Yes, you can do this with the HTTP monitor, even if they're external.
You could create an ICMP node and assign the template to this node. If ICMP doesn't work, then the node will show as down in Orion, but the APM application monitor will still be polled.
Here's another option:
When you create a HTTP (or HTTPS) component monitor, you can specify the the URL you want to monitor. The default (http://${IP}:${PORT}/) uses the special value ${IP} which gets replaced with the IP Address. If you don't include the special variable, then there is nothing tieing this monitor to the node it is assigned to.
So, you could create a component monitor with a url ofhttp://www.google.com and assign this to any node and it will monitor google.com to see if they ever go down.
Thanks Josh. I've actually tried both approaches but I can't get either to work through our proxy. In the http monitor template, it specifies a proxy port. Is this supposed to be the port the assigned node uses to reach the URL? I have allocated a http monitor to a node and this http monitor specifies the proxy address. However, what I don't understand is how this poll actually works;
- is it the assigned node that polls and then just informs Orion of its ability to accomplish the task ?
- or, does Orion actually perform the poll on behalf of the assigned node somehow?
I can't quite get my head around it! Thanks in advance for clarifying this for me!
APM performs an HTTP request from the Orion Poller to the URL specified. If the URL has the text ${IP}, that is replaced with the IP Address of the target node. If the URL hasthe text ${PORT}, that is replaced by the port configured in the component monitor.
The Orion Poller is always the source of the connection. If the Orion Poller needs to use an HTTP Proxy to access the URL, you should enter that information in the configuration.
Does that make more sense?