From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ricardo Wurmus Subject: Re: [PATCH 0/3] Fix letsencrypt Date: Wed, 23 Mar 2016 21:44:51 +0100 Message-ID: <8760wc3ovw.fsf@elephly.net> References: Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 8bit Return-path: Received: from eggs.gnu.org ([2001:4830:134:3::10]:51719) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1aipeI-0002er-69 for guix-devel@gnu.org; Wed, 23 Mar 2016 16:45:03 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1aipeD-0000k6-7H for guix-devel@gnu.org; Wed, 23 Mar 2016 16:45:02 -0400 Received: from sender163-mail.zoho.com ([74.201.84.163]:24228) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1aipeC-0000ju-W4 for guix-devel@gnu.org; Wed, 23 Mar 2016 16:44:57 -0400 In-reply-to: List-Id: "Development of GNU Guix and the GNU System distribution." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: guix-devel-bounces+gcggd-guix-devel=m.gmane.org@gnu.org Sender: guix-devel-bounces+gcggd-guix-devel=m.gmane.org@gnu.org To: Leo Famulari Cc: guix-devel@gnu.org Leo Famulari writes: > Earlier today I found that I could not build letsencrypt. > > The recent update to python-hypothesis broke python-pyopenssl, which > blocked the build of letsencrypt. > > This patch series seems to be the minimum required to get letsencrypt > building again. I built letsencrypt on these patches and used it to > issue certificates for a new server. > > The update to python-pytest and python2-pytest will require ~200 > rebuilds [0]. > > This does not update python-pytest to the latest version, 2.9.1. I could > not get that to work, and I suspect it will require updates to some core > Python packages such as python-setuptools. > > In the meantime, I'd like to do this intermediate update to get > letsencrypt building again. Any objections? If it's okay, should it go > on master, or onto a staging branch? I’m okay with this. Thanks for taking it on! I hope not too many of the bioinfo packages fail to build because of this, but we’ll see ... :) ~~ Ricardo