HS instance Talk to second HS instance

[WanderingEngineer]

I found https://forums.homeseer.com/forum/ho...nnect-hs-to-hs thread where they had an event trigger to have a secondary monitoring HS instance do an action.

I was wondering if there's a way to have a "master" HS instance handling several houses/components, and sub-instances for other tasking.

In my case, I have 2 homes I need to monitor, one of which I would like subdivided into 2 parts: Permanent structure items (built in lights, locks, HVAC, etc.) and "take it with me" items (holiday lights, media triggers, TV switching, etc.

I know I'm asking for something more complex, but it's what my goal is.

Is there an easy(er) way to have one HS instance expose the elements it contains to another HS instance?

A hypothetical action: on a master unit, trigger an "I'm away" event, which would obviously turn off all the media items. I'd also like it to send an event to the other HS instances to lock all the doors, close garage, randomize lights in the house, etc.

Is this feasible without significant event handling? Can I expose (authenticated, hopefully) elements to a master unit?

[Dweber85rc]

Start with this. https://forums.homeseer.com/forum/3r...er-4-connector
The kicker here is you would need a direct connection(vpn) between the sites for this to work

[WanderingEngineer]

I have VPN connection, that's no issue at all. Thank you!

[ezmaass]

I guess the first thing I'd question is... do you NEED two HomeSeer instances? Or, do you just need one HomeSeer to control devices in two houses?

If I've understood your scenario, you have two homes, and you'd like to have certain events at your first home trigger control of devices at your second home and vice versa. If this is the case, I'd probably first look into keeping everything on a single system that just has visibility / control over the remote devices.

HS4 allows you to group controls (Categories, Rooms, Floors), so for management purposes, you could easily group controls in a way that logically allows you to visualize them as belonging to Home 1 or Home 2. As for controlling devices, I would imagine you could keep HS4 running at your primary home while simply placing a Z-Net Z-Wave interface at your second home anywhere on the network (it's just a Pi). This would assume a few things: (a) you have a VPN between the homes so HS4 could talk to the remote Z-Net interface, and (b) that Z-Wave is your main interface for devices... otherwise, you may need other remote interfaces.

This gives you the benefit of a single HS4 installation (and license), avoiding the complexity of trying to mirror stuff between systems, and obviously central control from one console. Additionally, you're not patching, upgrading, and running two HS4 instances.

The main down side of this approach? If your VPN goes down, the remote property loses all home automation. There would be no local HS4 instance to keep automation running there.

Another approach (assuming the above is unacceptable), short of mirroring devices and such across HS instances (which I'm admittedly not familiar with), may be to use something like MQTT or even the HS4 API to trigger events between the two systems. You could do something like create a virtual device on your second home HS4 called "occupancy" with a state of "home" or "away" - your primary HS4 could trigger a change in that virtual device state, which in turn, you may have the remote HS4 adjust various devices as a result - limiting your need to directly interact with ALL of the remote devices but rather just triggering higher level state changes that can then be acted upon by the local HS4.

[WanderingEngineer]

Thanks for the info!

I'll caveat all this by saying I'm brand new to HomeSeer, but not to automation or networking/controls. My day job is PLC/SCADA controls. I went with HomeSeer because I just do not want to fiddle around with chasing updates as I was with HomeAssistant/others. Anyway, I chose HomeSeer, and now I'd like to set it up and forget it, because I do this all day and want to have it done for me, as it were.

I've thought long and hard about need, and I think ultimately, yes, I need them to be their own instances; in fact, a future goal would be to have a "locked to each house" and a master station that talks to all of them and could trigger their events; a little bit like a cloud, except I'd have control of it. This master station would connect to equipment that would not stay with a house should I move (networking/Home Lab/NAS, etc.), and the rest that would remain with a sold house. I know, people may not want it, and in that case I'd be happy to take it with me, but I want isolation between the "house" and my "life".

In addition, I don't want to rely on the connection for automation to work, and I want to be able to have a two way trigger (like you said, a virtual device to say I'm home/away for both houses), and I'd like to be able to trigger that from either house. Having the local instance fully self-sufficient is, in my opinion, the way to go, and the whole point of using HomeSeer over others. The fact that it costs money just means I'm directly paying someone for their labor to program and design it, and in my view not a downside, though it is a cost.

There are also some things I do not want exposed outside of the local HomeSeer instance (I am not sure if it's possible), like I don't want unlock actions to be able to happen from the remote house, but I do want to be able to lock from either house, or see the lock state; having a controllable API/interface would potentially allow this, but like I said, I'm brand new to this, and the documentation wasn't immediately obvious.

Thank you again for your replies!

[aa6vh]

I had a similar situation when I converted from Vera to Homeseer, and wanted to continue to run Vera during the conversion. And I needed them to talk to each other.

It takes a little effort to set up, but once set up it is easy to modify and maintain. What I did was have the two systems send HTTP commands to each other, using their respective API to trigger events.

[WanderingEngineer]

Thanks! I see HTTP Connections a lot here; does HomeSeer handle HTTPS well/at all?

[wpiman]

I have Homeseer running on a VM at my house and a Znet there and at the beach house connected via a VPN. Works great so long as the VPN isn't down.

[ezmaass]

OK, I think for your criteria, having two distinct systems would be the way to go.

I value avoiding third party tech (such as plug-ins) as much as possible, since I (a) enjoy building from the ground up, and (b) believe it provides for better long-term sustainability and ease of configuration / operation. So, I'll throw that out there as a disclaimer since it'll be the lens through which I provide the following advice.

If I were in your shoes (with your stated goals), I'd stand up two HS4 instances - one at each property. I'd then designate one of them the "primary" instance - not a technical designation of any kind but a design designation, which all of my configuration would leverage as context to planning how events work. I'd probably look to bridge the two systems with an assured delivery protocol (queue technology of your choice) that can cope with a connection going up/down between the homes.

HomeSeer does have a JSON over HTTP/S API, which could do things like manipulate device states, interact with HSTouch, etc., but I'd probably prefer to stay away from HTTP as it's not fault tolerant (dealing with your VPN going up/down). This is where something like MQTT makes more sense.

Essentially, you'd just be running two HS4 environments and writing events to interact between the two via a reliable queuing protocol. As mentioned above, you could structure your controls in you "primary" HS4 in such a way to control both properties from one location. Some of those controls would act on "native" devices in the primary home, while others would issue commands to your remote HS4 in the secondary home, picked up by appropriate event handlers there, and acted upon.

If you wanted to go to the extreme scenario, you could use HSTouch to create your own interface that consolidated your home controls in exactly the way you wanted - e.g., screen layouts for each home, etc. This is what I do today - I have a wall mounted tablet with custom designed HSTouch screens with everything from blueprint views of the home through specialized screens to control things like Virtual Machines, my irrigation system, water control, fire alarm, etc. There's no reason you couldn't build exactly the custom view you'd like. It'll run on Android or Windows minimally (haven't checked other platforms), and you could even produce different views for different devices (e.g., your phone versus a wall tablet vs. a computer). Alternatively, you may just go for the simple, stock approach of logging into the web console of your primary HS4.

As for categorizing devices, yes you can do that, too. As mentioned earlier, each device can have a category, floor, and room meta-data associated with it. With some creativity, I'm sure you can come up with a naming convention that allows you to group your devices not only by home (1 or 2) but by asset class (permanently installed, moveable, etc.).

HomeSeer is very extensible if nothing else. I've found that most things are limited by nothing more than your imagination and creativity.