Hi,
I've configured Icinga2 in a distributed mode.
My main machine is in a datacenter, and will have a lot of satellites.
All of these satellites will be located behind very unstable connections,
implying a lot of flaps.
For now I only have one satellite and I’ve configured it with 3 zones :
datacenter, global and site1.
However, when the connection between the datacenter and site1 is down, no
alert are generated because the site1 satellite is running its own host
alive check locally.
I cannot configure the cluster zone check because I my node could be
unavailable for 10/20 minutes and the accepted lag is not configurable.
I really don't know if its advised to do that but I've been able to create
a service in my datacenter zone for the satellite in the site1 :
- which runs pings from the datacenter node
- which calls the event command "link_down" to submit a passive check
result (host down) for the satellite node
> apply Service "site_connection" {
> check_command = "delayed_ping"
> event_command = "link_down"
> check_interval = 2m
> retry_interval = 30s
>
> host_name = host.name
> zone = "datacenter"
>
> vars.site_id = host.vars.site_id
> vars.delayed_ping_fast_recovery = 1
> vars.ping_retries = 5
>
> assign where host.vars.is_sat
> }
Can I have something equivalent in the Host check_command directly and get
this check executed from the datacenter zone ?
Thanks !
Cyril
_______________________________________________
icinga-users mailing list
[email protected]
https://lists.icinga.org/mailman/listinfo/icinga-users
