Icinga

[Kynao]

Before adding Zabbix, i suggest to have a look into Icinga https://www.icinga.org/.

For the choosed monitoring solution, i suggest to add/suppress hosts automatically into the monitoring when a vm node is created/suppressed/renamed
and groups/services into Opennode when they are identified/defined into the monitoring solution.

[activesys]

Thats the plan.

[Kynao]

Nothing more to say then
Did you have a look into Icinga ?

[ilja_l]

Hi, Kynao

Icinga looks very nice and we have added integration with it on the todo list. However, as our development effort is driven also by the practical needs and deployments we have to manage, Zabbix was chosen as the initial monitoring solution.

thanks,
Ilja

[Kynao]

Wow, impressive, i didn't suspect you could even take Icinga into account (even if i submited the idea ) because of your announced Zabbix support...

[ilja_l]

Well, our plan is to pull data from the monitoring systems, which makes integration with already deployed systems much easier. The goal is to provide something, which can be deployed on top/next to already existing systems, be it Icinga, Nagios or Zabbix. The latter was chosen by us as the first step because of its proven qualities, previous experience and large user base.

We have a small team, covering every possible monitoring system is not possible at this stage. Our product will continue being open-source and we will be more than happy to accept contributions, if somebody decides to implement support for a certain monitoring system.

[Kynao]

Icinga is based on Nagios, quite a large user base
Whatever, my understanding, as no one can cover every solutions (here monitoring), were you choosed one solution (here Zabbix) and only one to be available on another subjects.

You talk about pulling information from the monitoring solution. Is pushing still on the road map ? Here is an example of what i mean :

- You're an Opennode administrator living in los angeles
- Alice, your friend, is an Icinga administrator living in New York

One day, because you are in a good mood, you decide to create a vm on OpenNode called "Damned good Windows VM"
=> Alice will see in her Icinga monitoring panel, a new node appear and called "Damned good Windows VM" on which an automatic autodiscovery service is started. The discovered services are HTTP, HTTPS, SMTP, IMAP, LDAP in this example. As a result, these informations are given back and you in Los angeles, in front of OpenNode, you see that your "Damned good Windows VM" is now populated with new informations corresponding to the discovered services. You also could follow the provided link to go into Icinga and directly to the "Damned good Windows VM" node for the details.

This is a "add event", that is create a node on OpenNode, Icinga is informed and start adapted process to give back results to OpenNode.
There are other events in OpenNode like suppressing a vm, renaming a vm.
There are other events in Icinga like services changing on a given node over time.

[ilja_l]

Ok, thanks, now I understand your idea better. Yes, we are planning to use API provided by the monitoring system to inform it about certain events, e.g. creating a new host.