Hello Rick,

Now i see that default route option is not appropriate. I forgot the MPLS 
requirements in that task.

Regarding the iSPF timers, i still think that the command i mentioned is valid. 
Because of this:

"Incremental SPF is scheduled in the same way as the full SPF. Routers enabled 
with incremental SPF and routers not enabled with
incremental SPF can function in the same internetwork."

http://www.cisco.com/en/US/docs/ios/iproute/configuration/guide/irp_ospf_incre_spf.html#wp1054075


Regards,

Antonio Soares, CCIE #18473 (R&S)
[email protected]

-----Original Message-----
From: Rick Mur [mailto:[email protected]] 
Sent: terça-feira, 2 de Junho de 2009 8:13
To: Antonio Soares
Cc: [email protected]
Subject: Re: [OSL | CCIE_SP] VOL2 - Section 1

The task 1.2 is an error in the lab. R2 and R5 were connected someday, so just 
enabling ppp would be sufficient. The solution
IPexpert gives is changing the interface to frame-relay and use the DLCI 
between these routers.
I also lost several hours solving that.

I also agree with the no arp frame-relay. It's not needed to turn of 
inverse-arp.

Indeed the default OAM timers are good enough, still I like to put on the 
minimal timers to ensure it's fast enough. Still the
default should be enough so 'oam-pvc manage' command should be the only command 
needed.

The question asks that all routes are available when any of the links get's 
disconnected. This means that you need all routes
available in L1 and a default is not enough. Besides that you need a /32 route 
to label switch traffic to the loopback that's not in
L1 and a default wouldn't do the job there either.
For the password I would also set an area-password or domain-password, depends 
on where authentication needs to be enabled (within a
level or the entire NET area), don't remember the question of that one.

'timers throtlle spf' is not a valid solutions, because I believe the task 
requires that partial SPF calculations are performed when
receiving an update. That means ispf is the only valid solution, since this 
enables partial calculation.
The throttle command sets timers for 'delaying' full SPF calculations.


Rick Mur
CCIE2 #21946 (R&S / SP)
[email protected]

On Tue, 2 Jun 2009 01:09:51 +0100
  "Antonio Soares" <[email protected]> wrote:
  Hello group,
  
  I'm now starting Section 1 and i have some comments i would like to see 
discussed:
  
  Task 1.2) In the 7200/ATM topology, R2 is not directly connected to R5. R2 
connects to R4 and R4 connects to R5. So if this true,
  this task is not so trivial as it seems. I lost several hours to make it 
work. Basically i configured CRB in R4 and IRB in R2 and
  R5. With the release i was using, the bridge was not working. Then i moved to 
another release and the bridge started passing
  traffic. Then i found another problem with IP traffic from R1 to R5.
The solution was using PPP between R2 and R5 instead of HDLC.
  
  Task 1.3) Why do we need "no arp frame-relay" ? I'm convinced that this 
command has not effect at all. And we don't need to
  configure IETF in R1. The routers are smart enough to communicate even if the 
encapsulation is IEFT in one side an CISCO in the
  other end.
  
  Task 1.4) The default OAM timers are enough to make it faster than ISIS.
  
  Task 2.2) Is this minimal configuration ? My options were area-password and 
domain-password. And instead of using L2-to-L1 route
  leaking, why don't we simply send the default route from R1 and R5 to
R2 ?
  
  Task 3.3) Isn't the command "timers throttle spf" also valid for iSPF ?
  
  
  Thanks.
  
  Regards,
  
  Antonio Soares, CCIE #18473 (R&S)
  [email protected]
  

Reply via email to