Yes,
I forgot about the getting that default route into the enterprise
from the CE. That would leave the "default-information orginate" .
Nigel
>From: "Howard C. Berkowitz" >Reply-To: "Howard C. Berkowitz" >To:
[EMAIL PROTECTED] >Subject: Re: (correction) Method and Process
Scenario 5: OSPF [7:42139] >Date: Sun, 21 Apr 2002 12:37:52 -0400 > >
>See inline > > > > > >>From: "Howard C. Berkowitz" > >>Reply-To: "Howard
C. Berkowitz" > >>To: [EMAIL PROTECTED] > >>Subject: Re: (correction)
Method and Process Scenario 5: OSPF [7:42139] > >>Date: Sun, 21 Apr 2002
09:08:21 -0400 > >> > >> >You can originate a 0.0.0.0 route from each of
the ISP connected >routers. > >> >Ideally, then each router on your
internal network that receives both > >> >0.0.0.0 routes from both
originating routers will route traffic to the > >> >internet based on the
the 0.0.0.0 route with the best metric. > >> > >>You're on the right
track. But what characteristics must the default > >>routes have to
assure a degree of load sharing? (I'm thinking of > >>something specific
to OSPF) > > > >NT: On redistribution of the default-route using OSPF's
default assigment of > >"E2", with a standard "metric XX" value at both
POP's will allow the both > >default routes to be equal-cost. Another
option here would be to use the > >"ospf cost" or "bandwidth"
configuration commands to balance the links. > >That would make the links
to the ISP load balanced, but it wouldn't >necessarily equalize the load
in getting to them from within the >enterprise. Again, you are on the
right track. > > > > > > >>What is their effect on load balancing from
the provider to you? > >> > >> > > >> >I'm not sure about OSPF, but where
I work we have 2 connections to the > >> >internet at difference POPs,
and this is the method we use. Seems to > >>give > >> >some load
balancing, however, based on the number of users at each > >>site.... >
>> >i.e. we have twice as many users at one site (which chooses it's
closest > >> >internet connection for exit to the net) as we do at the
other, so we > >>really > >> >get a lopsided load balance, but it's what
we expect. We are soon going > >>to > >> >be implementing BGP on the 2
routers that connect to the internet so >what > >>we > >> >can have
inbound redunancy from the internet, but we'll still leave the > >>
>lopsided load balancing in place as to really load balance across our >
>> >internet connections would each bandwidth on our OC-12, which we
don't > >> >want.... > >> > > >> >Mike W. > >> >----- Original Message
----- > >> >From: "Howard C. Berkowitz" > >> >Newsgroups:
groupstudy.cisco > >> >Sent: Saturday, April 20, 2002 2:51 PM > >>
>Subject: (correction) Method and Process Scenario 5: OSPF Multihoming >
>> >[7:42092] > >> > > >> > > >> >> Your enterprise runs OSPF internally
and only takes default from one > >> >> ISP, but at multiple POPs. What
would this suggest you could do to > >> >> achieve a degree of
load-sharing among the POPs? > >> >> > >> >> Assume you do not run BGP.
What can you do and what are its > >>limitations? > >> >> > >> >> Don't
focus on the configuration commands as what mechanisms will be > >>
>required. >
>_________________________________________________________________ > >MSN
Photos is the easiest way to share and print your photos: >
misconduct and Nondisclosure violations to [EMAIL PROTECTED]
------------------------------------------------------------------------
Get your FREE download of MSN Explorer at http://explorer.msn.com.
Message Posted at:
http://www.groupstudy.com/form/read.php?f=7&i=42158&t=42139
--------------------------------------------------
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
