Couple thoughts on this. Cisco's OSPF should prefer intra area routes over
inter unless the administrative distances are modified. By default, as many
have mentioned, they are all set to 110. However, internally, I believe
path cost is the 2nd tie break, with intra beating inter as the first. If
this is the case (which it should be), modifying path costs whether directly
or through other means will not influence the correct behavior. If this
wasn't the case, the default metric calculation would already have traffic
following the correct path as ABD is a lower cost path then ACD. Where they
all in the same area, traffic would flow properly.
Hence, the question is obviously what do you do? In my opinion, the design
does not optimally reflect the constraints of the physical topology. I
assume we are looking at a subset of a larger network inclusive of more
routers? I would suggest a modification in the OSPF configuration or
possible the routing strategy itself may be in order. What is the reasoning
behind not running in a single area? In cases like these, I find we often
spend more time looking for the complex technical kludge instead of stepping
back and asking the now famous question "What problem are we(you) trying to
solve?"
Pete
*********** REPLY SEPARATOR ***********
On 5/29/2001 at 11:38 AM Kevin Schwantz wrote:
>Thanks for the recommendations. Firstly, let me explain why I need the
>routing to behave in such a way. The reasons are purely geographical and I
>want to reduce latency. Routers A and B are in London and connected back to
>back via FastEth. Routers C and D are in and SanJose and NewYork
>respectively(Connected to both London routers via FR).
>I certaintly won't want traffic originating from RouterA ( London )
>destined
>for RouterD (NewYork) to have to go to SanJose first. It would be much
>better if the hop is A-B-D instead of A-C-D.
>
>Schwantz
>
>""EA Louie"" wrote in message
>[EMAIL PROTECTED]">news:[EMAIL PROTECTED]...
>> ... or route-map the router D network(s) to go through Router B at Router
>A
>>
>>
>> ----- Original Message -----
>> From: "Chris Larson"
>> To:
>> Sent: Tuesday, May 29, 2001 7:24 AM
>> Subject: RE: Wanna Be a CCIE? Try This One [7:6076]
>>
>>
>> > Place a summary route to null 0 for the networks on Router D on your
>OSPF
>> > routers and set the metrics appropriately for the summary route
>> >
>> > -----Original Message-----
>> > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]On Behalf Of
>> > Kevin Schwantz
>> > Sent: Tuesday, May 29, 2001 10:03 AM
>> > To: [EMAIL PROTECTED]
>> > Subject: Re: Wanna Be a CCIE? Try This One [7:6076]
>> >
>> >
>> > routerA routerB
>> > AREA0--------AREA0
>> > | |
>> > routerC routerD
>> > AREA1---------AREA1
>> >
>> >
>> > Since we are on the topic of OSPF, could someone help me out on the
>> scenario
>> > above?
>> >
>> > Routers A and B have interfaces in Area 0 and Area1. I want traffic
>from
>> > routerA destined for routerD to go via router B. This is not the case
>in
>> my
>> > network because I realise that routerA prefers Intra-Area routes and
>thus
>> > would route traffic to routerD via routerC.
>> > What tweaks must I make in order to force the traffic from routerA to
>> > routerD to go via routerB ? Someone suggested building a GRE tunnel
>> between
>> > routerA and routerB and then configure the tunnel to be in AREA1.
>> >
>> > Any suggestions?
>> >
>> > Kevin
>> >
>> >
>> > ""W. Alan Robertson"" wrote in message
>> > [EMAIL PROTECTED]">news:[EMAIL PROTECTED]...
>> > > Guys,
>> > >
>> > > The actual traffic will not be routed up to area 0... Area 0 has
>been
>> > > extended
>> > > down to R2, so R2 is now a backbone router. R2 has interfaces in 3
>> areas
>> > > now:
>> > > Area1, Area2, and Area0 by means of it's virtual link.
>> > >
>> > > Any traffic originating in Area2 destined for Area1 will be routed
>> > directly
>> > > by
>> > > R2. This satisfies the "Interarea traffic must traverse the
>backbone"
>> > rule,
>> > > because R2 *is* a backbone router.
>> > >
>> > > This is not theory... It is fact.
