From: iron_tri (iron_tri@msn.com)
Date: Sat Jul 12 2003 - 15:41:39 GMT-3
It worked for me. OSPF made its topology changes during office hours as
required in the lab. Dialer Watch, however, served its purpose in the lab
by backing up the Frame Relay circuit. I used a combination of the two to
satisfy the lab requirement. Maybe I interpreted the lab wrong, but I think
I met the requirements by using both Demand Circuit and Dialer Watch. I
didn't interpret the question as stating that Dialer Watch couldn't bring up
the link during certain hours. I interpreted the requirement as stating
that the Frame circuit should be backed up. I don't disagree with the way
that Dialer Watch functions, but I used both to satisfy the requirement.
I guess we have different interpretations of the required solution.
----- Original Message -----
From: "SHARMA,MOHIT (HP-Germany,ex1)" <mohit.sharma@hp.com>
To: "'iron_tri'" <iron_tri@msn.com>; <ccielab@groupstudy.com>
Sent: Saturday, July 12, 2003 12:29 PM
Subject: RE: Dialer watch
> Hi,
>
> I did try this in the lab. However because dialer-watch is NOT dependent
on
> the interesting traffic to dial ( it dials as soon as the watched traffic
> disappears from the routing table), defining time-range with the
dialer-list
> will have no effect. It will only reset the idle-timwout if there is
> intersting traffic.
>
> SO I really dont see how you can make dialer-watch work here.
>
> With ospf Demand circuit, if you use time-range to let the ospf pass, and
> the ospf topology changes during the time when the acl is inactive, it
will
> bring down the ospf demand circuit ( although that may be fine as per the
> lab requirement, as it says that the link should only come up during
office
> hours and it does not care if the topology is not complete during
non-office
> hours))
>
> PLease corrct me if I am wrong.
>
> Thanx,
>
> Mohit.
> -----Original Message-----
> From: iron_tri [mailto:iron_tri@msn.com]
> Sent: Saturday, July 12, 2003 7:35 PM
> To: SHARMA,MOHIT (HP-Germany,ex1); ccielab@groupstudy.com
> Subject: Re: Dialer watch
>
>
> Why wouldn't Dialer Watch work in this scenario? You might have to use a
> combination of Dialer Watch and Demand Circuit (depending upon lab
> requirements and protocol), but it would all work with a Time Range based
> access list just fine as I have done this scenario in a lab environment.
As
> long as the time you are trying to dial matches the time range in the
access
> list, it will work correctly. If you are running Demand Circuit, then
OSPF
> will take care of topology changes, and Dialer Watch can be used to back
up
> the Frame circuit. Even if you aren't using OSPF and Demand Circuit,
Dialer
> Watch and a Time Range based access list will work. Define the extended
> access list, use the time range parameters, make sure you have allowed for
> interesting traffic, and apply the access list under the dialer list
command
> as your interesting traffic.
>
> HTH
> ----- Original Message -----
> From: "SHARMA,MOHIT (HP-Germany,ex1)" <mohit.sharma@hp.com>
> To: "'Jonathan V Hays'" <jhays@jtan.com>; <ccielab@groupstudy.com>
> Sent: Saturday, July 12, 2003 9:40 AM
> Subject: RE: Dialer watch
>
>
> > Hi All,
> >
> > Sorry about replying to my own mail. I overlooked the fact that its as
> ospf
> > P2M network, and hence the /32 route in this case.
> > ( Thanx Tomasz)
> > Otherwise can somebody please suggest what would be the recommended
method
> > to use, when the requirement says that, you should backup the
frame-relay
> > link and use time-range based dialer-list.
> >
> > My thoughts-
> > Dialer-watch cannot be used, as does not take care of dialer-list so no
> > time-range.
> > Backup interface cant be used as if its a physical int., it wont know
that
> > serial is down.
> > ospf demand circuit or floating static are the only choices left, based
on
> > the conext. Then again ospf demand circuit would break down because of
the
> > time circuit.
> >
> > Any comments, would be appreciate highly.
> >
> > Thanks as always.
> >
> > Smiles,
> >
> > Mohit.
> >
> > -----Original Message-----
> > From: SHARMA,MOHIT (HP-Germany,ex1) [mailto:mohit.sharma@hp.com]
> > Sent: Saturday, July 12, 2003 5:32 PM
> > To: 'Jonathan V Hays'; ccielab@groupstudy.com
> > Subject: RE: Dialer watch
> >
> >
> > HI Jonathan,
> >
> > Thanx for the reply.
> > I am a bit confused here, since the host is the part of the connected
> > subnet, you will not see this host router in the routing table anyways.
So
> > how will dialer watch work here??? Since the first pre-requisite is that
> the
> > watched route should be exactly in the routing table.
> >
> > Thanx,
> >
> > Mohit.
> >
> > -----Original Message-----
> > From: Jonathan V Hays [mailto:jhays@jtan.com]
> > Sent: Saturday, July 12, 2003 5:27 PM
> > To: 'SHARMA,MOHIT (HP-Germany,ex1)'; ccielab@groupstudy.com
> > Subject: RE: Dialer watch
> >
> >
> > So don't watch the subnet, watch host on the opposite end of the link
> > instead.
> >
> > dialer-watch list 10 ip 10.1.1.1 255.255.255.255
> >
> > A good way to test is to remove the VC from the frame switch.
> >
> > HTH
> >
> > > -----Original Message-----
> > > From: nobody@groupstudy.com [mailto:nobody@groupstudy.com] On
> > > Behalf Of SHARMA,MOHIT (HP-Germany,ex1)
> > > Sent: Saturday, July 12, 2003 11:02 AM
> > > To: ccielab@groupstudy.com
> > > Subject: Dialer watch
> > >
> > >
> > > HI All,
> > >
> > > Has any one attemped the lab 36 from ipexpert. IN the Dial
> > > section, they
> > > have a requirement that ISDN should come up when FR fails.
> > > They r using
> > > dialer-watch, but they use the IP address of the serial
> > > interface of the
> > > other router in the dialer-watch list. How will this work,
> > > since this subnet
> > > is directly connected????
> > >
> > > Also has somebody ever tried using Dialer-watch with
> > > time-range. I believed
> > > that since dialer-watch is not based on interesting traffic,
> > > time-range will
> > > have no effect in dialer-list used with dialer-watch.
> > >
> > > Can somebody please sehd some light here.
> > >
> > > Thanks,
> > >
> > > Mohit.
> > >
> > >
> > > ______________________________________________________________
> > > _________
> > > You are subscribed to the GroupStudy.com CCIE R&S Discussion Group.
> > >
> > > Subscription information may be found at:
> > > http://www.groupstudy.com/list/CCIELab.html
> >
> >
> > _______________________________________________________________________
> > You are subscribed to the GroupStudy.com CCIE R&S Discussion Group.
> >
> > Subscription information may be found at:
> > http://www.groupstudy.com/list/CCIELab.html
> >
> >
> > _______________________________________________________________________
> > You are subscribed to the GroupStudy.com CCIE R&S Discussion Group.
> >
> > Subscription information may be found at:
> > http://www.groupstudy.com/list/CCIELab.html
>
>
> _______________________________________________________________________
> You are subscribed to the GroupStudy.com CCIE R&S Discussion Group.
>
> Subscription information may be found at:
> http://www.groupstudy.com/list/CCIELab.html
This archive was generated by hypermail 2.1.4 : Wed Aug 06 2003 - 06:52:37 GMT-3