If a poller goes down?

[stega]

Hello everybody,
I'm new in this forum and I have to thanks a lot of you for the threads that helped me for my environment configuration with Centreon 2.0 Rc3.

My environment:
1 Master server with Centreon, DB, Nagios3, etc and 10 host to ping
2 Slave pollers with only Nagios3 and 10 host to ping on each server

I configured correctly my slave pollers and they work fine!!

On Master I set the ping check fot the 2 Slave server's IPs.
If I put the power off on a Slave server I can see the correct "KO" status for the ping check done by Master but I still have "OK" status for the IPs checked by Slave powered off. I cen see that "last check" date is stopped to the time in wich I powered off the poller.
Is it correct that hosts checked by a slave still have "OK" status while checks are stopped?
Is there a way to set their status if they are not under monitoring?

I set their dependencies from Slave too.

Thanks a lot,
Stega

[YodaNC]

Hi,

It's the normal working.
You can write an event handler that will change the status of hosts monitored by your poller when it's not monitoring. You can set the event handler on a service which checks if Nagios is runing on the poller.

[stega]

Hello,
here I am again for asking you help about distributed monitoring.

My Environment:
1 Master server with Centreon, Nagios and DB
2 Slave Poller server with Nagios only

My Configuration (example):
Master --> checks if pollers are alive
Poller1 --> checks if IP 10.10.10.10 is alive
Poller2 --> checks if IP 20.20.20.20 is alive

Question:
If host 10.10.10.10, checked by Poller1, goes down, who manage its notification handler?
Who will manage the notification (by a customized or implemented action)?
The local Nagios on Poller1, or the Master when it will receive the new host status by the Poller1?

I hope that my question is clear and I hope that someone can help me.
Thanks for every answer you'll give me.
Stega