Devuan bug report logs -
#261
eudev: System with encrypted root fs and swap fails to shutdown cleanly
Reply or subscribe to this bug.
Toggle useless messages
Report forwarded to devuan-bugs@lists.dyne.org, owner@bugs.devuan.org
:
bug#261
; Package eudev
.
(full text, mbox, link).
Acknowledgement sent to "Nik Wrt" <nikwrt@gmx.de>
:
New bug report received and forwarded. Copy sent to owner@bugs.devuan.org
.
(full text, mbox, link).
Message #5 received at submit@bugs.devuan.org (full text, mbox, reply):
Package: eudev
Version: 3.2.2-13
Severity: normal
On a fresh Devuan ASCII install with encrypted root and swap (no LVM, no RAID), I am experiencing exactly the same symptom described in the first post of https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=792552 i.e.
Stopping remaining crypto disks
sda2_crypt (busy)...
and the system hangs for a long time before powering off.
This behavior on debian stretch was determined to be caused by a bug in the udev init scripts when using sysvinit, see https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=791944
The relevant fix is in the commit
https://salsa.debian.org/systemd-team/systemd/commit/e803efca59978aa5bb1d8806247f986d0c0f7e67
Information forwarded to devuan-bugs@lists.dyne.org, owner@bugs.devuan.org
:
bug#261
; Package eudev
.
(full text, mbox, link).
Acknowledgement sent to Daniel Reurich <daniel@centurion.net.nz>
:
Extra info received and forwarded to list. Copy sent to owner@bugs.devuan.org
.
(full text, mbox, link).
Message #10 received at 261@bugs.devuan.org (full text, mbox, reply):
[Message part 1 (text/plain, inline)]
On 17/11/18 22:38, Where all about bugs in Devuan packages is reported
wrote:
> Package: eudev
> Version: 3.2.2-13
> Severity: normal
>
> On a fresh Devuan ASCII install with encrypted root and swap (no LVM, no RAID), I am experiencing exactly the same symptom described in the first post of https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=792552 i.e.
>
> Stopping remaining crypto disks
> sda2_crypt (busy)...
>
> and the system hangs for a long time before powering off.
>
> This behavior on debian stretch was determined to be caused by a bug in the udev init scripts when using sysvinit, see https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=791944
>
> The relevant fix is in the commit
>
> https://salsa.debian.org/systemd-team/systemd/commit/e803efca59978aa5bb1d8806247f986d0c0f7e67
>
> _______________________________________________
> Devuan-bugs mailing list
> Devuan-bugs@lists.dyne.org
> https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/devuan-bugs
>
I've developed a patch based on the one for systemd[-udev] that should
resolve the issue. Can you please test and confirm if it does?
Thanks,
Daniel.
--
Daniel Reurich
Centurion Computer Technology (2005) Ltd.
021 797 722
[signature.asc (application/pgp-signature, attachment)]
Information forwarded to devuan-bugs@lists.dyne.org, owner@bugs.devuan.org
:
bug#261
; Package eudev
.
(full text, mbox, link).
Acknowledgement sent to Daniel Reurich <daniel@centurion.net.nz>
:
Extra info received and forwarded to list. Copy sent to owner@bugs.devuan.org
.
(full text, mbox, link).
Message #15 received at 261@bugs.devuan.org (full text, mbox, reply):
[Message part 1 (text/plain, inline)]
The fundamental bug in eudev is fixed in ascii-proposed
but cryptsetup does nothing to detect conditions where a root filesystem
is dependent on a crypt volume (directly or indirectly). In this case
it should not try to stop the encrypted filesystem at all due to a
chicken<->egg scenario (root is needed to stop its encrypted volume but
also needs to be unmounted in order to for cryptmount to stop the
encrypted volume.
The most reliable fix is to not run the stop functions in the initscript
at all on shutdown
(I tested on a vm with an encrypted lvm setup containing root - default
devuan encrypted setup)
--
Daniel Reurich
Centurion Computer Technology (2005) Ltd.
021 797 722
[signature.asc (application/pgp-signature, attachment)]
Information forwarded to devuan-bugs@lists.dyne.org, owner@bugs.devuan.org
:
bug#261
; Package eudev
.
(full text, mbox, link).
Acknowledgement sent to Tom <tgrom.automail@nuegia.net>
:
Extra info received and forwarded to list. Copy sent to owner@bugs.devuan.org
.
(full text, mbox, link).
