From: John Conzone (jkconzone@xxxxxxxx)
Date: Mon Oct 02 2000 - 07:52:18 GMT-3
No. If you had a host on R2's E0 then yes.
----- Original Message -----
From: "Maljure, Sanjay" <smaljure@cibernetworks.com>
To: "'Steve McNutt'" <lpd@jacksonville.net>; <ccielab@groupstudy.com>
Sent: Monday, October 02, 2000 2:38 AM
Subject: RE: dlsw filtering/ETC...(CCBOOTCAMPL16)
> Hi
> I have another basic question....
>
> HostA-Ring1--TR-R1--IPCLOUD--R2-E0
>
> In the above case, lets say I want to advertise reachability for host A
> So on R1 I will do a "dlsw icanreach ...."
> Will I convert the mac-address of hostA to non-canonical format when
> I use it in the dlsw icanreach command?
>
> Thanks
> Sanjay
>
>
>
> If your doing MAC address stuff (filters and static ICANREACH statements)
> you gotta convert the Address to non cannonical format because DLSW+
> converts your ethernet MACs as soon as they leave the TB group and enter
> DLSW land.
>
> LLC2 is standardized across token ring, Ethernet and FDDI, so no weirdness
> there.
>
> AFAICT, with DLSW the two big things to watch out for are bit ordering in
> mac addresses and bridge loops in redundant topologies (no spanning tree
> saftey net here). you'll know when you have a bridge loop real quick
cause
> DLSW will quickly peg the CPU in your routers hehe.
>
> In regard to your questions about where frame translation is taking place,
> I'll make a stab at it, since I *think* I get how it works.
>
> With DLSW data link layer termination occurs locally. Any ring group you
> configure is just a construct to allow you to bind DLSW to a source route
> bridge, it's not used for communication between DLSW peers. why would you
> need a ring group if you are only talking to a transparent bridge?
>
> I belive that in normal (non passthru) operation once DLSW establises a
> circuit it just passes payload between peers, with the actual frame being
> created at the destination ring or bridge group construct DLSW is bound
to.
>
> -hope this helps some.
>
> -----Original Message-----
> From: nobody@groupstudy.com [mailto:nobody@groupstudy.com]On Behalf Of
> Mark Lewis
> Sent: Sunday, October 01, 2000 7:09 PM
> To: jeffsapiro@yahoo.com; ccielab@groupstudy.com
> Subject: Re: dlsw filtering/ETC...(CCBOOTCAMPL16)
>
>
>
> Well, again in the archives (so you probably know already) I saw a comment
> which was that the addresses are always sent by dlsw in non-canonical
format
> - doubt that helps though.
>
> Actually, I've got a related question:
>
> if I config. the following:
>
> r1:
>
> (NO source-bridge ring-group here)
> dlsw local-peer peer-id 10.1.1.1 (yep, I know it's better using a loopback
> as the id)
>
> dlsw remote 0 tcp 10.1.1.2
> dlsw bridge-group 1
>
> int e0
> ip address 10.1.1.1 255.255.255.0
> bridge-group 1
>
> bridge 1 protocol ieee
>
>
> r2:
>
> source-bridge ring-group 75
>
> dlsw local-peer peer-id 10.1.1.2
> dlsw remote 0 tcp 10.1.1.1 lf 1500
>
> int to0
> ip addr 10.1.1.2 255.255.255.0
> source-bridge 2 2 75
> source-bridge spanning
>
>
> Q: How do the frames pass from r1 to r2 ? Can sb. confirm (or set me
> straight) that they are translated from ethernet to token ring &
visa-versa
> ?
>
> Q2: If I add the command 'source-bridge ring-group 75' to r1 does it mean
> that the frames now cross the link & arrive at the destination
UNtranslated?
>
> I've seen so many config.s with one or the other (ie the 'source-bridge
> ring-group' command both on the ethernet router & not).
>
> In CCBootcamp lab 16, they have a q. about dlsw translational bridging and
> there is NO 'source-bridge ring-group' on the ETHERNET router.
> However, on one or two of the earlier CCBootcamp labs there was this
> command.
>
> I obviously need to pin this one down 'cos again it impacts crucially on
> filtering...
>
> Any ideas anybody?
>
> Mark
>
> P.S. Sorry Jeff, I've probably more confused than you!
>
>
> >From: Jeff Sapiro <jeffsapiro@yahoo.com>
> >Reply-To: Jeff Sapiro <jeffsapiro@yahoo.com>
> >To: ccielab@groupstudy.com
> >Subject: dlsw filtering/icanreach
> >Date: Sun, 1 Oct 2000 15:19:42 -0700 (PDT)
> >
> >Checking back in the archives I can't get a clear
> >answer about when to convert canonical addresses. any takers?
> >
This archive was generated by hypermail 2.1.4 : Thu Jun 13 2002 - 08:25:22 GMT-3