On Sun, 2 Nov 2014 12:57:53 +0000 Justin B Rye
<justin.byam....@gmail.com> wrote:
Jim McCloskey wrote:
> Bernhard Úbelacker <bernha...@vr-web.de> wrote:
>>> Can someone else reproduce this issue?
[...]
> Is it significant that all of the reports of this problem that I've
> seen (not that there are very many) involve the amd64 architecture?
No, because my case was an i386 machine.
I see text in /etc/init.d/kexec-load:
# If running systemd, we want kexec reboot only if current
# command is reboot
followed by checks for
pidof systemd > /dev/null 2>&1
and
systemctl list-jobs systemd-reboot.service | grep -q systemd-reboot.service
which look as if they should work... if they were in kexec instead of
in kexec-load!
But no, if I copy that stanza into /etc/init.d/kexec, it doesn't seem
to help... the init.d script runs, but it doesn't detect that systemd
is running, so it doesn't check for systemd-reboot.service. Is it
possible that systemd has already closed down by that point?
--
I think the more relevant question here should be: Why does K*kexec*
scripts in rc6.d executed in a shutdown? They are never meant to be.
IMO this looks like a systemd bug rather than kexec-tools.
Another interesting bit is, yesterday I fired up a jessie VM for the
purpose of tracing this issue, and I was be able to reliably reproduce
it at the beginning. Afterwards I performed an apt-get upgrade, few
dozen packages got updated, and now I can't reproduce it *most* of the
time. It happened few times even after upgrade, though.
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