Message #20 received at 261@bugs.devuan.org (full text, mbox, reply):
Hello, I just wanted to report this issue also effects Devuan ASCII
when using OpenRC INIT instead of sysV INIT.
Information forwarded
to devuan-bugs@lists.dyne.org, Devuan Dev Team <devuan-dev@list.dyne.org>
:
bug#261
; Package eudev
.
(Mon, 27 Jan 2020 13:33:01 GMT) (full text, mbox, link).
Acknowledgement sent
to svante.signell@gmail.com
:
Extra info received and forwarded to list. Copy sent to Devuan Dev Team <devuan-dev@list.dyne.org>
.
(Mon, 27 Jan 2020 13:33:04 GMT) (full text, mbox, link).
Message #25 received at 261@bugs.devuan.org (full text, mbox, reply):
Package: eudev
Version: 3.2.2-13
tags 261 moreinfo
Please check if the problems remains with eudev-3.2.9-4. And do also check the
Debian bugs, fixes and information for udev and cryptsetup e.g.:
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=791944
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=720340
Information forwarded
to devuan-bugs@lists.dyne.org, Devuan Dev Team <devuan-dev@lists.dyne.org>
:
bug#261
; Package eudev
.
(Thu, 30 Jan 2020 10:48:01 GMT) (full text, mbox, link).
Acknowledgement sent
to svante.signell@gmail.com
:
Extra info received and forwarded to list. Copy sent to Devuan Dev Team <devuan-dev@lists.dyne.org>
.
(Thu, 30 Jan 2020 10:48:03 GMT) (full text, mbox, link).
Message #30 received at 261@bugs.devuan.org (full text, mbox, reply):
On Wed, 2020-01-29 at 15:15 -0800, Tom wrote:
> On Tue, 28 Jan 2020 13:06:14 +0100
> Svante Signell <svante.signell@gmail.com> wrote:
>
> > Package: eudev
> > Version: 3.2.2-13
> > tags 261 moreinfo
> >
> > Please check if the problems remains with eudev-3.2.9-4. And do
> > also check the Debian bugs, fixes and information for udev and
> > cryptsetup e.g.:
> > https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=791944
> > https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=720340
> I don't see that a backport exists for ASCII for eudev version 3.2.9-
> 4. Only 3.2.2-13. Are you asking me to create a backport by hand or
> did you already have something to test in mind?
Since 3.2.9-4 is only available in ceres/unstable yet I propose that
you download eudev=3.2.9-4 libeudev1=3.2.9-4 from unstable and install
them by dpkg -i ..., do shutdown/reboot tests, and report back.
Please do also read the above Debian bugs on cryptsetup problems and
fixes. I don't have any box with an encrypted partition so I cannot
test.
Thanks!
Information forwarded
to devuan-bugs@lists.dyne.org, Devuan Dev Team <devuan-dev@lists.dyne.org>
:
bug#261
; Package eudev
.
(Fri, 31 Jan 2020 00:33:01 GMT) (full text, mbox, link).
Acknowledgement sent
to Tom <tgrom.automail@nuegia.net>
:
Extra info received and forwarded to list. Copy sent to Devuan Dev Team <devuan-dev@lists.dyne.org>
.
(Fri, 31 Jan 2020 00:33:03 GMT) (full text, mbox, link).
Message #35 received at 261@bugs.devuan.org (full text, mbox, reply):
On Thu, 30 Jan 2020 11:48:33 +0100
Svante Signell <svante.signell@gmail.com> wrote:
> On Wed, 2020-01-29 at 15:15 -0800, Tom wrote:
> > On Tue, 28 Jan 2020 13:06:14 +0100
> > Svante Signell <svante.signell@gmail.com> wrote:
> >
> > > Package: eudev
> > > Version: 3.2.2-13
> > > tags 261 moreinfo
> > >
> > > Please check if the problems remains with eudev-3.2.9-4. And do
> > > also check the Debian bugs, fixes and information for udev and
> > > cryptsetup e.g.:
> > > https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=791944
> > > https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=720340
>
> > I don't see that a backport exists for ASCII for eudev version
> > 3.2.9- 4. Only 3.2.2-13. Are you asking me to create a backport by
> > hand or did you already have something to test in mind?
>
> Since 3.2.9-4 is only available in ceres/unstable yet I propose that
> you download eudev=3.2.9-4 libeudev1=3.2.9-4 from unstable and install
> them by dpkg -i ..., do shutdown/reboot tests, and report back.
