RE: [RPRWG] Topology Discovery Timer.
If we focus on the SONET/SDH & Ethernet PHYs, each have a means of expressing
"neighbour status".
SONET --> LOS, LOF, AIS.
Ethernet -> Link Status.
Each of these signals are available in the ms time range.
Why couldn't those signals be leveraged? They do not need to be referred to
directly as "SONET AIS" signal for example in the core spec. Some specific
inter-working sections could go about explaining how to map the SONET/Ethernet
link status signals to the 802.17 link status signal.
jld.
marconi networks.
To: "'Anjan Kumar HG'" <AnjanHG@xxxxxxxxxxxxx>, "IEEE 802. 17 Group (E-mail)"
<stds-802-17@xxxxxxxx>
cc: (bcc: Jeanlou Dupont/MAIN/MC1)
Subject: RE: [RPRWG] Topology Discovery Timer.
Clearly, each node needs to quickly recognize (within a few ms) that a
neighbor has gone away.
This could be done by looking for interrupts from the PHY, but that may not
be considered PHY agnostic.
In addition, I the RPR protocol could greatly benefit from a topology and
capabilities discovery protocol. For example, in 802.5 there was a MAC
frame that allowed one adapter to query another adapter's ring state. This
was implemented in microcode on the adapter; even in the system's OS was
hung, the adapter could still be queried. This is a valuable little trick.
Ray Zeisz
Technology Advisor
LVL7 Systems
http://www.LVL7.com <http://www.lvl7.com/>
(919) 865-2735
-----Original Message-----
From: Anjan Kumar HG [mailto:AnjanHG@xxxxxxxxxxxxx]
Sent: Friday, May 04, 2001 1:44 AM
To: IEEE 802. 17 Group (E-mail)
Subject: [RPRWG] Topology Discovery Timer.
Hi all,
I think there are two implementation to send Topology Discovery packet
on the ring. One implemenation is "periodical" and another "need basis"
(Ex: wrap detection)
Should we implement both? How do you think about this?
Regards,
Anjan.