From: Jim Brown (Jim.Brown@xxxxxxxxxxxxx)
Date: Sun Aug 18 2002 - 16:54:08 GMT-3
It looks like an extraneous command left in the documentation example. It is
even the same in the 12.2 docs?
You wouldn't get the points in the lab if you configured DLSW+ based on the
docs in this example, I'd put money on it.
Somebody please correct me if I'm wrong, but you don't need the source
bridge ring-group command on an Ethernet only DLSW+ peer.
-----Original Message-----
From: Frank B [mailto:frank@buff-net.com]
Sent: Sunday, August 18, 2002 1:34 PM
To: 'Phil'; 'Jon O'Nan'
Cc: ccielab@groupstudy.com
Subject: RE: DLSw+ Translation Between Ethernet and Token Ring
Well I'm confused...if the RIF is terminated locally, the token-ring
router (Router B in this example) never knows the RIF info from the
other end anyway. It should never know the existence of the virtual
ring at the other end. And since DLSW+ takes care of the bit-swapping
based on the destination media (assuming Ethernet to be canonical, DLSW+
transports it as non-canonical) I don't see how the virtual ring is
being used here?
Secondly, source-route bridge-group is actually a SRB command. I
believe you need the interface command source-bridge ring bridge
virtual-ring in order to use the virtual ring for SRB. Why you need it
in DLSW+ is the router uses SRB to the virtual ring where DLSW+ picks it
up. The dlsw bridge-group bridge-group-number command is how the
Ethernet segment is introduced to DLSW+ as described in the last
paragraph.
Therefore I don't believe it's required either! Could it be a typo?
Here's a different opinion, another Cisco link explicitly states that
it's not required:
http://www.cisco.com/warp/public/701/45.html
"Note: source-bridge ring-group on router B is not needed."
This example has no SR/TLB...just token-ring on Router A and Ethernet on
Router B--essentially the same though.
If someone has some actual data disproving my theory I'm really
interested in seeing it...Man, I just thought I was beginning to catch
on ;-)
Tossing in my 2 cents. Aloha, Frank
-----Original Message-----
From: nobody@groupstudy.com [mailto:nobody@groupstudy.com] On Behalf Of
Phil
Sent: Sunday, August 18, 2002 7:05 AM
To: Jon O'Nan; ccielab@groupstudy.com
Subject: Re: DLSw+ Translation Between Ethernet and Token Ring
Jon,
Ring 31 is needed to represent the ethernet segment to the TokenRing
world. Without it htere will be no way to have packet exchange between
the tokenring and ethernet segments.
Phil
Jon O'Nan <jononan@columbus.rr.com> escreveu: Can someone explain to
me why a virtual ring is needed on Router A - the dlsw
peer with the ethernet segment.
Thanks
Jon
http://cisco.com/univercd/cc/td/doc/product/software/ios121/121cgcr/ibm_
c/bcp
rt2/bcddlsw.htm#36509
------------------------------------------------------------------------
---Router A
hostname Router A ! . . . source-bridge ring-group 31 dlsw local-peer peer-id 128.207.111.1 dlsw remote-peer 0 tcp 128.207.1.145 dlsw bridge-group 5 ! interface loopback 0 ip address 128.207.111.1 255.255.255.0 interface Ethernet 0 no ip address bridge-group 5 ! . . bridge 5 protocol ieee ! .
Router B
hostname Router B ! . . . source-bridge ring-group 500 source-bridge transparent 500 1000 1 5 dlsw local-peer peer-id 128.207.1.145 dlsw remote-peer 0 tcp 128.207.111.1 dlsw bridge-group 5 . . interface loopback 0 ip address 128.207.1.145 255.255.255.0 interface ethernet 1/2 no ip address bridge-group 5
. . . interface tokenring 2/0 no ip address ring-speed 16 source-bridge 7 1 500 source-bridge spanning ! . . . bridge 5 protocol ieee
This archive was generated by hypermail 2.1.4 : Sat Sep 07 2002 - 19:48:28 GMT-3