Your message dated Wed, 28 Sep 2022 12:21:12 +0200
with message-id <f67c8fb2-3dfe-5b06-70e9-ccbf8be76...@debian.org>
and subject line Re: include whether machine was rebooted after upgrading in
the bugscript
has caused the Debian Bug report #719527,
regarding include whether machine was rebooted after upgrading in the bugscript
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.)
--
719527: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=719527
Debian Bug Tracking System
Contact ow...@bugs.debian.org with problems
--- Begin Message ---
Package: systemd
Version: 204-2
Severity: normal
During DebConf13, I talked a bit more with Lennart about the upgrade
process of systemd. He mentioned that updates without rebooting are
generally not tested and not guaranteed to work. Regardless of how we
actually encourage users to reboot their machine after upgrading to a
new systemd version, he suggested the following:
The bugscript should include whether the machine was rebooted since the
last systemd upgrade. This would allow us to discard issues that might
be caused by people not rebooting. A simple way of tracking this
information is creating a file in /run on upgrades and checking whether
that file exists in the bugscript.
I would’ve gone ahead and implemented that, but it’s unfortunately not
straight-forward, since we don’t get the current (or old) and new
version in postinst maintscripts. I can see the following options:
1) Just touch a file in /run uncoditionally on each postinst run. This
would introduce false-positives when a user is running systemd 204
and runs dpkg -i systemd_204-2.deb again (or uses apt-get install
--reinstall).
2) Put the package version number into postinst by using sed +
postinst.in or some mechanism like that.
Are there any objections to option 2?
--- End Message ---
--- Begin Message ---
From the associated MR
https://salsa.debian.org/systemd-team/systemd/-/merge_requests/68/
"In the early days of systemd I vaguely remember that we had much more
problems when systemd transferred state on a daemon-reexec. I do not
remember any such issues since at least buster, certainly not bullseye,
so I'm closing this MR (and the corresponding bug report)"
OpenPGP_signature
Description: OpenPGP digital signature
--- End Message ---
