Ad Widget

**dimir** · 03-04-2018, 11:10

2) Yes. It does need a database but you can choose from MySQL, PostgreSQL, Oracle, IBM DB2 and SQLite. Of course, any other than SQLite does not need to run locally.
3) Yes. The client (or more precisely, agent) talks to proxy and proxy talks to server. The agent does not distinguish proxy from server. If you like I can make things more complicated by adding that agent can work in passive and active mode, and the proxy can also be active and passive. For your case it could be easier to have Proxy in active and agents in passive mode. For active proxy you don't need to have open hole in the firewall for the server.
4) Yes, if I understood the question correctly. The Proxy is the separate process and server has built-in functionality to talk to it.
5) No, you don't have to. As I explained in 3) if you set up proxy in active mode it will initiate connections to the server for both pull requests and pushes.
6) Yes, basically. In Zabbix there are many ways to automate things: a) active agent auto-registration, b) network discovery (or agents discovery), c) low-level discovery (host prototypes), d) Zabbix API etc. On top of that you use Templates and Groups based on specifics of discovered hosts.

Join our community on IRC at #zabbix@FreeNode to get started. Or if you like to get more information about our commercial services send a note to [email protected] . We have partnership program that you might be interested in https://www.zabbix.com/partners

**Pito78** · 03-04-2018, 14:24

Hello Dimir, thanks for the quick reply, I tried to put into practice what you told me, but I could not. The proxy is seen by the server and is updated every 60 seconds. But I can not activate the agent. On the server (in the cloud) I opened the 10051 port ... I enclose the configuration of the proxy, of the proxy agent and of the client to be monitored, can you please verify that it has done everything correctly? Thank you so much Maurizio