From: Nigel Taylor (nigel_taylor@xxxxxxxxxxx)
Date: Sun Feb 04 2001 - 13:51:37 GMT-3
Frank,
OK... I'm satisfied now... After messing around with some of
the configs and watching some debugs I finally got this thing to do what I
wanted based not entirely on the requirements. As I mentioned in my earlier
post "why would one map r4 and r3 through r2. In the event of a failure of
the frame link to r2 then r3 and r4 will have no connectivity.
Based on what I did using a full mesh I'm now able to loose either frame
connection from r3 or r2 and maintain full routing and connectivity to the
whole network. I must admit this is truly the way to learn...
Now I'll follow the requirements of the lab and lock in those issues....
One other thought that I will attempt is implementing IX throughout the
network. Also ISIS comes to mind maybe even some tunneling and then
replacing my ISDN with Async dialup.. to see if this will provide a
solution.
Thanks for the extra pair of eyes... and words to all that listened...
Nigel..
----- Original Message -----
From: Frank Liang <liangfrank@hotmail.com>
To: <nigel_taylor@hotmail.com>; <Justin.Menga@computerland.co.nz>;
<fwells12@hotmail.com>; <ccielab@groupstudy.com>
Sent: Sunday, February 04, 2001 3:59 PM
Subject: Re: ISDN questions ...!
> I have just reviewed the lab reqirements, it says:
>
> "Configure Frame Relay using no subinterfaces. The Frame Relay cloud is a
full mesh, ensure that you use only the PVCs indicated on the topology
map".
>
> My understand is not to use all the dlci to make it fully mesh, just use
the dlci in the diagram to make it hub and spoke.
>
> The "ip ospf priority 0" in the spokes is to make sure the hub router r2
is the DR all the time.
>
> If you want to test it in fully mesh topology, the "ip ospf priority 0" is
not required. Any router can be the DR. In fact. this is what cause the
original problem you had, r3 and r4 can not form neighbor even the can ping
each other after you pull off r2. Because both r3 and r4 is DROTHER.
>
> FL
>
>
>
> >From: "Nigel Taylor" <nigel_taylor@hotmail.com>
> >To: "Frank Liang" <liangfrank@hotmail.com>,
<Justin.Menga@computerland.co.nz>, <fwells12@hotmail.com>,
<ccielab@groupstudy.com>
> >Subject: Re: ISDN questions ...!
> >Date: Sun, 4 Feb 2001 10:24:12 -0500
> >
> >Frank,
> > I noticed this... but I was under the impression that this
was a
> >typo... Based on the topology map I see that the drawing has R2
connecting
> >to each spoke. But, looking at config for R1 in the solution it makes
use
> >of the full mesh. In my config I'm using frame mappings from r3 to r2
and
> >r3 to r4. So with the config as they have it you cannot loose the frame
> >conn. from R2 to the cloud or R2 and R3 is isolated from the
network.......
> >
> >What I'm also trying to work out is the reasoning behind not using the
full
> >mesh to provide connectivity in the event that r2 looses the frame
circuit.
> >Also why make it the hub..? Wouldn't R4 be a better choice for being a
hub
> >since it physically connected to the rest of the network....?
> >
> >Another thing I observed was the frame cloud is configured as a broadcast
> >cloud but the "ip ospf priority" assigned to r3 and r4 uses a value of
"0".
> >This is doesn't allow a DR to elected on the segment when r2 lost it's
frame
> >connection to the cloud.....
> >
> >Well, I think these labs are a good resource but I must remember they are
> >"FREE". I'm I thinking to much here or does the lab requirements not
> >provide a clear outline as to the requirements....? Can someone read the
> >Lab sheet and tell me if it's clear as to how the solution provided makes
> >sense and completes the requirements..
> >
> >
> >Nigel..
> >
> >
> >
> >
> >----- Original Message -----
> >From: Frank Liang <liangfrank@hotmail.com>
> >To: <nigel_taylor@hotmail.com>; <Justin.Menga@computerland.co.nz>;
> ><fwells12@hotmail.com>; <ccielab@groupstudy.com>
> >Sent: Sunday, February 04, 2001 1:51 PM
> >Subject: Re: ISDN questions ...!
