From: Fernando, Emil (emil.fernando@atosorigin.com)
Date: Sun May 16 2004 - 22:58:31 GMT-3
Hi Howard,
Howard,
First let me thank you for the prompt reply.
I understand that these commands are more effective when used in large
environments. As you have mentioned in your reply,
with the ospf demand circuit we can suppress ospf hellos and reflooding of
aged out SLAs. My understanding is that routers with a demand circuit will
not age out all LSA that the router received over the demand circuit and
will not resend locally generated and other LSAs over that circuit. If this
router has a another normal(non demand cuircuit) connection, then all LSAs
will be flooded over that connection. Pls correct me if I am wrong. I am
just trying to understand what router does in this situation.
Please could you tell me what router is actually doing when there is
interface with ip ospf database-filter, cisco doc says it
stops LSA flooding on that interface. Does that mean the router treat the
interface as a demand circuit by only blocking LSA floods and sending OSPF
hellos? and what happen if it stops sending LSA to the neighbour. will the
Neighbour loose all lsa info from that client? What it does with locally
generated LSAs? are they also suppressed on this interface?
I can understand that Ip ospf flood-reduction simply increase LSA age out
time and therefore reduce floods to others.
I appreciate your advice on this.
Thanks
Emil
-----Original Message-----
From: Howard C. Berkowitz [mailto:hcb@gettcomm.com]
Sent: Monday, 17 May 2004 10:49 AM
To: ccielab@groupstudy.com
Subject: Re: OSPF LSA filtering
At 7:59 AM +1000 5/17/04, Fernando, Emil wrote:
>Hi Group,)
>
>In OSPF there are two interface commands ( ip ospf database-filter and
ip
>ospf flood-reduction ) that we can use to improve bandwidth usage.
>Can someone help me to understand when I should use these commands and
which
>command is more effective than the other one. What OSPF does when these
>commands ara enabled ? Can I use these commands on a tail site to improve
>the routing performance of the tail site ?
>
Before trying to use these features, first, how have you determined
there is a bandwidth problem that needs to be fixed, or how would you
approach it? Second, if there is a bandwidth problem, there may be
better ways to fix it than tuning OSPF at the protocol level.
Still within OSPF, one reason you may have a bandwidth problem is
that your areas are too large, or have too many inter-area/external
routes injected into them. Due to such sizing, there may be too many
LSAs to transfer, and they may not, in any case, be useful to tail
routers.
Flooding reduction is intented for large, stable networks, as often
found in ISPs or some well-run enterprise networks. You can think of
it as a means of applying the logic in the demand circuit extension
to general OSPF, which stops reflooding LSAs simply because they have
aged out, and only refloods when a significant topological event
occurs. The desire to minimize updates in large, stable networks is
one reason many ISPs use ISIS as their IGPs, because ISIS has always
had good facilities for minimizing updates in what can be single
areas of over 1000 routers. See
http://www.cisco.com/en/US/products/sw/iosswrel/ps1834/products_feature_guid
e09186a008008011e.html
-- ============================================================================ === "What Problem are you trying to solve?" ***send Cisco questions to the list, so all can benefit -- not directly to me*** **************************************************************************** **** Howard C. Berkowitz presentation downloads by anonymous ftp from www.netcases.net.
This archive was generated by hypermail 2.1.4 : Wed Jun 02 2004 - 11:12:12 GMT-3