Zabbix Proxy and Node difference

[Sparky]

Hi Guys

I've been through the manual, but I want to make sure I understand the difference between a proxy and a node correctly.

Having a second node means that the node will do all processing and placing of data into its own database, as well as passing all this data to the parent node for processing.

Having a proxy only collects data and passes it to the Zabbix server for processing and input into the database.

Am I understanding the setups correctly? Scenario I am looking at, I am monitoring 4 sites with average of 100-150 devices at each site. 50% of the devices have only a ping check, 30% using SNMP and the last 20% using Windows agents. What are the pros/cons for each type of setup?

Edit : My reason for looking at a node/proxy at each site is that the links between the sites are not currently very stable and it's affecting especially the ping checks.

[walterheck]

In your case, go with proxies and not nodes. Nodes are useful when you want to be able to do full administration on each location you run a node at. Also, nodes are a nightmare to administer properly

Proxies are basically just a place where data gets collected and buffered, then sent through to the main server.