RE: Rip over isdn

From: ccie2be (ccie2be@nyc.rr.com)
Date: Tue Apr 05 2005 - 19:08:56 GMT-3


Simon,

I think of ip rip triggered as rip's equivalent to ospf's demand-circuit.
They both bring up the isdn link when there's a topology change. Besides
being specific to a particular routing protocol, that I can't think of any
other difference. Do you agree that these 2 commands are analogous?

But, getting back to the issue of ip rip triggered vs snapshot routing, if
the only LAB requirement is to not have the isdn link flap, both will
achieve that goal, don't you think?

Therefore, (assuming I don't forget the no peer neighbor-route command), in
the lab, I should always consider both options and if there's nothing that
implies I need to use snapshot routing, I can use ip rip triggered which is
easier to configure and holds less change for configuration mistakes.

What do you think?

Tim

-----Original Message-----
From: simon hart [mailto:simon.hart@btinternet.com]
Sent: Tuesday, April 05, 2005 5:43 PM
To: ccie2be; Group Study
Subject: RE: Rip over isdn

Tim,

I think I would not sue the ip rip triggered and plumb for snapshot routing.
A number of reasons

1. I think (but need to test) if you miss out the no peer neighbor route
the link will flap with ip rip trigger.

2. You cannot control call costs (if that is your primary consideration)
with ip rip triggered. If rip is unstable then the line will go up and down
sporadically. At least with snapshot you know when it is going to come up.

3. Relates to 2, snapshot provides more control. I would use ip rip trigger
when there is concern on fixed line bandwidth.

Simon

-----Original Message-----
From: nobody@groupstudy.com [mailto:nobody@groupstudy.com]On Behalf Of
ccie2be
Sent: 05 April 2005 22:24
To: Group Study
Subject: Rip over isdn

Hi guys,

If rip is running over an isdn link using ppp encap and the task states or
implies that that isdn link shouldn't flap, what are the config options that
should be considered?

Here's my list:

1, ip rip trigger

2. snapshot routing

Given these 2 options, what's the "best" practice? How do I decide which
option to use?

Also, I think other options such as dialer watch, backup int, etc aren't
good because then, in essence, the isdn link is down unless there's a
problem elsewhere in the network. Do you all agree?

Are there any options I should be considering?

TIA, Tim



This archive was generated by hypermail 2.1.4 : Tue May 03 2005 - 07:54:53 GMT-3