Bug#1307: Netbase fails due to running daemons
Karl Ferguson writes ("Bug#1307: Netbase fails due to running daemons"):
> > Package: netbase
> > Version: 1.13-1
> Ok, the package works fine if I run the postinst script, no worries. But if
> ANY daemons it's meant to restart are running it'll barf the error. ie: if
> I didn't kill named/smail/inetd/rpc.portmap/rpc.mountd/rpc.nfsd it would barf -
> only AFTER I manaully killed all these process it installed fine. Sounds
> bad, but maybe a "killall -TERM named"... etc in the postinst script would
> be alright?
Ah. Is netbase using start-stop-daemon (as it should be) ?
chiark:~> start-stop-daemon --help
Debian GNU/Linux start-stop-daemon 0.93.68. Copyright (C) 1995
Ian Jackson. This is free software; see the GNU General Public Licence
version 2 or later for copying conditions. There is NO warranty.
Usage: start-stop-daemon --start | --stop | --version|--help options ...
Options: --test --oknodo --exec <executable> --pidfile <pid-file>
--quiet|--verbose --user <username>|<uid> --name <process-name>
--signal <signal> --startas <pathname>
-- <... all of the rest are arguments to daemon ...>
Be careful - try not to call without --exec. `start-stop-daemon --stop'
would send a SIGTERM to every process, if it weren't specially prevented.
Exit status: 0 = done 1 = nothing done (=> 0 if --oknodo) 2 = trouble
As you can see, start-stop-daemon gives an exit status of 1 if it
doesn't do anything. netbase should probably be using --oknodo.