RE: ccbootcamp 8 - bgp always-compare-med

From: rsevier (rsevier@xxxxxxxxxxxxxxxxxxx)
Date: Sun May 20 2001 - 21:32:37 GMT-3


   
I meant the redistribution session in my last post, not bgp session.

-----Original Message-----
From: nobody@groupstudy.com [mailto:nobody@groupstudy.com]On Behalf Of
rsevier
Sent: Sunday, May 20, 2001 4:56 PM
To: Roman Rodichev; kenyeo@email.com; bob.chahal@ntlworld.com;
ccielab@groupstudy.com
Subject: RE: ccbootcamp 8 - bgp always-compare-med

Ken - check your filters on your redistribution you have bad routes in your
BGP table. Look at the next hop for the non-optimal route.

I think that the concept for redundancy is valid and another consideration
that I would like to offer is to keep an eye on the extra bgp session that
needs to be maintained. You would need to understand all the consequences
of any line failure to not inject bad routes into your network.

As for the lab, both ways satisfies the requirements in my opinion, I was
just wondering if I was incorrect.

Thanks for the input,
Raymond

-----Original Message-----
From: Roman Rodichev [mailto:rodic000@hotmail.com]
Sent: Sunday, May 20, 2001 4:09 PM
To: kenyeo@email.com; bob.chahal@ntlworld.com;
rsevier@zealousolutions.com; ccielab@groupstudy.com
Subject: RE: ccbootcamp 8 - bgp always-compare-med

Ken, you are right, in a real world scenario (for me, the ccie lab is a real
world scenario) you would not want to depend on one eBGP router doing
redistribution of all external routes (even the ones from other eBGP routers
in the network) into IGP. What do others think?

>From: "Ken Yeo" <kenyeo@email.com>
>Reply-To: "Ken Yeo" <kenyeo@email.com>
>To: "Bob Chahal" <bob.chahal@ntlworld.com>, "Roman Rodichev"
><rodic000@hotmail.com>, <rsevier@zealousolutions.com>,
><ccielab@groupstudy.com>
>Subject: RE: ccbootcamp 8 - bgp always-compare-med
>Date: Sun, 20 May 2001 17:56:46 -0500
>
>How about R6 is crashed.
>
>However the instruction didn't say any requirement for redundancy, is the
>step is not really needed??
>
>Thanks,
>Ken Yeo
>
>-----Original Message-----
>From: Bob Chahal [mailto:bob.chahal@ntlworld.com]
>Sent: Sunday, May 20, 2001 3:44 PM
>To: Roman Rodichev; rsevier@zealousolutions.com; kenyeo@email.com;
>ccielab@groupstudy.com
>Subject: Re: ccbootcamp 8 - bgp always-compare-med
>
>
>Yea but it's a good question because how would the IBGP connection actually
>be brought down. The physical link between 6 and 5 but then the ISDN kicks
>in. Or the physical link between 5 and 1 but if that happened then R7 would
>never be able to get to the OSPF routes anyway. Of course the IBGP link
>could be down because of a misconfiguration and then your point is valid.
>
>Good question though. Any one else any thoughts?
>
>
>----- Original Message -----
>From: "Roman Rodichev" <rodic000@hotmail.com>
>To: <rsevier@zealousolutions.com>; <kenyeo@email.com>;
><ccielab@groupstudy.com>
>Sent: Sunday, May 20, 2001 8:56 PM
>Subject: RE: ccbootcamp 8 - bgp always-compare-med
>
>
> > Raymond, it makes sense to do mutual redistribution on both R1 and R6.
>What
> > happens if IBGP is brought down. AS1 (i forgot the number, that's the
>one
> > with R7) will have no way of knowing OSPF routes.
> >
> > i.e. let's say R6 is rebooted, how will R7 know about Ethernet
>connection
>on
> > R5?
> >
> >
> > >From: "rsevier" <rsevier@zealousolutions.com>
> > >To: "Roman Rodichev" <rodic000@hotmail.com>, <kenyeo@email.com>,
> > ><ccielab@groupstudy.com>
> > >Subject: RE: ccbootcamp 8 - bgp always-compare-med
> > >Date: Sun, 20 May 2001 09:30:07 -0700
> > >
> > >I agree that bgp always-compare-med is not needed. however, I have
>another
> > >one for you. Does there need to be redistribution on r1 of ospf into
>bgp.
> > >I know that redistribution of bgp into ospf is needed. I have just
> > >completed 8a and didn't find a reason to for the redistribution on r1
>of
> > >ospf into bgp. Our solution worked fine with out it because of the it
>is
> > >being redistributed on r6. Can I get any input as to why it is in the
> > >answers from Marc.
> > >
> > >thanks in advance
> > >Raymond
> > >
> > >-----Original Message-----
> > >From: nobody@groupstudy.com [mailto:nobody@groupstudy.com]On Behalf Of
> > >Roman Rodichev
> > >Sent: Saturday, May 19, 2001 11:31 PM
> > >To: kenyeo@email.com; ccielab@groupstudy.com
> > >Subject: Re: ccbootcamp 8 - bgp always-compare-med
> > >
> > >
> > >I thought so too
> > >
> > >
> > > >From: "Ken Yeo" <kenyeo@email.com>
> > > >Reply-To: "Ken Yeo" <kenyeo@email.com>
> > > >To: <ccielab@groupstudy.com>
> > > >Subject: ccbootcamp 8 - bgp always-compare-med
> > > >Date: Sun, 20 May 2001 01:06:44 -0500
> > > >
> > > >on r1
> > > >
> > > >I believe bgp always-compare-med is not needed.
> > > >
> > > >Anyone can comfirm that?
> > > >
> > > >Thanks!
> > > >Ken Yeo
> > > >**Please read:http://www.groupstudy.com/list/posting.html



This archive was generated by hypermail 2.1.4 : Thu Jun 13 2002 - 10:30:47 GMT-3