is the
most commonly deployed Local Area Network technology in use today.
As deployed today, it is specified by both a physical layer/medium
access control (MAC) layer specification[802.3].
In order to provide connection of different LANs together into
a larger network, 802.3 LANs are often bridged together[802.1d].

In this section, the terms 802.3 and Ethernet are used interchangeably.
This section describes some issues in providing link-layer indications
on Ethernet networks, and shows how bridging affects these indications."

Is this more accurate?

> 
> It is wrong to say that all media have Link Integrity Tests eg 10BASE5 (yes, it
> is still in use).

OK...

Perhaps it is worth introducing the concept more explicitly.

"In order to determine whether the physical medium is ready for
  frame transfer, IEEE 802.3 Ethernet specifies its own link
  monitoring mechanism, which is defined for some, but not all
  classes of media.
  Where available, this Link Integrity Test operation is used
  to identify when packets are able to be received on an
  Ethernet segment.  It is applicable to both wired and optical
  physical layers, although details vary between technologies
  (link pulses in twisted pair copper, light levels in fibre)."

Is that a reasonable description of the variation between media?


> The default I see is BPDU Hello every two seconds so within the default 20s,
> more than two will have been lost.  But this paragraph and the following one
> confuse me with references to peer device.  This is broadcast, a degenerate case
> being only two devices, and if there is a bridge anywhere in the topology, then
> I expect to see BPDU every 2s. Spanning Tree will not work without them so not
> seeing them says either no bridges or Spanning Tree has been disabled (does
> happen, used to be common with slow - eg WAN - links)


I guess the peer LAN connected device could be either a bridge, an
end-station (router or host) or a repeater (in which case any other
device becomes visible on the LAN).

I there are bridges anywhere on the LAN, then you would expect
to see Hellos (although, topologies without bridges are valid too).

In section 8.10.2 of 802.1D 1998, (Table 8-3, the Hello timer has a
range of 1.0-10.0 seconds).

Also, there's an inequality:

Bridge_Max_Age >= 2 x (Bridge_Hello_Time + 1.0 seconds)

So I assumed that the inference of at least 2 messages were lost
could be applied.

Should I be more explicit about the maximum timeout?

I'm wondering if explicit statements about what a 'peer' may be,
and which devices are STP participants would help.  How about this?:

"...
For these technologies, even though the link-layer appears
available, no forwarding will occur until it is determined that
the port can be connected to the network in a loop-free
environment.

(new text)
   For host which are providing indications to upper layer protocols,
   even if the host itself does not implement bridging or STP,
   packet delivery across the network can be affected by their
   directly adjacent LAN peer which may be a bridge, or if this is
   a repeater, by the presence of STP bridges upon that LAN.

   Where the host is not running STP itself,
(end new text)
No explicit indication that forwarding has begun is sent from
a switch to its peer devices.  Therefore, a host may not know
when STP operations have completed, and when it is safe to
inform upper layers to transmit packets.
..."

Does this provide sufficient context to help disambiguate the
role of each device, and whether it is running STP?

Thanks for your work on this.

Greg