Gearman Data Retention in Collectors

[slansing]

Well yes, it will start updating the central XI server once a connection is re-established, but it will only be with check results that came into the remote server after the connection resumes. It's not going to detect when a connection was severed, then lump up all that data and send it over the pipe and dump it into the main central server.

[rajasegar]

Well yes, it will start updating the central XI server once a connection is re-established, but it will only be with check results that came into the remote server after the connection resumes. It's not going to detect when a connection was severed, then lump up all that data and send it over the pipe and dump it into the main central server.

Ok. This is not the behaviour I was looking for.
About about the other questions?
2) Where does the alerts notifications get generated? The internal or central server?
3) Does this work with Nagios Core internal servers as well?
4) Any potential issues with this type of setup?

[slansing]

2) This is up to you, when you start pushing data from one server to the central one, you will effectually be adding those configurations to the main system as passive hosts/services, you can alert on these just as you would with any host/service pairing. You could also do this on the internal system if you wish, where the main checks take place, though latency between those two systems more than likely does not need to be worried about unless they are very far apart geographically.

3) Yes, you could set up a core server to run the checks, then push to XI, it is a bit more complicated though, XI already has the ability to push and receive passive results from other servers so very little configuration is required.

4) As I spoke briefly of in (2), latency is a potential issue when it comes to alerts, though usually a non-issue for most use cases. Also, as you had pointed out, if the link goes down, you will have to rely on the internal server for alerting/checking and would need to look at it rather than the central server, both could be pretty easily managed through a Nagios Fusion instance.

[rajasegar]

2) This is up to you, when you start pushing data from one server to the central one, you will effectually be adding those configurations to the main system as passive hosts/services, you can alert on these just as you would with any host/service pairing. You could also do this on the internal system if you wish, where the main checks take place, though latency between those two systems more than likely does not need to be worried about unless they are very far apart geographically.

3) Yes, you could set up a core server to run the checks, then push to XI, it is a bit more complicated though, XI already has the ability to push and receive passive results from other servers so very little configuration is required.

4) As I spoke briefly of in (2), latency is a potential issue when it comes to alerts, though usually a non-issue for most use cases. Also, as you had pointed out, if the link goes down, you will have to rely on the internal server for alerting/checking and would need to look at it rather than the central server, both could be pretty easily managed through a Nagios Fusion instance.

Ok.
BTW, I am sorry to say this but Fusion seriously lacks featues. I have also not heard any updates on the issues raised ages ago.

[slansing]

That is fine, I'm offering suggestions. If you have no further questions we'll be closing this thread up.