>
> Please do also read the above Debian bugs on cryptsetup problems and
> fixes. I don't have any box with an encrypted partition so I cannot
> test.
>
> Thanks!
>
Sure. just please give me some time to put aside for building the bpo
and testing it. I'm not sure exactly how much of Debian's disk crpyto
problems will effect me because I'm not using systemd, but OpenRC init.
Will report on it. It sounds to me most of the LUKS mounting issues
come from the way systemd manages mounts not the luks mapper software
itself.
Information forwarded
to devuan-bugs@lists.dyne.org, Devuan Dev Team <devuan-dev@lists.dyne.org>
:
bug#261
; Package eudev
.
(Fri, 31 Jan 2020 02:33:01 GMT) (full text, mbox, link).
Acknowledgement sent
to Tom <tgrom.automail@nuegia.net>
:
Extra info received and forwarded to list. Copy sent to Devuan Dev Team <devuan-dev@lists.dyne.org>
.
(Fri, 31 Jan 2020 02:33:03 GMT) (full text, mbox, link).
Message #40 received at 261@bugs.devuan.org (full text, mbox, reply):
On Thu, 30 Jan 2020 11:48:33 +0100
Svante Signell <svante.signell@gmail.com> wrote:
> On Wed, 2020-01-29 at 15:15 -0800, Tom wrote:
> > On Tue, 28 Jan 2020 13:06:14 +0100
> > Svante Signell <svante.signell@gmail.com> wrote:
> >
> > > Package: eudev
> > > Version: 3.2.2-13
> > > tags 261 moreinfo
> > >
> > > Please check if the problems remains with eudev-3.2.9-4. And do
> > > also check the Debian bugs, fixes and information for udev and
> > > cryptsetup e.g.:
> > > https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=791944
> > > https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=720340
>
> > I don't see that a backport exists for ASCII for eudev version
> > 3.2.9- 4. Only 3.2.2-13. Are you asking me to create a backport by
> > hand or did you already have something to test in mind?
>
> Since 3.2.9-4 is only available in ceres/unstable yet I propose that
> you download eudev=3.2.9-4 libeudev1=3.2.9-4 from unstable and install
> them by dpkg -i ..., do shutdown/reboot tests, and report back.
>
> Please do also read the above Debian bugs on cryptsetup problems and
> fixes. I don't have any box with an encrypted partition so I cannot
> test.
>
> Thanks!
>
Ok, so I built and installed eudev from Ceres on ASCII, version 3.2.9-4
of eudev and libeudev1. I rebooted 3 times for good measure. This did
NOT solve the failing to unmount cryptdisks issue.
Information forwarded
to devuan-bugs@lists.dyne.org, Devuan Dev Team <devuan-dev@lists.dyne.org>
:
bug#261
; Package eudev
.
(Fri, 31 Jan 2020 11:18:02 GMT) (full text, mbox, link).
Acknowledgement sent
to svante.signell@gmail.com
:
Extra info received and forwarded to list. Copy sent to Devuan Dev Team <devuan-dev@lists.dyne.org>
.
(Fri, 31 Jan 2020 11:18:04 GMT) (full text, mbox, link).
Message #45 received at 261@bugs.devuan.org (full text, mbox, reply):
On Thu, 2020-01-30 at 18:12 -0800, Tom wrote:
>
> Ok, so I built and installed eudev from Ceres on ASCII, version
> 3.2.9-4
> of eudev and libeudev1. I rebooted 3 times for good measure. This did
> NOT solve the failing to unmount cryptdisks issue.
Thank you for trying. Next steps would be to check if any version of
eudev/udev solves this issue. I suspect that this bug is only related
to cryptsetup. Did you have any version of udev/eudev working properly?
If you can please do the following:
1) Downgrade to 3.2.7-6 from Beowulf and check if it works.
2) If not, try udev from Debian/Stretch (corresponds to Devuan/ASCII).
Either 232-25+deb9u12 from stretch or 241-5~bpo9+1 from stretch
backports. You can find source/binaries from snapshot.debian.org.
3) Optionally try some newer Debian version, e.g. 241-7~deb10u2 from
buster or 244-3~bpo10+1 from buster-backports.
If neither of the above works, which I suspect, the problem is not due
to udev/eudev. It is due to the cryptsetup package(s), and should be
reassigned to it.
Then you can try out the workarounds mentioned in #237 and
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=791944
You also have this bug that was tagged as wontfix.
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=720340
Thanks!
Send a report that this bug log contains spam.