[Olsr-users] Basic problem? router_id from loopback

[Kim Hawtin]

Hi Andreas,

There is a paper that describes all the concepts around implementing “anycast" services DNS, NTP and DHCP on IPv4 with OSPF, around 2003 ish. 

I implemented this at work on OSPF around 2006 and later for fun on OLSR V1. 

> On 27 Feb 2021, at 1:12 am, Andreas Martens1 <(spam-protected)> wrote:
> I've been trying to get this working for some time and I feat I'm just not understanding something... This might be a n00b problem! 
> 
> I have a group of nodes in a few different wireless networks. Because they have an IP address for each NIC I wanted to give them a unique IP address that could be routed across all the networks. 
> So I added an IP address to the lo device: 
> 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000 
>     link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 
>     inet 127.0.0.1/8 scope host lo 
>        valid_lft forever preferred_lft forever 
>     inet 172.20.0.1/32 scope global lo 
8<---
> I then added [interface=lo] to the olsrd2.conf: 
> [global] 
>         fork 1 
> [olsrv2] 
>         originator default_accept 
>         nhdp_routable true 
>         tc_interval 8.0 
>         tc_validity 40.0 
> [nhdp] 
>   mpr * 
> [interface=lo] 

You are not exporting the route here for the interface?
Is ‘lo' the actual interface name?

> but this seems to have stopped it working, as far as I can tell it's using 127.0.0.1 as its router_id now instead of the id of a NIC.   
> 
> What does the config need to look like so it distributes routes to the 172.20.0.0/32 addresses configured on loopback? 

The anycast technique on OLSR V1 works fine, don’t have the config handy from the setup in the old lab though.

We ran services on the lo:1 interface, using the same IP and machines on the mesh found the closest one.  From memory, I had to add lo:1 with its own routable network space as a /32.
DNS and NTP work ok as anycast on the mesh, but DHCP is ..er.. not really a thing unless you have helper machines along the whole route.
Question I suppose is, how do you do this now with the ‘ip’ command? Is that different?

Regards,

Kim