Devuan bug report logs - #762
Debian removed essential startup scripts

Reply or subscribe to this bug.

Toggle useless messages

---

Message #5 received at submit@bugs.devuan.org (full text, mbox, reply):

[Message part 1 (text/plain, inline)]

Package: mdadm
Version: 4.2+20230508-2
Severity: critical

With version 4.2+20230227-1, Debian did remove all init scripts
(/etc/init.d/mdadm and /etc/init.d/mdadm-waitidle) for no needs.

This renders all systems that are using md devices unbootable!

-- Package-specific info:

IMPORTANT:
  please do not forget to include all relevant system information with this
  bug report. You could run
    /usr/share/bug/mdadm/script 3>&1
  as root and attach or include the output.


-- System Information:
Distributor ID:	Devuan
Description:	Devuan GNU/Linux 5 (daedalus/ceres)
Release:	5
Codename:	daedalus ceres
Architecture: x86_64

Kernel: Linux 6.1.12 (SMP w/8 CPU threads)
Kernel taint flags: TAINT_PROPRIETARY_MODULE, TAINT_CPU_OUT_OF_SPEC, TAINT_FIRMWARE_WORKAROUND, TAINT_OOT_MODULE
Locale: LANG=C, LC_CTYPE=C.UTF-8 (charmap=UTF-8), LANGUAGE not set
Shell: /bin/sh linked to /bin/dash
Init: sysvinit (via /sbin/init)

Versions of packages mdadm depends on:
ii  debconf [debconf-2.0]  1.5.82
ii  eudev [udev]           3.2.12-1
ii  libc6                  2.36-9
ii  libeudev1 [libudev1]   3.2.12-1

Versions of packages mdadm recommends:
ii  kmod  30+20230519-1

Versions of packages mdadm suggests:
iu  exim4-daemon-light [mail-transport-agent]  4.96-16

-- Configuration Files:
/etc/logcheck/ignore.d.server/mdadm [Errno 13] Permission denied: '/etc/logcheck/ignore.d.server/mdadm'
/etc/logcheck/violations.d/mdadm [Errno 13] Permission denied: '/etc/logcheck/violations.d/mdadm'

-- debconf information:
  mdadm/initrdstart_msg_errblock:
* mdadm/autocheck: false
  mdadm/initrdstart_msg_errmd:
* mdadm/initrdstart: all
* mdadm/autoscan: true
  mdadm/initrdstart_msg_errexist:
  mdadm/initrdstart_msg_errconf:
  mdadm/autostart: true
  mdadm/initrdstart_msg_intro:
* mdadm/start_daemon: true
* mdadm/mail_to: root@ethgen.de
  mdadm/initrdstart_notinconf: false
Gruß
   Klaus
-- 
Klaus Ethgen                                       http://www.ethgen.ch/
pub  4096R/4E20AF1C 2011-05-16            Klaus Ethgen <Klaus@Ethgen.ch>
Fingerprint: 85D4 CA42 952C 949B 1753  62B3 79D0 B06F 4E20 AF1C

[signature.asc (application/pgp-signature, inline)]

---

Message #8 received at 762@bugs.devuan.org (full text, mbox, reply):

Control: tags -1 debian

Klaus,

On Tue, Jun 13, 2023 at 10:25:59AM +0100, Klaus Ethgen wrote:
> Package: mdadm
> Version: 4.2+20230508-2
> Severity: critical

src:mdadm is not forked in Devuan and we use Debian's packages directly without
recompilation.

This is clearly a Debian issue, see the Changelog entry for 4.2+20230227-1[1]

Please report to Debian's BTS.

Thanks

Mark

[1]  https://tracker.debian.org/media/packages/m/mdadm/changelog-4.220230508-2

---

Message #15 received at 762@bugs.devuan.org (full text, mbox, reply):

[Message part 1 (text/plain, inline)]

Hi Mark,

Am Di den 13. Jun 2023 um 10:36 schrieb Mark Hindley:
> On Tue, Jun 13, 2023 at 10:25:59AM +0100, Klaus Ethgen wrote:
> > Package: mdadm
> > Version: 4.2+20230508-2
> > Severity: critical
> 
> src:mdadm is not forked in Devuan and we use Debian's packages directly without
> recompilation.
> 
> This is clearly a Debian issue, see the Changelog entry for 4.2+20230227-1[1]

Yes. But it is done by ignorance and reading from this log message, they
do not show enlightenment as in debian, that would not break anything.
(Well, except of that, who uses SysV init. But that users are cursed
anyway in debian.)

So, the package needs to be forked for devuan to prevent harm from
devuan users.

Regards
   Klaus
-- 
Klaus Ethgen                                       http://www.ethgen.ch/
pub  4096R/4E20AF1C 2011-05-16            Klaus Ethgen <Klaus@Ethgen.ch>
Fingerprint: 85D4 CA42 952C 949B 1753  62B3 79D0 B06F 4E20 AF1C

[signature.asc (application/pgp-signature, inline)]

---

Message #20 received at 762@bugs.devuan.org (full text, mbox, reply):

On Tue, Jun 13, 2023 at 10:41:44AM +0100, Klaus Ethgen wrote:
> Yes. But it is done by ignorance and reading from this log message, they
> do not show enlightenment as in debian, that would not break anything.

