From: P729 (p729@xxxxxxx)
Date: Fri Jul 26 2002 - 01:03:43 GMT-3
Yes. Multiple routing protocols with overlapping major nets and
redistribution in the same router can be oh so much fun at times. As a
general rule, the way I dealt with it during my preparation was to analyze
each one from its own perspective, choose my core routing protocol, control
which interfaces participated in each protocol (to the extent/if possible)
and strictly filter my redistributions so that only networks that were
"native" to my edge protocol(s) were redistributed into my core.
Admittedly, sometimes it's hard to decide which is "core" and which is
"edge"--the Fatkid mondo redistribution lab comes to mind. There may not
always be one right way to do it--so try 'em all! <g>
Regards,
Mas Kato
https://ecardfile.com/id/mkato
----- Original Message -----
From: "Bruce Williams" <bruce@williamsnetworking.com>
To: "P729" <p729@cox.net>
Cc: <ccielab@groupstudy.com>
Sent: Thursday, July 25, 2002 5:59 AM
Subject: RE: OSPF keeps bringing up DDR across asyn line
> Okay, then the solution would be to filter the demand circuit route so it
is
> not redistributed into the "redistributed protocol". Correct?
>
> Bruce
>
> -----Original Message-----
> From: P729 [mailto:p729@cox.net]
> Sent: Thursday, July 25, 2002 1:18 AM
> To: Bruce Williams
> Cc: ccielab@groupstudy.com
> Subject: Re: OSPF keeps bringing up DDR across asyn line
>
>
> The problem occurs when the "demand circuit" is also known to the
> redistributed protocol. When it flaps, the redistributed protocol
withdraws
> the route and round-and-round she goes...
>
> Regards,
>
> Mas Kato
> https://ecardfile.com/id/mkato
> ----- Original Message -----
> From: "Bruce Williams" <bruce@williamsnetworking.com>
> To: "Joe A" <GroupStudy@comcast.net>; "'Donny MATEO'"
> <donny.mateo@sg.ca-indosuez.com>
> Cc: <ccielab@groupstudy.com>; "'Ivan'" <limmt@lycosasia.com>;
> <nobody@groupstudy.com>
> Sent: Wednesday, July 24, 2002 5:36 AM
> Subject: RE: OSPF keeps bringing up DDR across asyn line
>
>
> > I have a question about this. Once the redistributed routing protocol is
> > redistributed, will it keep on sending LSAs about the redistributed
routes
> > every time the redistributed routing protocol sends and update. I would
> > assume that it would not have to send LSAs every time because the
> > information is not new. LSAs are sent when something changes. If that is
> not
> > correct, please explain.
> >
> > Bruce
> >
> >
> > -----Original Message-----
> > From: nobody@groupstudy.com [mailto:nobody@groupstudy.com]On Behalf Of
> > Joe A
> > Sent: Tuesday, July 23, 2002 11:51 PM
> > To: 'Donny MATEO'
> > Cc: ccielab@groupstudy.com; 'Ivan'; nobody@groupstudy.com
> > Subject: RE: OSPF keeps bringing up DDR across asyn line
> >
> >
> > The issue in using a demand circuit in area 0 is that EVERYTHING must
get
> > flooded into area 0. Check out page 801 in the Solie book, Area 0
> > Guidelines. Also check out page page 498 paragraph 3 in the same book.
> >
> > Redistributing into OSPF certainly can cause a demand circuit to flap.
> > Remember, a demand circuit suppresses hellos, not LSAs. And all LSAs
get
> > flooded into area 0, so the link is going to flap up and down in 'sync'
> with
> > the redistributed routing protocol's periodic updates.
> >
> > - Joe
> >
> > -----Original Message-----
> > From: Donny MATEO [mailto:donny.mateo@sg.ca-indosuez.com]
> > Sent: Tuesday, July 23, 2002 12:43 AM
> > To: Joe A
> > Cc: ccielab@groupstudy.com; 'Ivan'; nobody@groupstudy.com
> > Subject: RE: OSPF keeps bringing up DDR across asyn line
> >
> >
> >
> > I'm affraid I dont' quite understand this. I have a perfectly running
> > demand circuit under area 0 and it's not a stubby area either.
> > What's the issue of using demand circuit under area 0 ?As far as I know
is
> > that if your topology is stable that it shouldn't have triggered the
line.
