Hi all,
I run a very stable machine (Pi3/raspbian, about 1 year old) with logitechmediaserver 7.9.1~1491339164 on it
The machine has a cron entry that restarts LMS weekly at 4:00 on Sundays. The gist of it is:
Today I received a mail telling me LMS had failed. I looked in server log and LMS was caught in a restart loop. The problem happened at 4:00.
I ran /usr/sbin/squeezeboxserver --help and got this:
Which corresponds to:
I reinstalled logitechmediaserver_7.9.1~1492077569_arm.deb, so not the same version but one I had lying around in the machine.
This "solved" the problem as the server restarted ok this time.
The uptime of the machine was 78 days, dmesg or syslog didn't show anything extraordinary. Load was a bit higher than normal due to LMS restarting incessantly, but still very manageable for the system. LMS doesn't work on Sundays on that machine :)
I guess something happened on the way to saving the DB when LMS stopped, and then it failed to restart?
In any case I would like to know how to make my crontab restart script less naive and avoid this problem in the future. Any idea what might have happened and how to make restart more sturdy?
Thanks in advance.
I run a very stable machine (Pi3/raspbian, about 1 year old) with logitechmediaserver 7.9.1~1491339164 on it
The machine has a cron entry that restarts LMS weekly at 4:00 on Sundays. The gist of it is:
Code:
/usr/sbin/service "$SERVICE" stop
sleep 5
/usr/sbin/service "$SERVICE" startCode:
2018-10-07 04:00:02 squeezeboxserver_safe stopped.
2018-10-07 04:00:08 squeezeboxserver_safe started.
2018-10-07 04:00:15 Logitech Media Server died. Restarting.
2018-10-07 04:00:20 Logitech Media Server died. Restarting.
...Code:
Number found where operator expected at /usr/share/perl5/Slim/Schema.pm line 2890, near "} 5"
(Missing operator before 5?)
Bareword found where operator expected at /usr/share/perl5/Slim/Schema.pm line 2892, near "$)sDebug"
(Missing operator before sDebug?)
Unrecognized character \xF4; marked by <-- HERE after $at<-- HERE near column 8 at /usr/share/perl5/Slim/Schema.pm line 2893.
Compilation failed in require at /usr/share/perl5/Slim/Player/Song.pm line 18.
...Code:
tail -2885 /usr/share/perl5/Slim/Schema.pm | more
sub hasLibrary {
return $initialized;
}
sub _connect {
my ( $class, $dsn, $sql ) = @_;
$sql ||= [];
my ($driver, $source, $username, $password) = $class->sourceInformation;
...This "solved" the problem as the server restarted ok this time.
The uptime of the machine was 78 days, dmesg or syslog didn't show anything extraordinary. Load was a bit higher than normal due to LMS restarting incessantly, but still very manageable for the system. LMS doesn't work on Sundays on that machine :)
I guess something happened on the way to saving the DB when LMS stopped, and then it failed to restart?
In any case I would like to know how to make my crontab restart script less naive and avoid this problem in the future. Any idea what might have happened and how to make restart more sturdy?
Thanks in advance.