[icinga-users] Distributed Monitoring otherwise

Szabó István szistvan at mail.vcsk.hu
Tue Dec 4 11:26:13 CET 2012


Hi!

Thanks for your response!

The problem is complex, I'm trying to describe it:
The design:
[Location _1] --- Internet --- [Center] --- Internet --- [Location _2]

[Location _1] ~ 20 devices and 10 services
[Center] ~ 40 devices and 25 services
[Location_2] ~  30 devices and 10 services

I would like to achieve:
In the above mentioned three locations independent Icinga observes and 
alarms.
In case of internet problems arrise, the local alarm works ( eg service 
stops).
However, we would like to arrive in [Center] data from every [Location], 
and in case network problems arrise the [Location] to be able to alert, 
independently.

Can be solved?


On Thu, 29 Nov 2012, Assaf Flatto wrote:
> If server 1 can not send to serve2 , the data remains on server 1 , server 2 
> (depending on your configuration) will either be dead/unaware of the issue/ 
> display an error stating that the data presented is invalid( see freshness ).
>
>
>
> On 29/11/12 09:07, Szabó István wrote:
>> Hi!
>> 
>> In the Distributed Monitoring description I read that  "Server #1",
>> "Server #2" sends the data to the "Central Monitoring Server".
>> What's happening with the data when "Server #1" cannot send data to the
>> center due to network problem?
>> It is possible that during a network failure data measured on "Server #1"
>> (eg CPU load) will be lost?

-- 
   (O__	  ------------------------------------------------------
   //\                       /      Varosi Csokonai Konyvtar
  //  )                     |       Tel.: 59/503-152
  V__/_    szistvan at tux.hu   \      szistvan at mail.vcsk.hu


More information about the icinga-users mailing list