Re: lesson from failure

From: Dillon Yang (dillony@gmail.com)
Date: Mon Jul 25 2005 - 21:39:01 GMT-3


Hi, Brian:

  Yes, it will not cost me more than 1 minute in my racks, but not in the real lab. When I finished the relevant "isdn switch-type" and "isdn spid" on each router, I began the verification from L2 step by step. After I noticed the command could not generate a successful feedback, I stoped to check where the problem was. The symptom is R5 can initiate the link while R3 can not with the same config lines. Surprising thing was that R5 showed the message "software-error" and dumped a lot of values of buffers and registers and rebooted eventually. When the proctor saw this, he was so kind to stop me to continue this nonsensical command and said "The router ofter reboots! Just do Layer 3 config". So I added quickly the commands such as "dialer string", "dialer-group",etc. And the PING was successful for each other.
  Since it is unnecessary to verify the L2 config to insure the L3 config, the command "isdn test" is useless.

TIA
dillon

----- Original Message -----
From: "Brian McGahan" <bmcgahan@internetworkexpert.com>
To: "Dillon Yang" <dillony@gmail.com>; "Group Study" <ccielab@groupstudy.com>
Sent: Tuesday, July 26, 2005 12:57 AM
Subject: RE: lesson from failure

Dillon,

4. DO NOT use "isdn test" command that has no assurance the L3
connectivity and waste my half hour.

The point of using the "isdn test call" command is to test layer
2 connectivity only. If you use this command and layer 2 connectivity
is successful it means that any problem you have is with your
configuration, and is completely unrelated to the layer 1 or layer 2
hardware.

Also this test should not take more than 5 minutes maximum. If
it is taking you a half hour to put the ISDN switch-type on, define the
SPID value(s) (if required), and issue the "isdn test call" command then
you need to work on your speed of core configurations.

HTH,

Brian McGahan, CCIE #8593
bmcgahan@internetworkexpert.com

Internetwork Expert, Inc.
http://www.InternetworkExpert.com
Toll Free: 877-224-8987 x 705
Outside US: 775-826-4344 x 705
24/7 Support: http://forum.internetworkexpert.com
Live Chat: http://www.internetworkexpert.com/chat/

> -----Original Message-----
> From: nobody@groupstudy.com [mailto:nobody@groupstudy.com] On Behalf
Of
> Dillon Yang
> Sent: Sunday, July 24, 2005 11:23 PM
> To: Thomwin Chen; Group Study
> Subject: Re: lesson from failure
>
> 1. If ACL, use numerical ACL as possible.
> 2. If you want to prevent advertising update, use passive-interface as
> possible.
> 3. Do map FR interface to ping itself even if you are instructed with
"you
> do not need".
> 4. DO NOT use "isdn test" command that has no assurance the L3
> connectivity and waste my half hour.
> 5. DO config ISDN part at the last step or the 4-point section will
reduce
> your 10 points or more.
>
>
> Hi, Thomwin:
>
> Yes, the class B address will cause RIP advertise on all the
interface,
> so you need "passive-interface" to restrict the advertising.
>
> HTH
> dillon
>
>
> ----- Original Message -----
> From: "Thomwin Chen" <thomwin_chen@yahoo.com>
> To: "Dillon Yang" <dillony@gmail.com>; "Group Study"
> <ccielab@groupstudy.com>
> Sent: Monday, July 25, 2005 10:01 AM
> Subject: Re: lesson from failure
>
>
> > if you are using RIP,
> > let's say you have :
> > fa0/0 : 172.16.1.1/24
> > fa0/1 : 172.16.2.1/24
> >
> > passive-interface fa0/1 won't make RIP stop advertising
172.16.2.0/24,
> you should explicitly filter it. (RIP use network 172.16.0.0
statement)
> >
> > but if you are using eigrp & ospf the network statement is more
> granular, so you can control more specific what eigrp will advertise
and
> enable. (network 172.16.1.1 0.0.0.0)
> >
> > Dillon Yang <dillony@gmail.com> wrote:
> > 1. If ACL, use numerical ACL as possible.
> > 2. If advertise update, use passive-interface as possible.
> >
> > HTH
> > dillon
> >
> >



This archive was generated by hypermail 2.1.4 : Sun Sep 04 2005 - 17:00:31 GMT-3