> >
> >
> > > Nigel,
> > > by looking at the fatkid 401 diagram, the frame cloud is hub and
spoke. If
> >you look at the solution, on R3, it maps r2 and r4 ip address to the same
> >dlci 102. R2 is a hub!
> > >
> > > FL
> > >
> > > >From: "Nigel Taylor" <nigel_taylor@hotmail.com>
> > > >Reply-To: "Nigel Taylor" <nigel_taylor@hotmail.com>
> > > >To: "Justin Menga" <Justin.Menga@computerland.co.nz>,
"fwells12"
> ><fwells12@hotmail.com>, <ccielab@groupstudy.com>
> > > >Subject: Re: ISDN questions ...!
> > > >Date: Sun, 4 Feb 2001 08:35:28 -0500
> > > >
> > > >Justin,
> > > > I'm using OSPF demand circuit and seeing the results I
> > > >described... I'm still trying to work through why R3 stays in a 2WAY
> > > >neighbor state with R4. In this scenario the frame cloud is full
mesh.
> > > >
> > > >Any thoughts....
> > > >
> > > >Thanks
> > > >Nigel..
> > > >
> > > >----- Original Message -----
> > > >From: Justin Menga <Justin.Menga@computerland.co.nz>
> > > >To: 'Nigel Taylor' <nigel_taylor@hotmail.com>; fwells12
> > > ><fwells12@hotmail.com>; <ccielab@groupstudy.com>
> > > >Sent: Sunday, February 04, 2001 6:07 AM
> > > >Subject: RE: ISDN questions ...!
> > > >
> > > >
> > > > > Well that just leaves OSPF demand circuit....
> > > > >
> > > > > Regards,
> > > > >
> > > > > Justin Menga CCIE #6640 MCSE+I CCSE
> > > > > WAN Specialist
> > > > > Computerland New Zealand
> > > > > PO Box 3631, Auckland
> > > > > DDI: (+64) 9 360 4864 Mobile: (+64) 25 349 599
> > > > > mailto: justin.menga@computerland.co.nz
> > > > >
> > > > >
> > > > >
> > > > > -----Original Message-----
> > > > > From: Nigel Taylor [mailto:nigel_taylor@hotmail.com]
> > > > > Sent: Sunday, February 04, 2001 8:58 PM
> > > > > To: fwells12; ccielab@groupstudy.com
> > > > > Subject: Re: ISDN questions ...!
> > > > >
> > > > >
> > > > > fwells12,
> > > > > The lab says I cannot use the backup interface
> >command,
> > > > > floating static's or watch groups....
> > > > >
> > > > > For more detail info.. it the 401 Adv. OSPF lab on www.fatkid.com.
> > > > >
> > > > > Nigel..
> > > > > ----- Original Message -----
> > > > > From: fwells12 <fwells12@hotmail.com>
> > > > > To: <ccielab@groupstudy.com>
> > > > > Sent: Sunday, February 04, 2001 2:42 AM
> > > > > Subject: Re: ISDN questions ...!
> > > > >
> > > > >
> > > > > > Make sure your idle-timeout is at least a minute, and that you
have
> >the
> > > > > > backup command on the serial interface that is being backed up.
Set
> >the
> > > > > > backup delay to be short coming up so routes start propagating
> > > > > immediately.
> > > > > >
> > > > > >
> > > > > > ----- Original Message -----
> > > > > > From: Nigel Taylor <nigel_taylor@hotmail.com>
> > > > > > To: Price, Jamie <JPrice@isgteam.com>; Cisco Group Study
> > > > > > <cisco@groupstudy.com>; CCIE_Lab Group Study
> ><ccielab@groupstudy.com>
> > > > > > Cc: Bryant Andrews <Bryant_Andrews@hotmail.com>
> > > > > > Sent: Saturday, February 03, 2001 11:25 PM
> > > > > > Subject: Re: ISDN questions ...!
