Abnormal shutdown process
Dear ladies and gentlemen,
Since 10 July 2014 my Debian system does not shut down well anymore. I
have worked many hours trying to solve the problem. Now I have decided
to seek some help from the Debian community.
I use the GNOME desktop environment to shut down my computer. My
computer then starts the shutdown process and reports the shutdown
process on my screen. Since 10 July 2014 it often reports that there is
a problem in terminating some processes. My computer then eventually
takes drastic measures to shut the system down. This succeeds and
finally the system is powered off. By powering the system off, all
messages on my screen are lost for careful study.
I have found a method for shutting down the system in such a way that
the messages remain on my screen. The method is to use the shutdown
command, with such options that the system is halted in the end, not
powered off. I also specified that the processes get no less than 10
seconds to end gracefully. The command and the resulting messages are
below.
# shutdown -t 10 -hH now
The system is going down for system halt NOW! (pts/1) (Fri Jul 25
21:36:18 201
INIT: Switching to runlevel: 0
INIT: Sending processes the TERM signal
[info] Using makefile-style concurrent boot in runlevel 0.
[ ok ] Stopping anac(h)ronistic cron: anacron.
[ ok ] Stopping virtual private network daemon:.
[ ok ] Stopping deferred execution scheduler: atd.
saned disabled; edit /etc/default/saned
speech dispatcher disabled; edit /etc/default/speech-dispatcher
Stopping kerneloops:
[warn] PulseAudio configured for per-user sessions ... (warning).
[ ok ] Stopping MTA: exim4_listener.
[ ok ] Stopping resolvconf ... done.
[ ok ] Stopping GNOME Display manager: gdm3.
[ ok ] Stopping Avahi mDNS/DNS-SD Deamon: avahi-daemon.
[ ok ] Shutting down ALSA ... done.
[ ok ] Stopping network connection manager: NetworkManager.
[ ok ] Asking all remaining processes to terminate ... done.
Currently running processes (pstree):
init --- rc --- startpar --- sendsigs --- pstree
|- rpc.statd
|- rpcbind
|- rsyslogd --- 3*[{rsyslogd}]
\- 2*[tracker-extract]
[FAIL] Killing all remaining processes ... failed.
[ ok ] Stopping rcpbind daemon ...
[ ok ] Stopping NFS common utilities: idmapd statd
[....] Deconfiguring network interfaces ... invoke-rc.d: ----------
invoke-rc.d: WARNING: 'invoke-rc.d firestarter restart' called
invoke-rc.d: during shutdown sequence
invoke-rc.d: enabling safe mode: initscript policy layer disabled
invoke-rc.d: ----------
done.
[info] Saving the system clock.
[info] Hardware clock updated to Fri Jul 25 21:36:32 CEST 2014.
[ ok ] Deactivating swap ... done.
[info] Will now halt.
[535.273160] System halted.
-
After "[ ok ] Asking all remaining processes to terminate ... done."
the system seems to hang for about ten seconds. When the option -t 10 is
omitted, the behaviour is the same. Does anyone have an insight in why
some processes do not terminate on the request? Additional information
is below.
On 10 July 2014 I installed a VPN client sofware package with GDebi
package installer, I tested it, and then I deinstalled it with Synaptic.
That day I also suffered a nasty bug in the Nautilus file manager. When
I removed two folders that were in my Downloads folder, all the contents
of my Downloads folder were lost. This was probably because my Home
folder was opened in another Nautilus window, and that Home folder
contained two folders that were identical to the ones I removed.
Because of the incident, I spent most time on checking and improving
the integrity of my system. The current situation is:
- Synaptic tells that no packages are broken.
- The command debsums -l tells that all packages have their md5sums.
- The command debsums -c tells that all packages are essentially
intact.
- The command debsums -se tells that of all the package configuration
files only two files have changed. These are
/etc/NetworkManager/NetworkManager.conf and
/etc/apt/apt.conf.d/10periodic. After inspection, I do not consider
these configuration files the cause of the problem.
- My filesystem has already been checked two times. This happens
automatically after each series of 25 mounts.
I do not have and never had the firestarter package installed.
Kind regards,
Maarten.
Reply to: