[dhcwg] DHCP hackathon in Prague?

[Tomek Mrugalski <[e-mail]>]

IETF is about rough consensus and running code. We had a lot of the
consensus part lately, how about a bit of running code for a change? ;)

Possible DHCP hackathon in Prague was mentioned briefly during Dallas
meeting. I'd like to follow up on the idea. I'm not sure if you recall,
but there was a hackathon in Dallas. Folks who participated found it
very useful and had good results (outcome of the last one:
https://www.ietf.org/proceedings/92/slides/slides-92-iesg-opsplenary-3.pdf,
more details
https://www.ietf.org/registration/MeetingWiki/wiki/ietfhackathon)

I'd love to hear your opinion on this. The hackathon would likely take
two days (Saturday and Sunday preceding IETF). Interested people would
meet, decide on common unresolved software issues, exchange ideas, hack
the code and maybe even do a bit of interop testing!

For now, we don't have any specific topics agreed on, but the ones
mentioned so far are:

- stateful issues - the draft got IESG approval recently. It would be
good to check what sort of tweaking is required for existing
implementations.
- secure dhcpv6 - that's a challenging one. We could check whether there
are existing implementations that could support this already. If not, we
could develop a plan to extend existing code base and start writing code.
- privacy - there's a lot of going on in the DHCP privacy area. We could
work on existing implementations or even start working on a privacy
oriented client from scratch. I know of at least one DHCP library that
would make this feasible in 2 days.

Other areas could be worked on, too. For example shared v4 allocation,
dhcpv6 load balancing, stable privacy addresses or maybe YANG modules
for dhcp? Feel free to propose your own topics.

In my opinion, it would make sense to go ahead with the hackathon if
there is enough people willing to work on specific topics. So the
questions are:

1. are you willing to participate in the DHCP hackathon, i.e. travel to
Prague 2 days earlier and spend your time with fellow DHCP geeks hacking
the code?
2. what topics are of particular interest for you?
3. do you have your own implementation or a patch for existing one? If
not, that's ok, there's plenty of open source projects that could
benefit from your time and skills.

Question 3 is completely optional. If you don't want to reveal your
secret plans, that's perfectly ok. I'm merely asking to gauge which
areas are likely to have the biggest code base, so perhaps we could plan
some sort of interop testing.

If there's enough interest, Bernie and I will submit this proposal to
whoever is in charge of the hackathon planning. We'll probably organize
a wiki somewhere and move the detailed discussion off the dhc list.
(There was a separate list for '92 hackathon.)

Bernie & Tomek