From mboxrd@z Thu Jan 1 00:00:00 1970 From: ludo@gnu.org (Ludovic =?utf-8?Q?Court=C3=A8s?=) Subject: Re: [PATCH 4/4] gnu: Add duplicity. Date: Mon, 26 May 2014 08:59:37 +0200 Message-ID: <874n0drpnq.fsf@gnu.org> References: <87wqdbptse.fsf@member.fsf.org> <87r43i7qdr.fsf@gnu.org> <87tx8doyvt.fsf@gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable Return-path: Received: from eggs.gnu.org ([2001:4830:134:3::10]:41857) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1Woosu-0006eB-P3 for Guix-devel@gnu.org; Mon, 26 May 2014 02:59:57 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1Woosl-0001Mv-KG for Guix-devel@gnu.org; Mon, 26 May 2014 02:59:48 -0400 Received: from hera.aquilenet.fr ([2a01:474::1]:41447) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1Woosl-0001KK-Dl for Guix-devel@gnu.org; Mon, 26 May 2014 02:59:39 -0400 In-Reply-To: <87tx8doyvt.fsf@gmail.com> (Eric Bavier's message of "Mon, 26 May 2014 01:08:38 -0500") 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: Eric Bavier Cc: Guix-devel@gnu.org Eric Bavier skribis: > Ludovic Court=C3=A8s writes: > >>> + (arguments >>> + `(#:python ,python-2 ;setup assumes Python 2 >>> + #:tests? #f ;TODO: tests cannot find lockf= ile module >> >> As Andreas notes, may be worth investigating, since that could be a real >> issue. Could it be that it=E2=80=99s expecting a different version of >> python2-lockfile or something like that? > > The "#:tests? #f" is a hold-over from a previous mistake. The tests can > indeed find the lockfile module. OK. > However, I am seeing some non-deterministic hangs in the python2.7 > threading module when running the tests. If the tests don't hang, there > are many, many ERRORs. I believe the errors are due to the tests > themselves, but I'm not sure about the threading hangs. Has enyone else > had issues with our python's threading module? Not me. You think it=E2=80=99s not specific to Duplicity? Ludo=E2=80=99.