Re: [v6ops] IPv6-Only Preferred DHCPv4 option

[Tomek Mrugalski <[e-mail]>]

On 06.12.2019 17:02, Ted Lemon wrote:
> On Dec 6, 2019, at 6:45 AM, Bernie Volz (volz) <[e-mail]
> <mailto:[e-mail]>> wrote:
>> which requires fewer changes on the DHCP server implementation
>
> This is almost certainly not true.  In order for this to work, you’re
> going to have to have some special-purpose code to support it.  Better
> to do the clean solution than a hack that minimizes changes to the
> server at the expense of more complication in the client-server
> interaction.
This part of the discussion seems to have been mostly settled. But in
case it isn't, here's another opinion. As a DHCP server vendor, I'd very
much prefer no code changes. For several reasons:

1. You can start using the option now, without any server upgrades, just
tweak the server configuration to send one extra option. This is
expected to be a significant factor in the adoption process.

2. The basic DHCPv4 mechanism of clients requesting options in PRL and
the server sending them is decades old and there are tons of policy
mechanisms around it (send only when requested, force sending to
specific devices or class of devices, treat devices that request that
option as a separate class, etc.) These are all well understood by
network admins.

3. Extra logic requires extra testing. There may be incompatibilities,
especially with firewalls and relays that attempt to track state. Some
of them may be hard to detect.

Having said that, I'm willing to implement code changes if I have to.

Tomek