From mboxrd@z Thu Jan 1 00:00:00 1970 From: ludo@gnu.org (Ludovic =?UTF-8?Q?Court=C3=A8s?=) Subject: bug#21010: eject fails because of hardcoded path to umount Date: Thu, 09 Jul 2015 00:05:11 +0200 Message-ID: <87h9pei9uw.fsf@gnu.org> References: <871tgig7w2.fsf@gmail.com> <20150708163144.GA11642@debian> 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]:41432) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1ZCxTg-0005I4-PT for bug-guix@gnu.org; Wed, 08 Jul 2015 18:06:05 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1ZCxTe-00031y-1Q for bug-guix@gnu.org; Wed, 08 Jul 2015 18:06:04 -0400 Received: from debbugs.gnu.org ([140.186.70.43]:43105) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1ZCxTd-00031r-Uf for bug-guix@gnu.org; Wed, 08 Jul 2015 18:06:01 -0400 Received: from Debian-debbugs by debbugs.gnu.org with local (Exim 4.80) (envelope-from ) id 1ZCxTd-00035q-KO for bug-guix@gnu.org; Wed, 08 Jul 2015 18:06:01 -0400 Sender: "Debbugs-submit" Resent-Message-ID: In-Reply-To: <20150708163144.GA11642@debian> (Andreas Enge's message of "Wed, 8 Jul 2015 18:31:44 +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-bounces+gcggb-bug-guix=m.gmane.org@gnu.org To: Andreas Enge Cc: 21010@debbugs.gnu.org Andreas Enge skribis: > On Wed, Jul 08, 2015 at 02:18:21PM +0200, Alex Sassmannshausen wrote: >> eject: unable to exec /bin/umount of `/mnt/usb' > > The culprit is in the source code of util-linux-2.25.2/sys-utils/eject.c: > 681: execl("/bin/umount", "/bin/umount", name, "-n", N= ULL); > 683: execl("/bin/umount", "/bin/umount", name, NULL); > 685: errx(EXIT_FAILURE, _("unable to exec /bin/umount of `%s'"= ), name); This should be fixed in core-updates. > This should be easy to patch, but there are other utilities with the same > problem. A hopeless case seems to be "whereis": It looks for files in fix= ed > lists of locations such as > static const char *bindirs[] =3D { > "/usr/bin", > "/usr/sbin", I think we should check what this list is used for, and then prepend entries accordingly. Volunteering? :-) Ludo=E2=80=99.