From: Derek Fage (DerekF@xxxxxxxxxxx)
Date: Tue Jul 13 1999 - 12:39:11 GMT-3
I figured it out (with a little help)
The problem was that the router with the on-demand circuit was also running
IGRP and performing mutual redistribution.
Due to the nature of the subnetting in use, the BRI interface was included
in the IGRP network statement, even though I had set it to passive.
The BRI going up/donw was causing IGRP to see a change and inject that into
OSPF as a type 5 LSA.
Using a route-map on the ospf redistribute igrp command to be a bit more
specific solved the problem.
Derek...
> -----Original Message-----
> From: Ron Trunk [SMTP:rtrunk@xatlantic.com]
> Sent: 13 July 1999 16:19
> To: ccielab@groupstudy.com
> Subject: Re: OSPF On-demand
>
> Derek
>
> You could configure your dialer list to define OSPF LSAs as uninteresting.
> This will stop LSAs from bringing up the ISDN line
>
> Ron
>
> -----Original Message-----
> From: Derek Fage <DerekF@itexjsy.com>
> To: ccielab@groupstudy.com <ccielab@groupstudy.com>
> Date: Tuesday, July 13, 1999 10:31 AM
> Subject: OSPF On-demand
>
>
> >I've got a bit of a problem with OSPF On-demand circuits.
> >
> >What I want to do is to use an OSPF On-demand circuit using ISDN to act
> as
> >the backup to ethernet interfaces between two routers.
> >
> >Due to the fact that it is possible for one ethernet interface to go down
> >and the other to remain up, I cannot use the backup interface command, so
> I
> >have configured the ethernet interfaces and the BRI interfaces in the
> same
> >area.
> >
> >When doing a debug whilst the ISDN is up I see messages indicating that
> the
> >demand circuit is running as expected:
> >15:47:56: OSPF: Dead event ignored for 137.20.224.5 on demand circuit
> BRI0
> >
> >The problem is that after two minutes the idle timer kicks in and the BRI
> >drops. At this stage it would appear that an LSA is generated indicating
> the
> >link state change and flooded out of all of the interfaces - including
> the
> >ISDN. This in turn brings the line up again and so it continues.
> >
> >Does anybody have any suggestions as to how to work around this ?
> >
> >Derek...
This archive was generated by hypermail 2.1.4 : Thu Jun 13 2002 - 08:21:42 GMT-3