From: Derek Small (Fuse) (dwsmall@xxxxxxxxxx)
Date: Tue Apr 11 2000 - 22:11:18 GMT-3
Just an FYI for everyone. The solution that Xuping posted did solve the
Snapshot routing problem. I'll have the completed solution up on Fatkid.com
in a week or two.
I don't want to abuse this group, but would anyone else be interested in
some more stuff like this (giving away time slots for solutions?) Seems like
it might be a win-win.
Paul, any problems with this?
Derek Small
dwsmall@fatkid.com
----- Original Message -----
From: xuping <cnmiecxu@public.bta.net.cn>
To: Derek Small (Fuse) <dwsmall@fatkid.com>
Cc: <ccielab@groupstudy.com>
Sent: Sunday, April 09, 2000 6:36 AM
Subject: My Answer: Problem with snapshot routing.
> >The solution for the SnapShot routing scenario is broken. The server
> >router, R2, is not sending or receiving RIP updates when snapshot brings
> the
> >BRI up. What is wrong?
> >
> >Hint, I have removed the dialer option from the "snapshot server 5"
command
> >under the BRI, that didn't do it.
>
>
> The option "dialer" after "snapshot server 5" is tell the snapshot
> process by which method to learn who is the snapshot client (the peer)
>
> In your R2
>
> interface BRI0
> ip address 172.1.4.1 255.255.255.0
> encapsulation ppp
> isdn spid1 0835866101 8358661
> isdn spid2 0835866301 8358663
> dialer idle-timeout 300
> dialer map ip 172.1.4.2 broadcast 8358662
> dialer load-threshold 1 either
> dialer-group 1
> snapshot server 5 dialer
> no fair-queue
> ppp multilink
> no shutdown
>
> when you delete the "dialer" option, the snapshot process will have no
idea
> about who make the connection, the snapshot process will try to bring up
> another connection using "dialer map ip 172.1.4.2 broadcast 8358662"
> statement for its routing purpose instead of utilizing the already
connected
> channel.
>
> solution 1:
> give snapshot process some other methed to identify the connection, say,
> give dialer map a remote-connection-name, and uses chap or pap to identify
> the connection by connection name. like this:
>
> username R3 password 0 cisco
>
> interface BRI0
> ip address 172.1.4.1 255.255.255.0
> encapsulation ppp
> isdn spid1 0835866101 8358661
> isdn spid2 0835866301 8358663
> dialer idle-timeout 300
> dialer map ip 172.1.4.2 name R3 broadcast 8358662
> dialer load-threshold 1 either
> dialer-group 1
> snapshot server 5
> no fair-queue
> ppp multilink
> ppp auth chap
> no shutdown
>
> solution 2:
>
> use dialer profile, unlike physical interface may defind mutiple dialer
map
> statement (making a physical BRI interface to be a point-to-multipoint
> style). a dialer profile define a pure point-to-point connection (since
> only one dialer string may defined in a dialer), snapshot process will not
> confused in such a enviroement.
>
> interface BRI0
> no ip address
> encapsulation ppp
> isdn spid1 0835866101 8358661
> isdn spid2 0835866301 8358663
> dialer pool-member 1
> ppp multilink
>
> interface Dialer 1
> ip address 172.1.4.1 255.255.255.0
> dialer idle-timeout 300
> dialer remote-name R3
> dialer pool 1
> dialer string 8358662
> dialer load-threshold 1 either
> dialer-group 1
> snapshot server 5
> ppp multilink
>
> I have repeat the problem and test both configs on my test bed(with
> defferent ISDN No.), both works.
>
> I suppose if your ISDN switch can send correct caller ID, your config will
> work without any change when delete "dialer" option.
>
> xuping
> cnmiecxu@public.bta.net.cn
>
>
>
>
>
>
This archive was generated by hypermail 2.1.4 : Thu Jun 13 2002 - 08:23:13 GMT-3