Hi All
I am sorry for posting this message as I know that very similar threads have been posted before but none of them have really explained simply what the pros and cons of each install type are (in a way that I understand).
I have been running HA in a VM on synology for a while. The discs on my synology never span down and I assumed that this was because of the VM accessing them. Because of this I moved to HA OS installed on a RPI 4 with an SSD. I seem to have some performace issues with that though so I am now considering other options. I am currently setting up a Truenas box that will have some sold state storage so I am considering using the solid state drives for HA.
It seems that my hosting options are:
HA OS on RPI 4
Docker container on TrueNas
VM on TrueNas
I think that I probably want supervised - so that seems to rule out docker. Is that correct? I know that when I try HA docker and it boots up there is no restore from backup option that I want. I also want to run the node-red add on.
Am I correct in saying that given the above I need HA OS either in a VM or running on my RPI4?
Are there any pros and cons between running as a VM or on a RPI?
thanks for the reply. HAOS is supervised isn’t it? I can certainly see the supervisor on my Pi when I run it on that.
If I run in docker I can’t have add ons can I?
So it sounds like I want HAOS on either the PI (which has an ssd) or in a VM (like you said!)
I plan to run most stuff separately from HA - for example mqtt and zigbe2mqtt will be separate docker containers as well as one instance of node-red. BUT I think that I need to have another HA hosted version of node red to get the HA nodes within node-red. I never managed to get the magic to work that allowed the communication between HA and node-red without installing NR as a HA addon.
Ok, any pointers on how to get it to work? At the time I did significant research into the subject and wasn’t able to come up with any solutions or find any information about how to set it up.