So, correct this misunderstanding by pointing out the error in a kind an
constructive way.

> (Well, except of that, who uses SysV init. But that users are cursed
> anyway in debian.)

But they still exist. So the correct place to fix this for both Debian
non-systemd users and Devuan users is in Debian.

Mark

---

Message #25 received at 762@bugs.devuan.org (full text, mbox, reply):

[Message part 1 (text/plain, inline)]

Hi Mark,

Am Di den 13. Jun 2023 um 10:47 schrieb Mark Hindley:
> On Tue, Jun 13, 2023 at 10:41:44AM +0100, Klaus Ethgen wrote:
> > Yes. But it is done by ignorance and reading from this log message, they
> > do not show enlightenment as in debian, that would not break anything.
> 
> So, correct this misunderstanding by pointing out the error in a kind an
> constructive way.

I don't think, that it is a "misunderstanding". The Debian packager
knows exactly what he does.

It is clearly not accidentally but clear knowing what they do.

It is that style that makes me frustrated about debian. I had many
discussions that mostly end with them, using their power to keep the
package in broken state. Sadly, my english is not that good to go into a
long therm battle.

Regards
   Klaus
-- 
Klaus Ethgen                                       http://www.ethgen.ch/
pub  4096R/4E20AF1C 2011-05-16            Klaus Ethgen <Klaus@Ethgen.ch>
Fingerprint: 85D4 CA42 952C 949B 1753  62B3 79D0 B06F 4E20 AF1C

[signature.asc (application/pgp-signature, inline)]

---

Message #30 received at 762@bugs.devuan.org (full text, mbox, reply):

Control: forwarded -1 https://bugs.debian.org/1037496

Klaus,

I have forwarded this request on your behalf.

Best wishes

Mark

---

Message #37 received at 762@bugs.devuan.org (full text, mbox, reply):

Klaus,

Could you please clarify the boot failure mode caused by the initscript
removal. We are pursuing this in Debian, but lorenzo has just pointed out that

On Tue, Jun 20, 2023 at 12:18:47PM +0200, lorenzo wrote:                                                                                                   
> Can you elaborate on this? I may be wrong, but I don't understand how the
> removal of mdadm script can render the system unbootable: the script runs
> mdadm in monitor mode and starts in runlevel 2. There is no script (removed)
> in runlevel S, I suspect a fail to boot will involve a problem in the kernel
> or in udev rules (that are still there)?

I have had a quick look and agree.

Is there some other change that has caused this? Or has the initramfs
integration broken. Have you edited the runlevels of the original initscripts?

Best wishes

Mark
-- 
Mark Hindley
GPG: 506C 15A4 2B0A F5A0 A854  23EE D28A 45BF 3287 D649

---

Message #42 received at 762@bugs.devuan.org (full text, mbox, reply):

[Message part 1 (text/plain, inline)]

Hi Marc,

Am Di den 20. Jun 2023 um 12:17 schrieb Mark Hindley:
> Could you please clarify the boot failure mode caused by the initscript
> removal. We are pursuing this in Debian, but lorenzo has just pointed out that
> 
> On Tue, Jun 20, 2023 at 12:18:47PM +0200, lorenzo wrote:                                                                                                   
> > Can you elaborate on this? I may be wrong, but I don't understand how the
> > removal of mdadm script can render the system unbootable: the script runs
> > mdadm in monitor mode and starts in runlevel 2. There is no script (removed)
> > in runlevel S, I suspect a fail to boot will involve a problem in the kernel
> > or in udev rules (that are still there)?
> 
> I have had a quick look and agree.
> 
> Is there some other change that has caused this? Or has the initramfs
> integration broken. Have you edited the runlevels of the original initscripts?

Hmm. I have to admit that I did not had the courage to boot the system
without that init script as I am depending on working md devices.

But I have a pretty old system with no initrd and separated /usr as it
was supported long ago. So I am usual extra sensitive to such changes.

Regards
   Klaus
-- 
Klaus Ethgen                                       http://www.ethgen.ch/
pub  4096R/4E20AF1C 2011-05-16            Klaus Ethgen <Klaus@Ethgen.ch>
Fingerprint: 85D4 CA42 952C 949B 1753  62B3 79D0 B06F 4E20 AF1C

[signature.asc (application/pgp-signature, inline)]

---

Message #47 received at 762@bugs.devuan.org (full text, mbox, reply):

Klaus,

On Tue, Jun 20, 2023 at 12:39:12PM +0100, Klaus Ethgen wrote:
> Hmm. I have to admit that I did not had the courage to boot the system
> without that init script as I am depending on working md devices.

OK, thanks. We have to be really careful about this then. Claiming non-existent
boot breakage as a justification considerably weakens any case we make. If you
want to persist with the assertion, your will need to verify: maybe try a boot
without the initscripts and with a recovery USB stick, or similar, to hand?.

