I want to set up a second Home Assistant server to experiment with various cool looking custom cards, like Mushroom, but I don’t want to do my testing on my main Home Assistant server. Because I tend to break things.
Hi, this is not a bad idea at all and it is possible to do.
The only thing is that it’s not certain that you can have all your devices from your working config in your test setup.
Depending on your setup you can virtualize HA so you don’t need a fullblown system just for testing.
Thanks for the input. I don’t need all the devices, maybe a couple of ESPHome devices for experimenting with front-end stuff.
I will either need an AP on a different local network or to find where Home Assistant defines its node name. I can’t have two instances of homeassistant.local on the same net.
You can have as many HA’s as you want. I had four at one time. now I have three. They aren’t always running but I use them for testing, etc. I use a HA container install so I just spin up a new container and give them each a different port than 8123 in the config.
the only issues you will have is connecting to devices that need a local dongle to connect like ZHA. Zwave used to be an issue as well but with zwavejs that has been resolved since it’s now separate from HA itself.
It won’t be a problem. I just need a sandbox to test front-end coding. I am YAML-challenged, but completely illiterate with CSS.
I only use the X86 native image file- no containers (other than HAOS) and no virtual machines. How would I specify the port of the second instance? (And the hostname “homeassistant”)?
Correct port only matters if you are trying to run two services on the same IP address. If you’re doing a separate host (like ha-test.local) then you don’t need to (and really shouldn’t) change the default port unless you have another really good reason.
Stephen, if you install HA on 2 different machines you don’t have to worry about ports or hostnames at at all.
You just open each instance with < IP >:8123
I never used docker but with my setup I can have multiple containers exposing the webUI on the same port too since the IP differs anyway.
I’m very confused. I’ve been running HA for a couple of years and I want to redo the lot but I want to migrate stuff over to a new server. I built a new HA server installed MQTT/Tasmota Integration etc. My Tasmota devices are all running v14. I attempted to move one over to the new server by changing the MQTT server address in the device but it’s not working.
I don’t understand what is the fundamental thing that dictates which HA server a Tasmota device connects to if they are both running the Tasmota Integration?
Do I even need to run MQTT config on the tasmotas? I’m using SetOption19 0.
There’s so much legacy info out there it’s really difficult to find current valid factual advice. Any help greatly apreciated.
Are you are trying to run two instances of Home Assistant on the same network? (That is the title of the thread).
If you are simply trying to migrate your Home Assistant server to a new host computer, then you are doing it the hard way.
Step 1- On your current Home Assistant, make a full backup. Save the backup .tar file to a flash drive.
Step 2. Shut down the original system.
Step 3. Install Home Assistant to your new host.
Step 4. Restore your Home Assistant configuration from the backup .tar file on the flash drive from step 1.
New version just released for mqtt payload and entities are setup differently in HA and the old config will not work. Resinchem tech just released how too.