> > > > > >
> > > > > >
> > > > > > > jamie,
> > > > > > > Ok, did some more testing and this thing will work
> >without
> > > >a
> > > > > > > name. I also changed the dialer idle-time-out value. The one
> >problem
> > > > > I'm
> > > > > > > seeing now is on router 2. When I pull the serial conn. from
> >router 3
> > > > > to
> > > > > > > the frame switch the bri line kick in and passes all the
routes.
> > > >However
> > > > > > > when I break the link from r2 to the frame switch,
nothing.....
> >R3
> > > > > never
> > > > > > > forms a full state with R4 and r2 and R3 doesn't pass routes
even
> > > >though
> > > > > > the
> > > > > > > bri line is up....
> > > > > > >
> > > > > > > Any thoughts..
> > > > > > >
> > > > > > > Nigel...
> > > > > > >
> > > > > > > ----- Original Message -----
> > > > > > > From: Price, Jamie <JPrice@isgteam.com>
> > > > > > > To: 'Nigel Taylor' <nigel_taylor@hotmail.com>; Cisco Group
Study
> > > > > > > <cisco@groupstudy.com>; CCIE_Lab Group Study
> ><ccielab@groupstudy.com>
> > > > > > > Cc: Bryant Andrews <Bryant_Andrews@hotmail.com>
> > > > > > > Sent: Sunday, February 04, 2001 1:49 AM
> > > > > > > Subject: RE: ISDN questions ...!
> > > > > > >
> > > > > > >
> > > > > > > > Just an FYI/gotcha
> > > > > > > >
> > > > > > > > I think setting the dialer idle-timeout to 1 causes
problems.
> > > > > > > >
> > > > > > > > We're all impatient and want the line to come down quickly
so we
> >can
> > > > > > prove
> > > > > > > > the solution but I've noticed that when the "dialer
> >idle-timeout" is
> > > > > > less
> > > > > > > > than the "dialer wait-for-carrier-time" then behaviour that
is
> > > > > > comparable
> > > > > > > to
> > > > > > > > what you mention here (flapping line) occurs.
> > > > > > > >
> > > > > > > > Jamie
> > > > > > > >
> > > > > > > > -----Original Message-----
> > > > > > > > From: Nigel Taylor [mailto:nigel_taylor@hotmail.com]
> > > > > > > > Sent: Sunday, February 04, 2001 12:38 AM
> > > > > > > > To: Cisco Group Study; CCIE_Lab Group Study
> > > > > > > > Cc: Bryant Andrews
> > > > > > > > Subject: ISDN questions ...!
> > > > > > > >
> > > > > > > >
> > > > > > > > I'm working through one of the labs at fatkid(401, Advanced
> >OSPF)
> > > >and
> > > > > =
> > > > > > > > have a couple of questions. The one part of the
configuration I
> > > >can't
> > > > > =
> > > > > > > > understand is the BRI's on R2 and R3. There is no
> >configuration
> > > >for
> > > > > =
> > > > > > > > the name of the calling device or called device. when I
tried
> >this
> > > > > the
> > > > > > =
> > > > > > > > routers would no connect, well at least not for more than a
> >second.
> > > >=
> > > > > > > > "debug ppp nego" showed called from <unknown> hang-up. Most
> > > >examples
> > > > > =
> > > > > > > > I've worked with makes use of the name option on the "dialer
> >map" =
> > > > > > > > command?
> > > > > > > >
> > > > > > > > My other question is specific to the "dialer idle-timeout 1
> >either".
> > > > > I
> > > > > > =
> > > > > > > > know what the command does but with one second.
> > > > > > > >
> > > > > > > >
> > > > > > > > Has anyone worked through this lab and got it working with
the
> > > > > solution
> > > > > > =
> > > > > > > > on the site... What was your observations on the bri line.
> > > > > > > > Mines keep bouncing up and down...
> > > > > > > >
> > > > > > > > TIA
> > > > > > > >
> > > > > > > > Nigel..
> > > > > > > >
> > > > > > > >
> > > > > > > >
This archive was generated by hypermail 2.1.4 : Thu Jun 13 2002 - 10:28:36 GMT-3