I've been working on a home lab scenario involving ISIS and it's been
driving me crazy. The behavior appears to be a split horizion issue even
though ISIS(like OSPF) is a link state protocol and shouldn't have a problem
with this. The scenario involves a hub router R4 and two spoke routers
r2,r3.
James,
I hate to give such a poorly researched answer, but I did quite a bit of
experimenting with ISIS a few months back. I don't have time to verify this
for sure right now, but here is what I remember:
ISIS knows nothing of NBMA. It only know serial point-to-point and
multi-access LANs.
Scott,
Thanks for your help. I had already tried the frame-relay map commands with
the broadcast ending. It didn't seem to work then either :(
You are correct in that on a multi-access LAN all the routers have to see
each other and these two spoke routers are definitely not. The DIS is not
quite
James,
OK, you tried the map statements with the broadcast keyword on the hub. Did
you add the broadcast keyword on the map statements between the spoke sites?
Regards,
Scott
James Haynes wrote:
Scott,
Thanks for your help. I had already tried the frame-relay map
commands with
the
Yes, the spoke routers frame-relay map statements all contain the broadcast
keyword as well. However, the routes are still not propagating. If I use the
show isis topology command on the spoke routers I can see the other spoke
router and the correct metric to it, but not the next hop, interface,
James,
I notice that you are using the same DLCI (104) between the spokes as the
one you use to get to the hub. Is this right? I was under the impression
that the non-NBMA support aspect of ISIS required either p-t-p subinterfaces
or fully meshed mulipoint.
Scott
James Haynes wrote:
Yes,
Add 'frame-relay map ip ' statements as well as the 'frame-relay map
clns' statements and it will resolve your problem.
-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]
Sent: Thursday, January 24, 2002 4:11 PM
To: [EMAIL PROTECTED]
Subject: Re: ISIS [7:33096]
James
Scott,
I have been reading many of the troubleshooting articles about ISIS at the
Cisco web site and every WAN scenario indeed involves a fully meshed
multipoint frame relay configuration or a p-t-p subinterface setup. What I'm
trying to do is not going to work at all. I thank you for your
Jeff,
That was my initial suggestion, which James said didn't cut it. Would you
add your thoughts on the full mesh issue?
Thanks,
Scott
Message Posted at:
http://www.groupstudy.com/form/read.php?f=7i=33128t=33096
--
FAQ, list archives, and
James,
You are quite welcome. I have recently begun to worry that all of my hard
work to fully understand ISIS was wasting away. I need to go back and
refresh every couple of months or all is lost because I don't get any real
world interaction with ISIS (at least not yet). So in that regard,
Mush is definitely what it feels like.My work with ISIS is limited to
playing with it in a lab as well. I'm going to fully mesh this later just so
I can see it work. Thanks again.
--
James Haynes
Network Architect
Cendant IT
A+,MCSE,CCNA,CCDA,CCNP,CCDP,
CQS-SNA/IPSS
s vermill wrote in message
11 matches
Mail list logo
