From mboxrd@z Thu Jan 1 00:00:00 1970 From: ludo@gnu.org (Ludovic =?UTF-8?Q?Court=C3=A8s?=) Subject: bug#26976: On Hydra, offload crashes while trying to build linux-libre source Date: Wed, 14 Jun 2017 22:10:59 +0200 Message-ID: <87r2yms52k.fsf@gnu.org> References: <87h90japz0.fsf@netris.org> <87h90h966f.fsf@gnu.org> <87pof5dnja.fsf@gnu.org> <87mva88oe9.fsf@gnu.org> <87k24q86hk.fsf@gnu.org> <87shj3tvzd.fsf@gnu.org> 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]:59108) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1dLEe7-0001Pk-5U for bug-guix@gnu.org; Wed, 14 Jun 2017 16:12:08 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1dLEe2-0007Vt-8l for bug-guix@gnu.org; Wed, 14 Jun 2017 16:12:07 -0400 Received: from debbugs.gnu.org ([208.118.235.43]:45328) by eggs.gnu.org with esmtps (TLS1.0:RSA_AES_128_CBC_SHA1:16) (Exim 4.71) (envelope-from ) id 1dLEe2-0007Vp-55 for bug-guix@gnu.org; Wed, 14 Jun 2017 16:12:02 -0400 Sender: "Debbugs-submit" Resent-Message-ID: In-Reply-To: <87shj3tvzd.fsf@gnu.org> ("Ludovic \=\?utf-8\?Q\?Court\=C3\=A8s\=22'\?\= \=\?utf-8\?Q\?s\?\= message of "Tue, 13 Jun 2017 23:32:06 +0200") 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.org@gnu.org Sender: "bug-Guix" To: Mark H Weaver Cc: Artyom Poptsov , 26976@debbugs.gnu.org, guix-sysadmin@gnu.org (+ Cc: guix-sysadmin) ludo@gnu.org (Ludovic Court=C3=A8s) skribis: > I=E2=80=99ve pushed a workaround that seems to work (=E2=80=98guix copy= =E2=80=99 commands that > previously segfaulted on hydra.gnu.org no longer do): > > https://git.savannah.gnu.org/cgit/guix.git/commit/?id=3D8e469b67f95cfe5= b95405b503b8ee315fdf8ce66 > > The idea is simply to explicitly close all the channel ports. That way, > when their finalizer gets called, it doesn=E2=80=99t invoke libssh code. > > (guix build offload) and (guix ssh) already explicitly close all the > channel ports they open, so I found that the culprit is =E2=80=98node-eva= l=E2=80=99 in > Guile-SSH. The patch changes =E2=80=98node-eval=E2=80=99 to explicitly c= lose the RREPL > channel upon completion. (BTW, performance-wise, it may be best to > avoid opening a new channel every time =E2=80=98node-eval=E2=80=99 is cal= led.) > > That may be good enough for Guix, but of course that=E2=80=99s a workarou= nd and > not a proper fix. > > I=E2=80=99ll do some more testing and then maybe try switching hydra.gnu.= org to > Guile 2.2 again. After some more testing, I=E2=80=99ve upgraded hydra.gnu.org to /gnu/store/4p56wc3lmr177ffw0zkjdwcmm3cb18wx-guix-0.13.0-2.de9d8f0 (from /gnu/store/cxmj38x6rh0ykq3d5dlqbxr5h1zgiiaf-guile2.0-guix-0.13.0-1.a6d728b). So far so good, but let=E2=80=99s keep an eye on the ongoing builds. We can always roll back if it turns out my optimism was misplaced. Ludo=E2=80=99.