You are not logged in.
Hi!
Little problem here..
When I shutdown (from gnome) the system, the daemons aren't stopped correctly but just killed.. Infact the "[DONE].." screen doesn't appears and at the reboot daemons like amuled said: "remove the ~/.aMule/mulelok file if amuled isn't running.."
How can I resolve? Thank you.
Last edited by dvd100 (2008-12-07 20:10:30)
Offline
up..
Offline
I think the problem is amuled, not all daemons. IMO, it's a matter of time. I mean amuled needs a lot of time to close all connections and save all data. Probably the solution comes from to stop amuled manually or insert a sleep sentence in the start/stop script.
Scientia et sapientia patrimonium humanitatis sunt.
Offline
Thanks for the answer.
I'll try with a 'sleep' sentece!
But why gdm doesn't save the session?
Offline
Ok, solved adding "sleep 10s" in /etc/rc.d/amuled.
For gdm this is the bug: http://bugzilla.gnome.org/show_bug.cgi?id=552387 I'll wait..
Offline
were exactly are you adding sleep 10 to the amuled script?
i use amule-cvs and use the start/stop script from AUR
#!/bin/bash
if [ "$(id -u)" != 0 ]
then
echo "You must be root to start aMule as daemon."
exit 2
fi
# general config
USECOLOR="YES"
. /etc/conf.d/amule.conf
. /etc/rc.d/functions
if [ -z $AMULEUSER ]
then
echo "Unable to run aMule. You must set the user in /etc/conf.d/amule.conf."
exit 2
elif [[ $AMULEUSER == "root" ]]
then
echo "Unable to run aMule. You must not run aMule as root."
exit 2
fi
confdir=$(su -l $AMULEUSER -c "echo ~")/.aMule
logfile=$(su -l $AMULEUSER -c "echo ~")/.aMule/amuled.log
lockfile=$(su -l $AMULEUSER -c "echo ~")/.aMule/muleLock
PID=`pidof -o %PPID /usr/bin/amuled`
[ -f $logfile ] && rm -f $logfile
case "$1" in
start)
stat_busy "Starting aMule"
if [ -f $lockfile ]
then
stat_fail
echo "Unable to run aMule, it is already running. If you're sure aMule is not already running, you can remove ${lockfile}."
exit 2
fi
[ ! -d $confdir ] && su -l $AMULEUSER -c "mkdir -p $confdir"
[ -z "$PID" ] && su -l $AMULEUSER -c "/usr/bin/amuled -o &> $logfile &"
i=0
while [ "$i" -lt 20 ]
do
if [ ! -z "$(grep "No other instances are running." $logfile)" ]
then
add_daemon amule
stat_done
exit 0
elif [ ! -z "$(grep "There seems to be an instance of aMule already running" $logfile)" ]
then
stat_fail
exit 2
else
i="$(($i+1))"
fi
sleep 1
done
stat_fail
;;
stop)
stat_busy "Stopping aMule"
if [ -f $lockfile ]
then
[ ! -z "$PID" ] && kill $PID &> /dev/null
else
stat_fail
echo "Unable to stop aMule. It is not running."
exit 2
fi
if [ $? -gt 0 ]; then
stat_fail
else
rm_daemon amule
stat_done
fi
;;
restart)
$0 stop
sleep 1
$0 start
;;
*)
echo "usage: $0 {start|stop|restart}"
;;
esac
exit 0
.
I get corrupt (journal recover) data at reboot because its closing down too fast.
help! Thank You!
Last edited by gav616 (2009-01-06 17:28:34)
Offline
stop)
stat_busy "Stopping aMule"
if [ -f $lockfile ]
then
[ ! -z "$PID" ] && kill $PID &> /dev/null
else
stat_fail
echo "Unable to stop aMule. It is not running."
exit 2
fi
if [ $? -gt 0 ]; then
stat_fail
else
rm_daemon amule
stat_done
fi
sleep 10
;;
Offline
that works when i reboot (i think)
but when i just shutdown its not 'sleep'ing to let amule shutdown properly, thus, i get my journal recovering data from the fast shutdown of amule.
do i need to ad that extra 'sleep 10' above, and then change all the other 'sleep's from '1' to '10' as well?
Last edited by gav616 (2009-01-15 14:04:44)
Offline
I don't know.. I suggest to ask to the amule package mantainer. Probably he wrote the init script and he could fix it..
Offline
Speaking of the package maintainer:
When I took the liberty of making that package I had to make a init script - since I have very little experience about bash - the approach I took was to study the following init scripts:
- the arch init script for mldonkey, gdm and some others
- the debian init script for mldonkey and amule
- the gentoo init script for mldonkey and amule
Now, I know that amule-remote exhibits the behavior (*) you describe here - and so does mldonkey.
I have a dedicated headless 24/7 server that run these all the time for real, so I can try and fix this.
But since I do not have any deeper knowledge of bash, the method I will use will be "trial & error".
I may succeed, I may not.
(*) Except for the data loss.
I have never experienced data loss due to this.
Last edited by wantilles (2009-01-15 21:18:00)
Offline
yesterday I could stand it no more
since the server has good uptime I stopped caring about, but each reboot amule wouldn't start complaining for the mulelock left by fast shutdown.
So, this is my 'stop' section, basically instead of waiting a fixed amount of time we wait 1 sec and then we look again for mulelock existence, but for a maximum amount of ten times to avoid amule preventing machine from shutdown
stop)
stat_busy "Stopping aMule"
if [ -f $lockfile ]
then
[ ! -z "$PID" ] && kill $PID &>/dev/null
count=0
while [ -f $lockfile ] && [ $count -lt 10 ];
do
sleep 1
let count=$count+1
done
else
stat_fail
echo "Unable to stop aMule. It is not running."
exit 2
fi
if [ $? -gt 0 ]; then
stat_fail
else
rm_daemon amule
stat_done
fi
;;
Hope it helps,
bye
Offline