>> > >
>> > > Alan
>> > >
>> > > ----- Original Message -----
>> > > From: "Andrew Larkins"
>> > > To:
>> > > Sent: Monday, May 28, 2001 10:13 AM
>> > > Subject: RE: Wanna Be a CCIE? Try This One [7:6076]
>> > >
>> > >
>> > > > agreed....to area 0 then on to the intended area
>> > > >
>> > > > -----Original Message-----
>> > > > From: Circusnuts [mailto:[EMAIL PROTECTED]]
>> > > > Sent: 28 May 2001 15:50
>> > > > To: [EMAIL PROTECTED]
>> > > > Subject: Re: Wanna Be a CCIE? Try This One [7:6076]
>> > > >
>> > > >
>> > > > Chuck- my answer is Yes. The traffic from the Virtual Linked
>> psuedo-ABR
>> > > > passes back to Area 0, before it's sent onto the intended Area
>(even
>> if
>> > > it's
>> > > > directly connected).
>> > > >
>> > > > Phil
>> > > >
>> > > >
>> > > > ----- Original Message -----
>> > > > From: Chuck Larrieu
>> > > > To:
>> > > > Sent: Sunday, May 27, 2001 8:59 PM
>> > > > Subject: Wanna Be a CCIE? Try This One [7:6076]
>> > > >
>> > > >
>> > > > > Ever wonder what the CCIE candidates talk about on the CCIE list?
>> > > > >
>> > > > > The following message came through today. I thought the bright
>folks
>> > on
>> > > > this
>> > > > > list might be curious, and might want to venture an answer.
>> > > > >
>> > > > > Begin original question:
>> > > > >
>> > > > > Guys,
>> > > > >
>> > > > > I wonder if there is anybody who remembers the discussion on
>Virtual
>> > > > > Links in OSPF. It was posted some time ago but I can't seem to
>find
>> > it.
>> > > > >
>> > > > > The scenario was something like this:
>> > > > > ________ _______ _______
>> > > > > |Area 0 | |Area1| |Area2|
>> > > > > | R0 |--| R1 |--| R2 |
>> > > > > |______| |_____| |_____|
>> > > > >
>> > > > > There is a virtual link from area 2 to Area 0 via Area1. Traffic
>> needs
>> > to
>> > > > > get to R1 in Area 1 from R2 in Area 2. Assume that the virtual
>link
>> > has
>> > > to
>> > > > > use R1 (To create the V.Link). Does the traffic flow passed R1
>(in
>> > Area
>> > > 1)
>> > > > > to Area 0 and then back to area 1, or does the actual flow just
>to
>> R1
>> > > from
>> > > > > R2.
>> > > > >
>> > > > > I cant remember the conclusion, and I cant seem to find it on the
>> > > > archives.
>> > > > > Quite interesting issues.
>> > > > >
>> > > > > End of original question
>> > > > >
>> > > > >
>> > > > > Chuck
>> > > > >
>> > > > > One IOS to forward them all.
>> > > > > One IOS to find them.
>> > > > > One IOS to summarize them all
>> > > > > And in the routing table bind them.
>> > > > >
>> > > > > -JRR Chambers-
>> > > > > FAQ, list archives, and subscription info:
>> > > > http://www.groupstudy.com/list/cisco.html
>> > > > > Report misconduct and Nondisclosure violations to
>> [EMAIL PROTECTED]
>> > > > FAQ, list archives, and subscription info:
>> > > > http://www.groupstudy.com/list/cisco.html
>> > > > Report misconduct and Nondisclosure violations to
>[EMAIL PROTECTED]
>> > > > FAQ, list archives, and subscription info:
>> > > http://www.groupstudy.com/list/cisco.html
>> > > > Report misconduct and Nondisclosure violations to
>[EMAIL PROTECTED]
>> > > FAQ, list archives, and subscription info:
>> > http://www.groupstudy.com/list/cisco.html
>> > > Report misconduct and Nondisclosure violations to
>[EMAIL PROTECTED]
>> > FAQ, list archives, and subscription info:
>> > http://www.groupstudy.com/list/cisco.html
>> > Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
>> > FAQ, list archives, and subscription info:
>> http://www.groupstudy.com/list/cisco.html
>> > Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
>> FAQ, list archives, and subscription info:
>http://www.groupstudy.com/list/cisco.html
>> Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
>FAQ, list archives, and subscription info:
>http://www.groupstudy.com/list/cisco.html
>Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
Message Posted at:
http://www.groupstudy.com/form/read.php?f=7&i=6273&t=6076
--------------------------------------------------
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
