You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Currently, I have a service configuration like this::
root
↳Cachet
↳Gameserver
↳Minecraft
↳GMod
Where "Cachet" is the configured root for cachet, so "Gameserver" are a component group and the services below are the actual components. These services (in this case Minecraft & Gmod) have configured a trigger which results in a status message being created by this zabbix-cachet-integration.
This is pretty limiting, since for every service/component there can only be a single trigger.
In Zabbix, you'd normally just create children for the services which would contain different triggers - but these children aren't being detected by this script. It'd be pretty great to have that kind of functionality.
You could configure different triggers per component, which would result in different status messages and severities. For example, a trigger for high latency could result in performance problems, whilst a server being fully unreachable would result in full failure status.
The text was updated successfully, but these errors were encountered:
Supplement: Would be great if different triggers would trigger different severitys in cachet (Operational/PartialOutage/MajorOutage/Performance issues). Also, if status messages created manually wouldn't be overwritten by the script.
I totally agree with this, we have the same problem.
For example we have
root
-SMS
--SMS Operator 1
--SMS Operator 2
We have all 3, SMS, SMS Operator 1 and SMS Operator 2 displayed in Cachet, but unless you are logged on, you will only see SMS. So if there is an error in SMS Operator 1, a non-logged in user will see the parent SMS as fully operational which is wrong.
In Zabbix when one item in Services is having and issue, all parent items would be seen as having an issue. I think to be usable zabbix-cachet needs to adhere to this behaviour. So in my example if SMS Operator 1 has an issue, both SMS Operator 1 and SMS needs to be set in problem state.
Currently, I have a service configuration like this::
Where "Cachet" is the configured root for cachet, so "Gameserver" are a component group and the services below are the actual components. These services (in this case Minecraft & Gmod) have configured a trigger which results in a status message being created by this zabbix-cachet-integration.
This is pretty limiting, since for every service/component there can only be a single trigger.
In Zabbix, you'd normally just create children for the services which would contain different triggers - but these children aren't being detected by this script. It'd be pretty great to have that kind of functionality.
You could configure different triggers per component, which would result in different status messages and severities. For example, a trigger for high latency could result in performance problems, whilst a server being fully unreachable would result in full failure status.
The text was updated successfully, but these errors were encountered: