From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:34100) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1eK5eY-00067R-MZ for guix-patches@gnu.org; Wed, 29 Nov 2017 11:56:07 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1eK5eU-0004Ff-41 for guix-patches@gnu.org; Wed, 29 Nov 2017 11:56:06 -0500 Received: from debbugs.gnu.org ([208.118.235.43]:56769) by eggs.gnu.org with esmtps (TLS1.0:RSA_AES_128_CBC_SHA1:16) (Exim 4.71) (envelope-from ) id 1eK5eT-0004FN-W9 for guix-patches@gnu.org; Wed, 29 Nov 2017 11:56:02 -0500 Received: from Debian-debbugs by debbugs.gnu.org with local (Exim 4.84_2) (envelope-from ) id 1eK5eT-00071j-Lw for guix-patches@gnu.org; Wed, 29 Nov 2017 11:56:01 -0500 Subject: [bug#26685] certbot service experience Resent-Message-ID: From: ludo@gnu.org (Ludovic =?UTF-8?Q?Court=C3=A8s?=) References: <87tw56dhlp.fsf@dustycloud.org> <87eft3a804.fsf@gnu.org> <87tw0z4u5k.fsf@gnu.org> <87h8wymivl.fsf@dustycloud.org> <871slsmxkq.fsf@dustycloud.org> <87a80gpl3c.fsf@gnu.org> <87bmjmrppp.fsf@gnu.org> <87shcxwsck.fsf@dustycloud.org> Date: Wed, 29 Nov 2017 17:55:23 +0100 In-Reply-To: <87shcxwsck.fsf@dustycloud.org> (Christopher Allan Webber's message of "Tue, 28 Nov 2017 23:45:31 -0600") Message-ID: <878tepqb2c.fsf@gnu.org> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: guix-patches-bounces+kyle=kyleam.com@gnu.org Sender: "Guix-patches" To: Christopher Allan Webber Cc: wingo@igalia.com, 26685-done@debbugs.gnu.org, clement@lassieur.org Hello Chris! Christopher Allan Webber skribis: > Yay! But oh nooooo! I really should have replied to this! I had > stopped working on it because I got overloaded with final-push situation > in ActivityPub. Understood, no problem! We=E2=80=99ve had seven months and I thought we=E2= =80=99d rather make progress on this than let everyone re-implement more-or-less the same service. > Maybe this is worth moving to the mailing list instead, but one reason I > didn't push it was I had discovered a significant and serious > issue... building a system based off of its config is no longer > determistic based off of the system config alone... because this > checks for the rpesence of nginx-server-configuration-ssl-certificate > at *build* time of the system config. The following code is the > culprit: OK, I agree it=E2=80=99s a problem. That said, the =E2=80=98file-exists?=E2=80=99 check comes from an earlier a= nd IIUC unrelated commit, f2d7a492df759ab7416ff4b244e37896835f04fa. > Relatedly, the current setup has these defaults: >=20 > (ssl-certificate nginx-server-configuration-ssl-certificate > (default "/etc/nginx/cert.pem")) > (ssl-certificate-key nginx-server-configuration-ssl-certificate-key > (default "/etc/nginx/key.pem")) >=20 > IMO they should probably default to #f so that looking for the files > doesn't fail to boot nginx. I agree (apparently it comes from a Sept. 2016 commit, 8c00b83818acbbc3dfd406465e50439c3418afc5.) Could you send a patch? :-) Thanks, Ludo=E2=80=99.