Hello 
we are using the JIRA module to send notifications and are experiencing some issues in our clustered setup.
Setup is:
2x HA master
2x web servers (separate form master)
To be able to actually send the notifications we needed to install icingaweb2 and the JIRA module on the master server as well, which we did.
We also added the needed permissions
chown icinga -R /usr/share/icingaweb2
chown icinga /etc/icingaweb2 -R
usermod -aG icingaweb2 icinga
With this done the notifications are sent when a check changes to critical and create the tiket correctly, but produce the following error message in the log:
[2021-01-15 11:23:45 +0100] information/Notification: Completed sending 'Problem' notification 'blub-ic-sl01!service-chronyd!jira' for checkable 'blub-ic-sl01!service-chronyd' and user 'company-ticketsystem' using command 'JIRA Service Notification'.
[2021-01-15 11:23:47 +0100] warning/PluginNotificationTask: Notification command for object 'blub-ic-sl01!service-chronyd' (PID: 4040191, arguments: '/usr/bin/icingacli' 'jira' 'send' 'problem' '--description' 'PROCS CRITICAL: 0 processes with command name 'chronyd' ' '--host' 'blub-ic-sl01' '--issuetype' 'Incident' '--project' 'BLUBCMS' '--service' 'service-chronyd' '--state' 'CRITICAL' '--summary' 'service-chronyd on blub-ic-sl01 is CRITICAL') terminated with exit code 255, output: PHP Fatal error: Uncaught Error: Class 'Icinga\Module\Monitoring\Object\Service' not found in /usr/share/icingaweb2/modules/jira/library/Jira/IcingaCommandPipe.php:58
Stack trace:
#0 /usr/share/icingaweb2/modules/jira/library/Jira/IcingaCommandPipe.php(41): Icinga\Module\Jira\IcingaCommandPipe->getServiceObject('blub-ic-sl01', 'service-chronyd')
#1 /usr/share/icingaweb2/modules/jira/library/Jira/IcingaCommandPipe.php(16): Icinga\Module\Jira\IcingaCommandPipe->getObject('blub-ic-sl01', 'service-chronyd')
#2 /usr/share/icingaweb2/modules/jira/application/clicommands/SendCommand.php(110): Icinga\Module\Jira\IcingaCommandPipe->acknowledge('JIRA', 'JIRA issue BLUB...', 'blub-ic-sl01', 'service-chronyd')
#3 /usr/share/php/Icinga/Cli/Loader.php(265): Icinga\Module\Jira\Clicommands\SendCommand->problemAction()
#4 /usr/share/php/Icinga/Application/Cli.php(162): Icinga\Cli\Loader->dispatch()
#5 /usr/share/php/Icinga/Application/Cli.php(152): Icinga\Application\Cli->dispatchOnce()
#6 /usr/bin/icingacli(7): Icinga\Application\Cli->dispatch( in /usr/share/icingaweb2/modules/jira/library/Jira/IcingaCommandPipe.php on line 58
Fatal error: Uncaught Error: Class 'Icinga\Module\Monitoring\Object\Service' not found in /usr/share/icingaweb2/modules/jira/library/Jira/IcingaCommandPipe.php:58
Stack trace:
#0 /usr/share/icingaweb2/modules/jira/library/Jira/IcingaCommandPipe.php(41): Icinga\Module\Jira\IcingaCommandPipe->getServiceObject('blub-ic-sl01', 'service-chronyd')
#1 /usr/share/icingaweb2/modules/jira/library/Jira/IcingaCommandPipe.php(16): Icinga\Module\Jira\IcingaCommandPipe->getObject('blub-ic-sl01', 'service-chronyd')
#2 /usr/share/icingaweb2/modules/jira/application/clicommands/SendCommand.php(110): Icinga\Module\Jira\IcingaCommandPipe->acknowledge('JIRA', 'JIRA issue BLUB...', 'blub-ic-sl01', 'service-chronyd')
#3 /usr/share/php/Icinga/Cli/Loader.php(265): Icinga\Module\Jira\Clicommands\SendCommand->problemAction()
#4 /usr/share/php/Icinga/Application/Cli.php(162): Icinga\Cli\Loader->dispatch()
#5 /usr/share/php/Icinga/Application/Cli.php(152): Icinga\Application\Cli->dispatchOnce()
#6 /usr/bin/icingacli(7): Icinga\Application\Cli->dispatch( in /usr/share/icingaweb2/modules/jira/library/Jira/IcingaCommandPipe.php on line 58
I assume that this message is created when the module/icinga/who ever tries to acknowledge the problem in the webinterface.
Furthermore recovery notifications are not appended to the existing JIRA ticket.
Also if there is already an existing AND open JIRA ticket for “host!service” and the check changes to critical again a new ticket will be created.Though I’m not sure if this is by design.
Any help or insights are much appreciated