Bug#571280: false positive for "init.d-script-missing-dependency-on-remote_fs /etc/init.d/rsyslog: required-stop"
Michael Biebl <biebl@debian.org> writes:
> I don't think your analysis is correct.
> The rsyslog header has a
> Required-Stop: umountnfs
> i.e. it is stopped before umountnfs.
> As /usr can be a remote fs in Debian, this means, rsyslog is stopped
> before /usr (or for that matter /var) is unmounted (fwiw the rsyslog LSB
> header wouldn't be correct, I certainly would have gotten a bug report
> until now :-) )
> What the rsyslog init script does though, is using the sendsigs omit
> interface, to not be killed by the sendsigs init scripts, so it can run
> longer during shutdown.
> I'm not quite sure where you see me saying that /usr or /var could be
> unmounted while rsyslogd is still running?
Oh, I see. You're explicitly depending on some of the components of
$remote_fs instead of using $remote_fs because $remote_fs includes
sendsigs, which you don't want. Sorry, I didn't look deep enough.
Yes, this should be fine and Lintian should cope. Will be fixed in the
next version.
--
Russ Allbery (rra@debian.org) <http://www.eyrie.org/~eagle/>
Reply to: