Nagios Support Forum

HI I am working with NSCA but not yet implemented, I noticed in the MySql database tehre are issues as below wo0uls you please advice

May 28 12:44:39 oracle-mon ndo2db: Error: mysql_query() failed for 'INSERT INTO nagios_servicechecks SET instance_id='1', service_object_id='299', check_type='0', current_check_attempt='1', max_check_attempts='3', state='0', state_type='1', start_time=FROM_UNIXTIME(1275047079), start_time_usec='151180', end_time=FROM_UNIXTIME(0), end_time_usec='0', timeout='60', early_timeout='0', execution_time='0.000000', latency='0.151000', return_code='0', output='', long_output='(null)', perfdata='users=0;5;10;0', command_object_id='21', command_args='check_users', command_line='/usr/local/nagios/libexec/check_nrpe -H 192\.168\.213\.76 -t 30 -c check_users' ON DUPLICATE KEY UPDATE instance_id='1', service_object_id='299', check_type='0', current_check_attempt='1', max_check_attempts='3', state='0', state_type='1', start_time=FROM_UNIXTIME(1275047079), start_time_usec='151180', end_time=FROM_UNIXTIME(0), end_time_usec='0', timeout='60', early_timeout='0', execution_time='0.000000', latency='0.151000', return_code='0', output='

Best Regards, Dave

This indicates that NSCA is working. The check results are being added twice... Try sending the NSCA commands to another server or don't *always send them back to the server they came from.

* You should be using a service event to start/stop the use of NSCA... To prevent loops in NSCA, ala STP.

Hi I had not not yet got NSCA working at the point when the Insert failures occured, I am concerned that unique constraiints are being violated how can I deal with this ???

Also I am having extreem issues with NSCA When I try to set any parameters the daemon will not start if I rem them out it starts fine

password=xxxxx
encryption_method=1

Any guidance would be most welcome
Regards, Dave

A unique constraint on Time being violated... No, it's not that the same check command is being run twice at the same time, it's that the results are being submitted twice.

Multiple results for a single service at a the same time are dealt with during the result check run, each run should be at least 2 seconds apart and for a single service only the last result is processed... Though this assumes a bug in the scheduler that causes a check to be run more then once a check_period(5min). The only way for this to happen is via the command interface(AKA passive checks).

start_time=FROM_UNIXTIME(1275047079), start_time_usec='151180'
start_time=FROM_UNIXTIME(1275047079), start_time_usec='151180'