Hello - I have 2 instances of Node-red running, talking to the same hub (one instance is test, one is live, both are on different platforms). On the MakerAPI configuration, the IP address of the node-red instance is specified. Does this mean that I need to configure another instance of MakerAPI? I suspect that this is required but wanted to confirm.
Also, is there any downside of running multiple configurations of MakerAPI? I currently have 2 (one for Homebridge, one for node-red) and will likely need a 3rd (for the test instance of node-red).
Yes, that is the correct/best way to do it, as you can only supply one destination IP/port for events to be sent to.
(OK... Technically if using the UNSUPPORTED websocket connection method for events you could do both node-red apps w/1 Maker API instance as you don't care about Maker API sending the events in that case)
More instances=more loading, at some base level. In then end loading is more a function of how Maker API is used than how many instances there are, though.
I've had as many as 8 Maker API instances running on 1 hub with no issues before.
I've got two instances of MakerAPI running on each HE (one NR, one HB). Exposing ~80-100 devices on each. My hubitats don't require routine reboots or any other deliberate intervention. Generally, they get restarted only when there's a platform update.
I was having some issues with devices on my main hub and during the height of the problems/for giggles created 1 maker instance for zigbee devices and another for z-wave. My thought at the time was to isolate requests to/from maker api to get around lockups/slowness affecting everything. One HE hub with multiple instances talking to a single Node Red server.. Seems to work fine even though probably unnecessary. Not your really situation but thought I'd mention it...
