From: Eric Hauptman (ehauptma@xxxxxxxx)
Date: Sat Apr 14 2001 - 04:51:16 GMT-3
I would like to thank all those who responded to my point-to-point serial link
problem. I was informed tonight that our provider, AT&T, had a loop somewhere
in the telco pointing back to
one end of the site, thus leaving one side up/down & the other side up/up.
However, going forward, I'm perplexed about the best way to back this circuit
up using my current ISDN B-channels on the 2610 router. I had it setup to use
floating static routes with higher AD for the BRI channels, which works fine as
long as the serial drops on BOTH sides. What happened yesterday is that when
one side went down the floating static kicked in and the BRI was dialed.
However, the other side's serial was still up/up, meaning the return traffic
was still being sent to the serial line, not the ISDN line, hence traffic
couldn't pass. We manaully solved the problem by removing the floating statics
and putting all static routes on both sides pointing to the BRI lines.
Does anyone have any ideas on the best way to deal with the situation in the
future? I assume some sort of dynamic routing protocal might be the answer,
but I was hoping to run this link without the added overhead of a routing
protocol. Thanks.
Eric H.
Bob Dixon wrote:
> Maybe a stupid question, but are you sure that your physical wiring is in
> place as it should be? In other words, maybe routerA's serial0 is cabled to
> a live circuit terminating on an unknown device that is exchanging
> keepalives as it should. This would mean serial0 on routerA shows up/up. Of
> course, it is highly unlikely that you could ping across the wan link in
> this scenario. routerB's serial 0 could be cabled to a cicuit that is open
> on the other end. That means serial 0 on routerB shows up/down. Just a
> thought.
>
> >From a perspective that may sound more like PC troubleshooting than router
> troubleshooting, you could try powering down the router that shows up/down,
> re-seat the WIC card, and then boot the router. I just turned up some t1's
> and t3's the other day. We had a weird problem where the t1 wic showed
> up/down (keepalive was set) and the ILEC could not loop the built-in csu on
> the WIC. After re-seating the WIC, the ILEC could latch the WIC and after
> dropping the loop, the circuit came up/up and ran clean.
>
> Bob
>
> ----- Original Message -----
> From: Prashanth <prashanthcm@yahoo.com>
> To: Curtis Call <curtiscall@home.com>; <ccielab@groupstudy.com>
> Sent: Saturday, April 14, 2001 2:26 AM
> Subject: RE: Point to Point Serial link
>
> > possible to send us the config?
> >
> >
> > --- Curtis Call <curtiscall@home.com> wrote:
> > > It might be that one side is expecting keepalives
> > > and the other is not
> > > expecting nor sending them.
> > >
> > >
> > > >-----Original Message-----
> > > >From: Eric Hauptman [mailto:ehauptma@home.com]
> > > >Sent: Friday, April 13, 2001 2:25 AM
> > > >To: Groupstudy
> > > >Subject: OT: Point to Point Serial link
> > > >
> > > >
> > > >I have a production network where one side of a
> > > point to point serial
> > > >link is showing up/down and the other side is
> > > showing up/up. I thought
> > > >that if one side went down they would both go down.
> > > I know on frame
> > > >relay one can be up/up and the other down, but I
> > > thought on a point to
> > > >point this would not be the case. Am I missing
> > > something basic here? I
> > > >am running the T1s into 2610 routers with built in
> > > WIC-T1-CSUs.
> > > >
> > > >Thanks.
> > > >
> > > >Eric
> > >
This archive was generated by hypermail 2.1.4 : Thu Jun 13 2002 - 10:29:45 GMT-3