I am planning to run openHAB as a VM on my Proxmox Server.
This VM will run a headless Arch Linux (latest version) with 1 GB of RAM and 12 GB of SSD storage (4 GB will be preserved for SWAP) on which I am going to install openhab. I hope this is enough?
Since I am very concerned about the security of my smarthome, I have a separate VLAN where all my smarthome devices can do their thing.
I was asking myself if it is possible to assign openHAB network two interfaces.
Interace_trust will connect to my trusted lan, so I can manage my smarthome components using openHAB, without a direct connection to the components themself. Interface_iot will communicate with the smarthome devices, without having access to my trusted lan network.
If the above is not possible, I will simply put the whole openHAB VM in the iot_vlan.
You may want to think about running it in a container vs a full VM, I’m testing OH3 currently in one and its staying under 512 MB of memory, 384 swap, and 8 GB hard drive space; haven’t noticed any issues with the OH interface side, granted this is a test system and doesn’t have my zwave hardware talking directly so mileage could vary.
A lesson I learned with Wemo and Nest, a separate VLAN while good in practice some IoT devices will not allow direct communications and force OH to talk to their cloud APIs. Not all devices are like this, so take inventory before you go this route.
@pascal’s solution would be a litte bit to overkill but yeah it would also do its thing^^
@anonymous.one
Not really a fan of containers, I don’t hate them but I don’t really like them either.
I am the kind of guy that likes to keep thing isolated. Also RAM/CPU/Storage usage is not an issue on my server.
But thank you for the hint with the smart devices!
I am currently only using products from Shelly. https://shelly.cloud/
The can be run with their cloud or without any cloud/hub or using something like openHAB.
The best thing is: There products are opensource.