Your message dated Fri, 02 May 2014 17:36:41 +0200
with message-id <5363bb89.2020...@debian.org>
and subject line Re: Bug#746577: closed by Michael Biebl <bi...@debian.org>
(Re: Bug#746577: systemd-sysv: for upgrade safety, systemd-sysv and
sysvinit-core must be coinstallable)
has caused the Debian Bug report #746577,
regarding systemd-sysv: for upgrade safety, systemd-sysv and sysvinit-core must
be coinstallable
to be marked as done.
This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.
(NB: If you are a system administrator and have no idea what this
message is talking about, this may indicate a serious mail system
misconfiguration somewhere. Please contact ow...@bugs.debian.org
immediately.)
--
746577: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=746577
Debian Bug Tracking System
Contact ow...@bugs.debian.org with problems
--- Begin Message ---
Source: systemd-sysv
Version: 204-10
Severity: critical
Justification: breaks the whole system
Control: retitle -2 sysvinit-core: for upgrade safety, systemd-sysv and
sysvinit-core must be coinstallable
For safety of upgrades from wheezy to jessie, the process of upgrading
packages and installing new ones *must not* change either the currently-
running init or the init that will manage the system on the next boot.
A transition away from sysvinit must only occur as a result of a
deliberate sysadmin action with that specific effect, not as a side
effect of any other operation.
To achieve this, systemd-sysv and sysvinit-core must be coinstallable,
with an alternatives-like mechanism for deciding which package actually
provides init (if this can be done with actual alternatives for
/sbin/init, that would be ideal, but I suspect it cannot be that simple).
Other potential providers of /sbin/init should ideally also be included
in this mechanism, but because the default-for-new-installations is changing
in jessie from sysvinit to systemd, the cooperation of those two providers
is more important than the others.
severity:critical because, depending on how an installation is configured,
an unanticipated conversion to systemd absolutely can cause the system
to be unbootable, or fail to carry out its intended function (e.g. because
a network server no longer starts on boot as intended).
-- System Information:
Debian Release: jessie/sid
APT prefers unstable
APT policy: (501, 'unstable'), (500, 'testing'), (101, 'experimental')
Architecture: amd64 (x86_64)
Kernel: Linux 3.14-1-amd64 (SMP w/8 CPU cores)
Locale: LANG=en_US.utf8, LC_CTYPE=en_US.utf8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
--- End Message ---
--- Begin Message ---
Am 02.05.2014 17:07, schrieb Zack Weinberg:
> Control: reopen -1
>
>> You didn't show any actual breakage, so closing the bug report.
>> If you can show us of actual breakage caused by the switch to
>> systemd where the system will not boot, we will certainly deal
>> with it.
>
> I believe you have missed the point of the bug report. This is not
> about specific cases that break; any known specific case that breaks
> can presumably be dealt with in the package metadata and/or scripts,
> it is a simple matter of programming. This is about providing a
> *safety net* for all of the cases that we don't find out about until
> it is too late to handle them in an automated manner. It is my
> contention that the best way to do that is to require an extra, manual
> action to switch from sysvinit to systemd, decoupled from package
> upgrades.
>
> Reopening. I will escalate this to policy if I have to, but I hope I
> don't have to.
The ctte decisions was about switching the default init for jessie, that
includes upgrades.
If you want to opt-out of that switch you can certainly do that, but the
default action is to do the switch.
Michael
--
Why is it that all of the instruments seeking intelligent life in the
universe are pointed away from Earth?
signature.asc
Description: OpenPGP digital signature
--- End Message ---
