From: Jason T. Rohm (jtrohm@xxxxxxxxxxx)
Date: Mon Oct 02 2000 - 20:21:19 GMT-3
I removed the dialer portion of the snapshot server command and converted
the idle time to be 300 seconds (5 minutes). On my routers, the idle timeout
does not start until the 5 minutes of active snapshot routing are over... is
this normal?
The issue is still here however, the route stops updating as soon as the
active period is over but stays in the routing table. It will sit there
until the line hangs up (even though it says the route has aged 200+ seconds
at that point).
Any more ideas? I am going to down-grade my IOS to 11.2.x to see if it is a
version compatability issue.
-Jason
-----Original Message-----
From: nobody@groupstudy.com [mailto:nobody@groupstudy.com]On Behalf Of
Mark Lewis
Sent: Monday, October 02, 2000 4:57 PM
To: jtrohm@athenet.net; ccielab@groupstudy.com
Subject: Re: Snapshot Routes Timeout
Hi,
Try making the active time equal to the idle-time. Also, I'm guessing that
you have 'snapshot server DIALER' config.ed on one of the boxes(?). Try
taking the 'dialer' option off the snapshot server command.
Hope that helps,
Mark
>From: "Jason T. Rohm" <jtrohm@athenet.net>
>Reply-To: "Jason T. Rohm" <jtrohm@athenet.net>
>To: "CCIE List" <ccielab@groupstudy.com>
>Subject: Snapshot Routes Timeout
>Date: Mon, 2 Oct 2000 16:26:43 -0500
>
>
>I found a bunch of questions on this in the archive, but not any real
>answers... so here goes...
>
>I have snapshot routing setup on a DDR circuit (CCIE Prep Lab 7 for those
>of
>you who have it).
>
>The access-lists deny IPX Sap/Rip updates and permit all other IPX traffic.
>
>At the end of the quiet period, the snapshot client will dial the server
>and
>receive updates, however, as soon as the idle timer expires (active time +
>2
>mins), the line will drop (normal), but a show ipx route shows all snapshot
>learned routes as being in holddown, as soon as the hold down expires, the
>routes are aged out.
>
>A debug of the snapshot server and client (debug snapshot) shows that the
>snapshot server and client are atleast recognizing that routing updates are
>coming in.
>
>I have reviewed the code that CCIE Bootcamp sent me, and the only real
>difference that I see is that they have the ends (server/client) reversed
>which shouldn't really matter since it is not really a hub and spoke
>configuration.
>
>Anyone have ideas?
>
>-Jason
>
This archive was generated by hypermail 2.1.4 : Thu Jun 13 2002 - 08:25:23 GMT-3