OSPF config Example

[rvi-ip]

I would like to know if OSPF can be configured to auto detect neighbors and thus build a “Mesh” or do I have to add each neighbor by hand in the config? Either way can some one please give me an example on OSPF config? I have searched the post and found a lot of post but nothing like Lonnie’s RIP config example where each command is listed line for line.



Best regards,

Rvi-ip

[timo]

Quagga documentation:

http://www.quagga.net/docs/docs-info.php

[lonnie]

This is a very straightforward config for a unit that has my ADSL lines. It is set to announce that it is a default route for other units. It has 3 areas, one of which is the special area 0 backbone. It announces three networks.

Current configuration:
!
hostname ospfd
password 1234
!
!
!
interface eth0
ip ospf cost 100
!
interface eth1
ip ospf cost 100
!
interface lo
!
interface tunl0
!
interface gre0
!
interface eth2
!
interface ecb
!
interface ipacct
!
interface beacon
!
interface wlanbr
!
interface cbq
!
interface eth3
!
interface eth4
!
router ospf
ospf router-id 10.0.0.1
network 10.0.0.0/16 area 0.0.0.0
network 204.50.71.112/28 area 0.0.0.4
network 204.50.234.88/30 area 0.0.0.5
default-information originate
!
access-list vtylist permit 127.0.0.1/32
access-list vtylist deny any
!
line vty
access-class vtylist
!
end


This is a script from an intermediate radio repeater.
Current configuration:
!
hostname ospfd
password 1234
!
!
!
interface eth0
ip ospf cost 100
!
interface wpci0
ip ospf cost 500
!
interface wpci1
ip ospf cost 500
!
interface lo
!
interface tunl0
!
interface gre0
!
interface eth1
!
interface ecb
!
interface ipacct
!
interface beacon
!
interface wlanbr
!
interface cbq
!
router ospf
ospf router-id 10.0.1.33
network 10.0.1.33/27 area 0.0.0.0
!
access-list vtylist permit 127.0.0.1/32
access-list vtylist deny any
!
line vty
access-class vtylist
!
end

[GlobalProof-TG]

Yes, in most of the cases there is no need to configure the neighbours in OSPF (as you have to do in BGP), unless you are in non-broadcast networks (like in example frame-relay/serial/point-to-point links). All the OSPF-enabled routers will detect each other via multicast(!!!).

The only thing you might do is to "manually" choose your DR (Designated router) by influating the priority ID.

Do not be afraid from OSPF. It can serve very well to your network if you know how to work with it. The main issue is that zebra/quagga is still not 100% reliable especially with the OSPF protocol.

I would like to know if OSPF can be configured to auto detect neighbors and thus build a “Mesh” or do I have to add each neighbor by hand in the config? Either way can some one please give me an example on OSPF config? I have searched the post and found a lot of post but nothing like Lonnie’s RIP config example where each command is listed line for line.



Best regards,

Rvi-ip

[Michal]

If you use WRAP + CM9 atheros, don´t use bridged wpci.

OSPF multicast don´t pass thru bridged wpci on WRAP. This is bad, we like use system router-bridge-router-bridge for speed up of main link. OSPF then don´t work. No problem on PC.

[GlobalProof-TG]

From what I understand, you are feeling sorry for being not able to run OSPF due to the structure of your network.
I think that you are able. You need to confirgure your OSPF neighbors manually and it will work (and use virtual-link if you are running a multi-are OSPF network and some area is not directly connected to the backbone 0.0.0.0).
Believe me, it works and it is not complex.

[knolan]

Looking at all the examples etc. on the forums, I still have a number of questions.

1. Does implementing OSPF mean removing all static routes from all nodes on the network, including the default routes.
2. If the node which is doing the default route to the internet, has 2 seperate routes to the internet, one which is prefered, How do I tell Star OS that both routes are default routes.
3. If I have two nodes which can have default routes, both at either end of the network, can I tell Star OS to use one node always, except if that node goes down, then to use the second backup node.

Regards,
Keith

[lonnie]

With OSPF you can remove all default routes except for the Internet gateway machines. They can advertise themselves as default routes for other machines.

OSPF will determine which one to use based on the costing you assign to each device.

It would help you to get that Cisco book on OSPF. Check the Library topic.

[bairdc]

1. Does implementing OSPF mean removing all static routes from all nodes on the network, including the default routes.
You can leave static routes if you want to. If you have static routes, they will be preferred over dynamic routes. However, by its nature, if a network segment for any of those static routes goes down, you will not have any way to automatically route around the problem. That is one of the big advantages of dynamic routing. If you're trying to set up a redundant network that will automatically route around problems, you need to ditch static routes.

2. If the node which is doing the default route to the internet, has 2 seperate routes to the internet, one which is prefered, How do I tell Star OS that both routes are default routes.
If you have a single device with two routes to the Internet, then that device should be configured to originate a default route into OSPF. StarOS will then have a default route to direct traffic toward that device. There is no reason for that device to announce two default routes. All StarOS needs to know is which device is the gateway device. Once the traffic gets there, then that device can determine which Internet link to send it out on. It's not something that your StarOS box even needs to know about.


3. If I have two nodes which can have default routes, both at either end of the network, can I tell Star OS to use one node always, except if that node goes down, then to use the second backup node.
I actually have my network set up this way. I've got two physically separate default routes. OSPF determines the active route based on the cost (metric) involved. If you have two routes to the same destination, but you only want one of them to be used unless failover is required, then its just a matter of assigning a higher cost to the non-preferred route.

Craig