> But I have a pretty old system with no initrd and separated /usr as it
> was supported long ago. So I am usual extra sensitive to such changes.

Yes, now I remember. Although I know you don't like it, booting without an
initrd has not been supported for a long time.  I am finding it hard to locate
the specific release, but I *think* it was Etch first released in 2007 and EOL
in 2010. Of course you are welcome to do it if you can get it to work, but you
are going to be continually disappointed by expecting Devuan or Debian to
support it in any way now. When booting without an initrd no longer works for
you, you may just have to accept that and start using one.

I am sorry if that sounds blunt or uncaring. It isn't meant to be -- just a
statement of fact. I had an old AMD K6 system that was made obsolete by the
change to gcc -march=686 default. I completely understand your pain. However,
Devuan (and Debian) has enough issues to deal with just to keep non-systemd init
viable and working.

With best wishes

Mark

---

Message #52 received at 762-done@bugs.devuan.org (full text, mbox, reply):

Version: 4.2+20230508-5devuan1

Source package mdadm (4.2+20230508-5devuan1) added to Devuan suite unstable.

This closes bug report 762.

Thanks

DAK managing the Devuan archive

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA512

Format: 1.8
Date: Sat, 01 Jul 2023 09:54:13 +0100
Source: mdadm
Architecture: source
Version: 4.2+20230508-5devuan1
Distribution: unstable
Urgency: medium
Maintainer: Mark Hindley <mark@hindley.org.uk>
Changed-By: Mark Hindley <mark@hindley.org.uk>
Closes: 762
Changes:
 mdadm (4.2+20230508-5devuan1) unstable; urgency=medium
 .
   * Fork for Devuan.
   * Revert "Adding debconf note when run on non-systemd systems."
   * Revert "Removing cron jobs in favour of systemd timers."
   * Revert "Removing sysvinit scripts in favour of systemd units."
     (Closes: #762).
   * Fixup lintian override for init.d-script-possible-missing-stop.
   * Remove unused lintian override systemd-service-file-missing-install-key.
   * d/t/control: add missing kmod dependency.
Checksums-Sha1:
 a28f396f9230ccf8a6be8371c1f97ca85e05fea3 1682 mdadm_4.2+20230508-5devuan1.dsc
 9695d567c99c60a9d5711c3358f10f535ab72d47 617020 mdadm_4.2+20230508.orig.tar.gz
 ac29c0ded3bef1e266c53ca6aed40297066765ca 89044 mdadm_4.2+20230508-5devuan1.debian.tar.xz
 8b49c08f8551444f30023b530e7ccde83faeb87b 5417 mdadm_4.2+20230508-5devuan1_source.buildinfo
Checksums-Sha256:
 7a58a3615c6116a341c23028d0303c78a58435d34e931fb2b8ba527c68b97a17 1682 mdadm_4.2+20230508-5devuan1.dsc
 c6a60f9948ac9c22e53ea78e06c64a4d8a06a5543b7adfe509561f8b3d9466bf 617020 mdadm_4.2+20230508.orig.tar.gz
 ccfb2e3610223925ff89599ebdf8336cfd3815f06bf54b0138d159168a0d1109 89044 mdadm_4.2+20230508-5devuan1.debian.tar.xz
 7be1130c0dafa42402368dcfec316066d6533811c8df6689ad88cbd6d728b261 5417 mdadm_4.2+20230508-5devuan1_source.buildinfo
Files:
 45db4e5006847131c672352a74c0b2af 1682 admin optional mdadm_4.2+20230508-5devuan1.dsc
 ffcef2839a36d0ee0b39cce9dfc7711a 617020 admin optional mdadm_4.2+20230508.orig.tar.gz
 33b85e48e05285bcf825e7a927565338 89044 admin optional mdadm_4.2+20230508-5devuan1.debian.tar.xz
 eb4f89fdb96bce4e31dc940d719cddf0 5417 admin optional mdadm_4.2+20230508-5devuan1_source.buildinfo

-----BEGIN PGP SIGNATURE-----

iQEzBAEBCgAdFiEEcuPLdzMV36LkZHQ9lFMhJFQZIvsFAmSf79EACgkQlFMhJFQZ
Ivttzgf9EQqo18rWdzRzK2glwD3F/Z5KElW9EMu0q5sgPYZOhEB6TxMvvrRPIyE8
1QVC3fNdvWKubpnnl8frq6geFZwH+xLH2HowPpcCCC1biSqdMTQOXIJSje/yI9MS
ki3LwvGM2e2YQQkXsOtX/ChMzTnIYJL2Vtf/CGq+G6UGSKATqBpSvFzst+SNsZUD
KwIOqbQuiAohgsKgqEBjwL9JI/oMUD+bNSA7SyPOeOgOAbofkf1PVZlvzIREA6qv
GX/qjAebC8mVnrI7p9y/fS6l0dhWdmsbngJCeIAPBpvxLzpTbUztReCpI7Lq1NKU
SUoMP7BgGeKXWtVEMssRM5bNN/HWYQ==
=cf40
-----END PGP SIGNATURE-----

---

Send a report that this bug log contains spam.

---