Slow shutdown in dependency-based boot system

[Sven Joachim]

On 2008-07-15 10:46 +0200, Petter Reinholdtsen wrote:

> Hm.  Can you add a call to 'ps -ef' in sendsigs, to report what
> processes are being killed?

These are the processes that get killed:

root        80     1  0 15:14 ?        00:00:00 udevd --daemon
daemon     955     1  0 15:14 ?        00:00:00 /sbin/portmap
statd      967     1  0 15:14 ?        00:00:00 /sbin/rpc.statd
root       984     1  0 15:14 ?        00:00:00 /usr/sbin/rpc.idmapd
102       1239     1  0 15:14 ?        00:00:00 /usr/bin/dbus-daemon --system
root      1386     1  0 15:15 ?        00:00:00 dhclient3 -pf /var/run/dhclient.wlan0.pid -lf /var/lib/dhcp3/dhclient.wlan0.leases wlan0


>  I see nothing obviously wrong with the
> shutdown sequence.  Are you using network file systems?

No, but the machine acts as an NFS server for data exchange with my
laptop.  They are not connected currently, though.

>> Any idea who could be the culprit?  I see that K10ifupdown and
>> K12mountoverflowtmp are new, while K07nfs-common and K11hwclock.sh
>> had been run before sendsigs in the old system.
>
> I doubt the K12mountoverflowtmp script is the culprit.  nfs-common
> might be related, but it seem unlikely.  ifupdown might be related,
> but it seem likely as it only clean up /etc/network/run/ifstate.

It is indeed nfs-common, if I call "/etc/init.d/nfs-common stop" before
the shutdown, sendsigs only takes one second again.  So why does it take
so long for sendsigs to kill rpc.{stat,idmap}.d?

Sven