zmfilter error producing infinite loop

[yajrendrag]

I have a script that runs when my filter detects an alarm. The script simply parses the path of the event files and builds a string that i publish to an mqtt broker.

Evidently, sometimes the script fails for some reason. But the problem is that zoneminder seems to re-execute the failed command - about every minute. i noticed this because my mqtt traffic is showing an alarm message from a specific camera every minute. But in the zoneminder GUI there are only a handful of alarms shown.

I modified my script to also contain the path to the zoneminder event and it's the same event over and over. For example, i see this message in the zoneminder logs:

Code: Select all

1526345270.397060	zmfilter	NULL	571	-2	ERR	Command '/var/lib/zm/scripts/alarm.sh /usr/share/zoneminder/www/events/4/18/05/12/19/52/42' exited with status: 1	zmfilter.pl	NULL

repeatedly - only the timecode changes. the path to the event (which is now over 2 days old) is the same.

I've tried killing the zmfilter.pl process, restarting zoneminder, but the filter processing of this event seems to continue forever. not sure how i can stop it.

I am running zoneminder 1.30.4 (regular release - ie, not a dev release) in ubuntu 16.04 which is running as an lxd container. i intalled using the easy way from here: http://zoneminder.readthedocs.io/en/sta ... untu-16-04

Any suggestions on how i can stop this filter from continuing to try and process this same event?

my log file is huge - 14M, but i will put it somewhere if the whole thing will be helpful..

thanks for any help.

[yajrendrag]

i finally got rid of the problem by deleting the event. the only thing unusual is that the event showed up in the event list in the gui with a

Code: Select all

(r)

after then event number. not sure what that is indicative of..

[iconnor]

The (r) indicates that it was resurrected by zmaudit.pl. So the db record was deleted, but zmaudit.pl found the files and re-created the db record.