Do IPv6 addresses have better persistence with DHCPv6?

Question

I have a LAN with about 25 devices that occasionally grab each other's IP address via DHCPv4. I know there are certain lease time limits and it can be set to very long periods to work it around to an extent, but it happens once in a while regardless. This irks me, and I recently switched to IPv6 in the internal network. Does DHCPv6 have better resistance to such IP address losses, such as combined with something like DUID/MAC of the device? Or does it work with the incremental IP addresses in the same range starting with the same seed? It would be such a waste of space if it worked like that though?

I've read some documentation about this but haven't been able to find a clear answer.

Answer 1

If you're seeing duplicates with DHCPv4, it's very likely you'd see the same with DHCPv6, because the problem is with your DHCP server and/or configuration. Both systems work exactly the same, addresses generally aren't recycled until they have to be. (when a lease expires, the address becomes a "tombstone" -- expired but still in the database as having been assigned to whomever. until the pool is depleted, tombstones don't get recycled, because there are any number of reasons a node may have failed to renew but still use the address.)

It's a common practice for DHCP servers to "ping" the address they are about to assign to ensure it isn't being used.

The only saving grace with IPv6 is the size. Your v4 pool may be too small resulting in too much churn. IPv6 pools will generally be many orders of magnitude larger. (A LAN is typically an insane ::/64, as required for SLAAC. But with DHCPv6, your LAN can be whatever size you want, if you never want SLAAC to function.)