[quote=“kenm, post:96, topic:181820”]
large Z-Wave network that’s running rock solid with HS3 and I don’t want to mess with it right now. [/quote]
I hear that. Mine is quite large too. I have 64 nodes migrated over right now. In order to migrate over and still maintain approval factor I purchased a second Z-Wave stick. I moved devices over in batches and got the automations written over on Home Assistant for things I moved over. I still have 4 water sensors to bring over and a few other straggler devices.
My procedure was to remove node in HomeSeer then add node in Home Assistant. For the first of each kind of device I got my naming standards documented so that the entity ids in Home Assistant were easy to work with. For things that HomeSeer still had to know about I used MQTT to push over to virtual devices in HS3.
I believe I read somewhere that the Z-NET is a ser2net raspberry pi. Good call on not trying to get that into Home Assistant without an integration for it.
I actually had it working for a while today but not sure what I did to get it working. Then I had to reboot HA and lost all the switches again. Here is an example of one of my switches. It does match one of the device types you posted.
@kenm I am all set now. I built a new HA system running on a docker VM and all the HS3 switches and scenes are showing up without issues now. The system I was running before was the VMDK that I installed on my ESXi host so I am not sure why it had issues. Either way all is good!
After approximately 2 1/2 months with Home Assistant, I’ve decided to stick with HS3 for now. I need a system that allows me the flexibility to do what I need, which HA does a great job of, but I also need a system that doesn’t break when an update is applied. That’s been my experience over the last 2+ months. I’ve had several cases of OS updates just not updating as well as the system just “breaking” after a HASS update, even when the things that break aren’t listed in “breaking changes”. Luckily I’m running in a VM, so I can restart from a VM snapshot and keep running.
I won’t say I won’t try again in the future, but for now I need rock solid reliability and I can get that with HS3. We’ll see what HA (and HS4) bring down the road.
I have probably the same amount of time with Home Assistant but have come to a different conclusion. I am almost fully migrated over to Home Assistant and plan on retiring my HS3. I do have licenses for HS4 but they won’t now be used.
I have had the same experiences with Home Assistant but I believe the long term future development is worth sticking with it. I have had the same downtime with Home Assistant upgrades. A few weeks ago all my panels were broken due to the most recent upgrade. But… a few days later after all the HACS developers jumped on it everything was resolved.
When I look at these Home Assistant forums compared to the HomeSeer ones it is very different. Home Assistant has plenty of new development, ideas and integrations. HomeSeer to me at least is a dead end.
Home Assistant: Excellent community support. Bullet proof automation engine. I hate that my Z-Wave takes 15 minutes to come up after a restart. I hate that I have to restart to apply anything of significance in the configuration. Z-Wave is not that great but I have found workarounds for all of my devices to get them working. I have made beautiful panel screens with the Lovelace ui based on this A different take on designing a Lovelace UI I am planning on looking at a 3D panel view like this but for right now I just had to have minimum viable panels 3D Floorplan using lovelace picture-elements card What used to take forever in HSTouch Designer is now really quick in Home Assistant. In HSTouch I had to make buttons in an image editor for everything at all different resolutions for each tablet size. With Home Assistant everything is generated on the fly on each panel and all the dimensions are percentages of display resolution. This means one panel configuration supports many devices. As an example of this I have one panel configuration being used on Amazon Fire 8 tablets, iPad Pro and a Windows 10 tablet.
HomeSeer: Z-Wave excellent support. Bullet proof event engine. HSTouch no support in years and HSTouch Designer is a real clunker. Private company. HS4 appears like the answer to a problem I don’t have (mobile admin interface).
After further discussion with my wife over a glass of wine (actually two) I’ve decided to continue running HA but with a different strategy. For now, all devices and events will remain active in HS3 with HA becoming the UI server as I develop replacements for my HSTouch screens. Next to automated lighting, this is probably the biggest impact on the family. The UI will be more of a learning/prototyping mechanism and training/proof-of-concept for my family rather than a finished product. The reason for this approach is because all of the data will be coming into HA via MQTT messages from HS3.
That is a good idea using Home Assistant as a ui server. You could use one of the MQTT HomeSeer plugins to get state over to Home Assistant for display.
As an example I had the following in my Home Assistant configuration /sensors/hs2hass_time_of_day.yaml (using split configuration)
I definitely don’t blindly update HA. I carefully read the blog posts, specifically breaking changes, and I also stay a release or so behind to see what others stunble upon. The biggest thing I’ve had was when the UI just never comes back after an update. Even after a full system restart from the command line, I can’t get a web interface. It’s as if the web server doesn’t get started. This has happened twice.
