I only have one router/one "next-hop" that itself is connected to multiple uplinks using multiple interfaces. With source address based routing, this step works. An uplink loss is detected using the mwan3 software. The result is that further connections are rewritten using my stateful prefix rewriting and redirected to another uplink.
If the host is not warned that some prefix is deprecated (reminder, it is not possible now through a few hops of routers), then the host would use the disconnected IP address space for the packet source header. Such packet may enter only the respective Carrier. Any other Carrier must drop it for spoofing protection (a BCP is requesting it). Source routing just does not make sense because the mode fundamental problem was not resolved. If you would do NAT in this situation, then do not create a problem for yourself: do NAT for all traffic, and preserve the current IPv4 design. IPv6 first hop is a really complex matter. For example: I did complain many times in 6man that they have an extraordinary choice for SASA (RFC 6724). IPv6 decides about the packet structure (for example: source IP address field) only after the next hop is decided. This field in the packet would be populated with the IP prefix advertised by the next hop router. People, not familiar with IPv6 may not believe in this. Eduard -----Original Message----- From: Jonas Lochmann <ripe-ipv6-wg@jonaslochmann.de> Sent: Friday, March 7, 2025 16:46 To: Vasilenko Eduard <vasilenko.eduard@huawei.com> Cc: ipv6-wg@ripe.net Subject: Re: [ipv6-wg] Re: IPv6 Multihoming with Load Balancing Hi, On Fri, Mar 07, 2025 at 07:20:44AM +0000, Vasilenko Eduard via ipv6-wg wrote:
You did mention RFC 8678. Then probably you have a multi-hop routing site because this RFC concentrates only on this aspect of the MHMP problem (all other problems are out of the scope). Then look to section 6 of our draft (comparison table) - you will have big challenges with the provider's addresses - this option is probably blocked for you. Actually, RFC 8676 is pretty useless yet, because there is no way to propagate ISP uplink loss through the site (to withdraw the particular carrier IPv6 PA address) - the blackholing is guaranteed. Then the advice to get your own address space and become the full BGP speaker is probably a good one.
I only have one router/one "next-hop" that itself is connected to multiple uplinks using multiple interfaces. With source address based routing, this step works. An uplink loss is detected using the mwan3 software. The result is that further connections are rewritten using my stateful prefix rewriting and redirected to another uplink.
Actually, the problem is very complex (you could look at the draft) - IPv6 flexibility on the 1st hop always translates to tremendous complexity. Are you sure that you need multi-homing in IPv6? This rabbit hole is very deep. You could stay with multi-homing in IPv4.
Then I get fast IPv4 and slow IPv6. Then I could disable IPv6 for a better user experience. Nothing that I want to do.