OpenBSD Journal

p2k16 Hackathon Report: espie@ on proot

Contributed by tj on from the unlikely-porters dept.

Our very first p2k16 hackathon report comes from none other than Marc Espie, who writes:

Lots of thanks to Gilles Chehade, Epitech Nantes, and Aymeric Fouchault for the organization. It was top-notch. The only complaint I might have is that the food was so good that I might have eaten too much.

What's to say about Nantes? I've already been to the city twice, and it is nice. I especially enjoyed "les machines de l'ile" which I had no time to see last time I was around. Definitely must-see if you're leaning towards interesting mechanical ideas, totally geeky look at the world, in a good sense.

And it's always a pleasure to meet with Aymeric and his boundless optimism.

As far as the ports crowd goes, there were a lot of the usual suspects. One notable missing person being Kili. Another one Jeremy.

As usual, Remi had aquaponey and couldn't come.

Cool to meet some newcomers like tsg and danj, or tb, and to finally meet jca (yeah, we managed to miss each other each time before that, go figure)

There were some unlikely porters that nobody expected: deraadt, beck, mlarkin and krw... I'm not sure they did a lot of ports work, though we got at least some interesting shit fixed thanks to them, and yet more new problems to solve thanks to deraadt.

This might be the one ports hackathon where I did what I had set up to do, namely finally deal with one of landry's longtime nags, and put together a proper script to create chroots for ports building, instead of doing it by frankensteining a proper release process (build, make install DESTDIR=/root.... do some more steps once, forget about them and keep going until things break every year)

When I came in I had no idea how I was going to achieve that. And now, one week later, I have something that works correctly for me, and even looks like it's tweakable in a sensible way so that other people will be able to use it.

As usual, it's in perl (what else) because I required a lot of unusual properties:

- do links instead of copies whenever possible
- record enough of what we did so that we can rescan the directory and tell about pieces we don't know about
- do chroot to chroot

etc. It's extensible.

As usual, I found and fixed some bugs in other tools like dpb, and Some of them were due to me doing things in a way that happened to work by chance, and some others which turned to be much easier to reproduce within a separate chroot.

All in all, a rather productive week. Didn't spend that much time writing code, but a lot of time thinking about stuff instead.

Oh, and watching the usual ports activity, and talking about strategy for the future. But that's another story. And I'm sure the other p2k16 reports will cover some of that.

Thanks for the report, Marc!

(Comments are closed)


Copyright © - Daniel Hartmeier. All rights reserved. Articles and comments are copyright their respective authors, submission implies license to publish on this web site. Contents of the archive prior to as well as images and HTML templates were copied from the fabulous original with Jose's and Jim's kind permission. This journal runs as CGI with httpd(8) on OpenBSD, the source code is BSD licensed. undeadly \Un*dead"ly\, a. Not subject to death; immortal. [Obs.]