HA’s development pace is exactly why I plan to stick around and see where things go but also why I just can’t use it as my production control system right now.
Notice the title of this topic. I have no intention of moving to HS4.
Hello all from a new HA user! I made the plunge into HA a little more than 24 hours ago after using HS since ~2000. I have HA up and running on an HP Prodesk 600 under Debian 10. Moved my Insteon and Zigbee devices over to it and loving the new UI! Now working on Z-Wave and attempting HS3:HA integration via the mathoc/homeseer pyHS3 approach
My Z-wave is a RaZberry in a Pi3B+ that I’ve been using with HS3/4 via ser2net on the Pi. I will be looking at continuing to use is as a pseudo-Z-Net directly in HA via: RPi as Z-Wave/ZigBee-over-IP server for Hass…
Success! All of my HS4 Z-Wave devices and events have populated!
There was an error in my copy/paste/edit for new homeseer: section in configuration.yaml (a rogue \t snuck in somehow). After enabling ASCII commands in HS4, a restart is required to effect the change.
Welcome! I think a lot of people have made the move from HS4 to Home Assistant. Personally my HS3 machine is now turned off. I’ll keep it around for a while in case I need to refer back to it. So far so good on Home Assistant. HomeSeer HS3 served me well over the years but HS4 is the answer to a problem I didn’t have. I don’t believe HS4 will get the level of development that Home Assistant gets and frankly HS4 looks botched. I would not trust my production system to it.
My user base has now become accustomed to the new panels around the house.
@dbrunt - Welcome to the party. The pyhs3/homeseer component is s great way to go as long as all of your device types are supported. I added some sensor support to pyhs3 a while back. For those devices not supported, MQTT is another option.
I’m using the mcsMQTT plugin to push sensor data over to HA for my new UI design. Using the Mosquitto add-on in HA as the broker.
Sticking with HS3 for automation while using HA as the UI server seems to be working out well so far.
HS4 was just officially released the day after I decided to venture into HA! I’ve been doing their beta thing since March but got fed up with their Alexa integration still not working 100% and their Zigbee beta plugin’s limited device support (10 or 12 I think). I embarked upon mcsMQTT, mosquito and zigbee2mqtt for my foray into Zigbee devices but getting that set up was while a useful learning curve, extremely tedious. The pyhs/homeseer component is working for me. I just had to change my Z-Wave motion detectors device type string from Z-Wave Notification to Z-Wave Sensor Binary and they showed up in HA so I can use them to automate my Insteon switches and Zigbee lights. It took me a while to figure out the HA automation process to turn on/off an Insteon switch. I eventually determined they are only entities and do not exist as devices. I was quite perplexed as to how to use them in the automation UI until I found some yaml code that referenced them as an entity. I copied/pasted/modified code into yaml and then looked at the automation via the UI and now I see how to do it…
Funny, I just made the leap 2 days ago and am doing automations impossible in Homeseer. HS4 was super disappointing, way worse than HS3, super buggy and the most basic stuff was hard to do (even compared to HS3). I sunk a lot of money into HS over the years, but it is what it is. I retired Homeseer last night after transitioning everything I needed over to HA.
I have sunk a lot of money as well into HomeSeer. It is what it is. I was happy to pay it at the time as I thought I was getting a well supported product.
I can’t remember the last time HS3 was upgraded let alone HSTouch Designer. When I used to call for some support I was directed to use the community forums.
Here on Home Assistant there are new substantive releases every 3 weeks. There is a sense of community and the Home Assistant integrations are far better supported than the paid HomeSeer plugins. I’m not sure now that the HomeSeer paid plugin model is sustainable. I really do appreciate the community in Home Assistant giving their time to improve it.
I might try HomeSeer HS4 as I do have it licensed. But then what is the point if everything is cut over to Home Assistant. My users are happy. My panels are massively more responsive than HomeSeer HSTouch / Designer.
Anyway. I think I just need to let it go. I was with HomeSeer for a long time and I think I’m disappointed that HS4 is a dead end in my opinion and not worth investing in.
This was pretty much the same situation I was in. I use a Z-Net which is basically a Pi with a Z-Wave module. Until it’s supported in HA I decided to keep my Z-Wave devices in HS3 and use MQTT for status updates to the HA UI.
