From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ludovic =?UTF-8?Q?Court=C3=A8s?= Subject: bug#40518: guix pull: Dependency guix-packages-base fails Date: Fri, 10 Apr 2020 10:31:46 +0200 Message-ID: <87r1wvyc5p.fsf@gnu.org> References: <3DTCFZZAVGJ4B.22Q96Q868VFQ8@wilsonb.com> <87lfn53skg.fsf@gnu.org> <1Z6E7NH7WBPL4.3SGTTMBBYVN5C@wilsonb.com> <87zhbk1xsn.fsf@gnu.org> <2EXPIM22USD15.3L4PAFHXMY9H2@wilsonb.com> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable Return-path: Received: from eggs.gnu.org ([2001:470:142:3::10]:43463) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1jMp51-00048E-U7 for bug-guix@gnu.org; Fri, 10 Apr 2020 04:32:04 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1jMp50-0004Mc-Mn for bug-guix@gnu.org; Fri, 10 Apr 2020 04:32:03 -0400 Received: from debbugs.gnu.org ([209.51.188.43]:43210) by eggs.gnu.org with esmtps (TLS1.0:RSA_AES_128_CBC_SHA1:16) (Exim 4.71) (envelope-from ) id 1jMp50-0004MV-JK for bug-guix@gnu.org; Fri, 10 Apr 2020 04:32:02 -0400 Received: from Debian-debbugs by debbugs.gnu.org with local (Exim 4.84_2) (envelope-from ) id 1jMp50-0003eH-G6 for bug-guix@gnu.org; Fri, 10 Apr 2020 04:32:02 -0400 Sender: "Debbugs-submit" Resent-Message-ID: In-Reply-To: <2EXPIM22USD15.3L4PAFHXMY9H2@wilsonb.com> (elaexuotee@wilsonb.com's message of "Fri, 10 Apr 2020 10:03:51 +0900") List-Id: Bug reports for GNU Guix List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: bug-guix-bounces+gcggb-bug-guix=m.gmane-mx.org@gnu.org Sender: "bug-Guix" To: elaexuotee@wilsonb.com Cc: 40518@debbugs.gnu.org Hi, elaexuotee@wilsonb.com skribis: >> It could be an out-of-memory (OOM) condition: this build phase is >> unfortunately very demanding currently (at least 2=C2=A0GiB or RAM I thi= nk). >> How much RAM do you have? > > I think we have a winner. This is running on a VM and I just realized it > it only runs with 500MB of ram. OK. > So, if I am understanding correctly, for some reason the original pull > didn't have an available substitute; however, when I later tried again > the new HEAD referenced a guix-packages-base which *did* have a > substitute available? That=E2=80=99s probably what happened, yes: if someone just pushed to repo,= then substitutes aren=E2=80=99t available yet; a few minutes later, they=E2=80= =99re available. You can drop the attached file in ~/.config/guix/channels.scm to pull from the latest commit that has been fully built (which is not exactly equivalent to =E2=80=9Csubstitutes available=E2=80=9D, but as close to it a= s it gets): --8<---------------cut here---------------start------------->8--- (use-modules (guix ci) (srfi srfi-1) (ice-9 match)) (define (latest-commit-successfully-built) "Return the latest commit for which substitutes are (potentially) available." (let* ((evaluations (filter (lambda (evaluation) (and (evaluation-complete? evaluation) (string=3D? "guix-modular-master" (evaluation-spec evaluation)))) (latest-evaluations "https://ci.guix.gnu.org"= )))) (any (lambda (evaluation) (match (evaluation-checkouts evaluation) ((checkout) (checkout-commit checkout)) (_ #f))) evaluations))) ;; Pull the latest commit fully built on berlin.guixsd.org. ;; WARNING: This could downgrade your system! (list (channel (name 'guix) (url "https://git.savannah.gnu.org/git/guix.git") (commit (pk 'commit (latest-commit-successfully-built))))) --8<---------------cut here---------------end--------------->8--- Use with care! If the ci.guix.gnu.org is lagging or compromised, you could find yourself not upgrading at all, or not upgrading to the latest commit. Ludo=E2=80=99.