> > If changing your area into totally stub area solve the problem , perhaps
> > it's because you have an unstable topology in the first place.
> >
> > Donny
> >
> >
> >
> > Joe A
> > <GroupStudy@comca To: "'Ivan'"
> > <limmt@lycosasia.com>
> > st.net> cc:
> > ccielab@groupstudy.com
> > Sent by: Subject: RE: OSPF keeps
> > bringing up DDR across asyn line
> > nobody@groupstudy
> > .com
> >
> >
> > 23-07-2002 12:09
> > Please respond to
> > Joe A
> >
> >
> >
> >
> >
> >
> > You could've also done a route filter to stop the distance vector
protocol
> > from causing an LSA. Another message referenced an unstable IGP, but
> don't
> > think that is always the case - the LSA is probably generated by the
> normal
> > periodic advertisments that would still happen in a completely stable
> > environment.
> >
> > -----Original Message-----
> > From: Ivan [mailto:limmt@lycosasia.com]
> > Sent: Monday, July 22, 2002 11:30 PM
> > To: GroupStudy@comcast.net
> > Cc: ccielab@groupstudy.com
> > Subject: RE: OSPF keeps bringing up DDR across asyn line
> >
> >
> > Hi,
> >
> > Yes, its correct.
> > I've changed the area 10 where the dialer interface is in into a totally
> > stubby area and that solves the issues.
> >
> > Thanks
> > Ivan Lim
> >
> >
> > -----Original Message-----
> > From: "Joe A" <GroupStudy@comcast.net>
> > Sent: Tuesday, July 23, 2002 10:50 AM
> > To: "'Ivan'" <limmt@lycosasia.com>, "ccielab@groupstudy.com"
> > <ccielab@groupstudy.com>
> > Subject: RE: OSPF keeps bringing up DDR across asyn line
> >
> > It might be because you are redistributing a distance vector protocol
into
> > OSPF, which generates an LSA over the link at each update period.
> >
> > -----Original Message-----
> > From: nobody@groupstudy.com [mailto:nobody@groupstudy.com]On Behalf Of
> > Ivan
> > Sent: Monday, July 22, 2002 8:16 AM
> > To: ccielab@groupstudy.com
> > Subject: OSPF keeps bringing up DDR across asyn line
> >
> >
> > Hi,
> >
> > I've configured AUX back to back to simulate DDR across ISDN.
> > I've also configure it as a IP OSPF Demand Circuit.
> >
> > However OSPF continues to bring up the link.
> > The link is under an OSPF area 10.
> >
> > The debug dialer shows the reason for the link being bring up:
> > -------------------------------------------------------------
> > Mar 4 03:59:57.249 EST: As1 DDR: rotor dialout [priority]
> > Mar 4 03:59:57.253 EST: As1 DDR: Dialing cause ip (s=110.99.25.2,
> > d=224.0.0.5)
> > Mar 4 03:59:57.257 EST: As1 DDR: Attempting to dial 1234
> > Mar 4 03:59:57.257 EST: CHAT1: Attempting async line dialer script
> > Mar 4 03:59:57.261 EST: CHAT1: Dialing using Modem script: isdn &
System
> > script
> > : none
> > Mar 4 03:59:57.269 EST: CHAT1: process started
> > Mar 4 03:59:57.269 EST: CHAT1: Asserting DTR
> > Mar 4 03:59:58.197 EST: CHAT1: Chat script isdn started
> > Mar 4 03:59:58.201 EST: CHAT1: Chat script isdn finished, status =
> Success
> > MU-R2#
> >
> > ----------------------------------------------------------------
> >
> > Since the hello is suppressed over the point to point dialer link,
> > I'm not sure what OSPF is sending out on the all router multicast
address
> > across the dialer interface. The other reason i can think of is when a
> > topology change that trigger the link but there is none.
> >
> > Appreciate if anyone can advice on this.
> >
> > Thanks
> > Ivan Lim
> >
> > ------------------------------------------------------------------------
> > cOntact @ Lycos <http://contact.lycosasia.com>
> > = 10MB for email and filestore + lots of other goodies...
This archive was generated by hypermail 2.1.4 : Sat Sep 07 2002 - 19:36:44 GMT-3