I’m considering migrating my smart home from Hubitat to Home Assistant Green. I have about 60 Z-wave and Caseta devices, Blue Iris, LG washer and dryer, Chamberlain garage door openers (using ratgdo/ESPHome), Google smart speakers, and a Lucid Air I’d like to integrate with HA.
Hubitat has been problem-free for two years, and software updates have never broken anything for me. I’m wondering if the same will be true for Home Assistant. I have a lot of projects and do not want to create a time sink. How has Home Assistant reliability been for you? Does it need regular maintenance? Do version updates tend to create problems once in a while?
Thank you, I appreciate hearing about your experience with HA.
You should check the ‘breaking changes’ (i REFUSE To calm them backwards incompatible changes) of every release and determine if the release is right for you.
You should plan on staying relatively current but you do not have to take every update.
My plan usually involves waiting for mid month on a 0.2 or 0.3 release or even skipping every other month avlfter I’ve reviewed breaking changes.
While you build you WILL break stuff all the time. Once you get to your stable product you can probably proceed relatively trouble free.
I have been using HA for about 4 years. I came from an Insteon / ISY environment and wanted to migrate from an end-of-life environment to something more flexible. I started out “playing” with Home Assistant to learn the different Advanced Installation Methods and ultimately settled on the DIY with Raspberry install (Raspberry Pi4 B with Home Assistant running on an SD card).
All my automations are done with Node-Red. This is a personal preference and there are other ways to do automations in Home Assistant (see Getting Started).
I consider my self technically competent and very capable of doing very complex automations but I try to keep things simple. I am very good about having good, reliable backups, reading the release notes, and as @NathanCu does, waiting for releases to settle - don’t update without being informed and being able to recover if something does go wrong. I have not had ANY downtime due to maintenance updates - I know others have but I believe they did not prepare properly.
All that said, I really love working with Home Assistant, this community, and learning a lot of new things. For me, this is a very enjoyable and worthwhile time sink.
There is an HA integration with Hubitat using HE’s Maker API. It will bring any or all of your devices from HE into HA. Then you can test the waters. It will also mean you won’t have to buy a separate Zigbee and ZWave radio yet that HA doesn’t have.
As you can see from loads of my responses, I do love HA… but wanted to add a different thought.
In my family HA-stuff all depends on ‘me’… if HA server goes down, I need to fix, if a switch does not work, I need to fix, if the screens donot show…etc. This might not be fully related to HA and more to home-automation in general… but HA does allow a lot of trickeries with dependencies on ‘those-that-know’
At the moment I am fine but I (close 60) am not introducing any oddities anymore and am removing dependencies on HA running
I went from Hubitat C5 to HA 3 months ago, the only reason for was I had to update the hardware to get the latest matter protocol and various other hardware updates, so crossroads for me, new Hubitat or dip into HA. I chose a mini PC as it was good for power and fairly cheap at £60
I had no issues with Hubitat, the dashboard was fairly restrictive and there was less UK device integrations than HA. I always felt the American market was more supported for device integration priority.
I think the Hubitat team is phenomenal at keeping on top of things and stability, but I’m always worried that it can fold or be sold like it was with smart things all those years ago.
With HA, there was a learning curve and it took a few weeks to get it into top order, about the same with Hubitat, but I have to say I’m impressed so far, there are numerous integrations, good user support and great dashboards, there are some features that are not so user friendly for beginners, but I do think Ha are trying to address this issue. Not everyone wants to learn yaml or spend hours trying to get things to work.
The updates can be too much, this is the issue with breaking, but generally if you have things working, little need to update.
I did run Hubitat and HA for a while together with the integration, just to ensure things were stable. I’m glad I did Hubitat 5 years ago it was a game changer, I don’t think HA was right for me then, but now I’m not tied to one bit of hardware and the community is massive.
Let us know how you go, take the red pill or blue pill😂
There are a few automations I’d like to be able to do, but can’t in Hubitat - examples are to get a voice notification from my Google speakers when the clothes washer or dryer are done, or close my garage door by command to Google Assistant or Alexa. In particular I’d like voice notifications on my Google speakers about motion from Blue Iris to not permanently stop music that is currently playing. Hubitat has been great but can’t do these things. Also some friends have been working on an integration with our Lucid Air cars that seems fun.
But I’m not as crisp-thinking I was 30 years ago, and I’m wondering if the learning curve and the fixing-what-isn’t-broken aspect of the journey from Hubitat would be worth the effort or not.
I would kind of go against the trend here and highly suggest that you don’t wait to update for several monthly releases (tho I do agree to waiting until later in the month to do the monthly update).
In theory it sounds reasonable - “if it ain’t broke don’t fix it” and all - but the issue then becomes that there is some breaking change in an integration you use caused by something external to HA or even some new integration or functionality that you want and then you will need to go thru every past release to figure out what all the breaking changes were in those so you can finally get up to speed with the current desired version. You can’t pick and choose what to update in HA. It’s all or nothing for anything in the HA core.
It’s just easier to update every month, deal with the changes needed (if any - most times there won’t be) and then be done with it.
However that does put you, like many of us, in the position that you need to pay attention to at least the release announcement thread to know if there are unintended issues that might effect you.
So my point is that HA is not a “set it and forget it” system unless you never want a new functionality and no integration that you use breaks.
If that’s your definition of a time suck then it might not be a good fit. It’s a time suck for me but I have the time and I enjoy the tinkering.
The pro is that HA is hugely flexible and powerful if you are willing to put in the time to learn a bit of yaml and Jinja2 (the templating engine). Neither are really that hard and you can start slowly enough that you don’t need to know any of that to do basic things and even some moderately advanced stuff too.
