I have been installing each alpha build and testing. While I really haven't had much to offer, since the earlier builds were clearly not ready for production use. I installed .52 today and since it is going out for a wider beta, I thought I would test again with my HS3 database and plug-ins. The first thing i noticed is that HS4 comes to a dead halt when starting up with an unfiltered Device View with all of my devices. My current HS3 configuration is:
Current Date/Time: 12/10/2019 4:16:04 PM
HomeSeer Version: HS3 Pro Edition 3.0.0.548
Operating System: Microsoft Windows 10 Pro - Work Station
System Uptime: 0 Days 0 Hours 2 Minutes 57 Seconds
IP Address: 192.168.2.36
Number of Devices: 3208
Number of Events: 1625
Available Threads: 798
HS4 is running on a different but identical machine. I installed all HS3 plug-ins and registered them under HS4. That all worked. Then I copied my scripts, config files, html directory and database from HS3. HS4 converted it from .hsd to .json and moved it to a new folder
After allowing Chrome to time out on the page 3-5 times, I was finally able to apply some filters so that the page would load normally. I noticed that the "All" choice on both locations is located alphabetically within the list, instead of at the top where it should be.
I also noticed that the filters are not persistent. When I restarted HS4, I was back to an unfiltered display, which caused it to come to a dead halt again while loading all of my devices. While it is in this state, the web browser is totally locked, I cannot launch setup, tools, events, etc. Running HS3, my large number of Devices does slow it down a bit. The initial unfiltered loading of the Device Manager can take just over a minute, but HS4 can take 12-13 minutes or more before I can apply a filter to reduce the number of that are displayed. Since the filters are not persistent, I will be faced with this every time I relaunch a browser view of HS4. This is on a reasonably powerful 4-core hyperthreaded Xeon powered server. The CPU usage remains below 20%, so I don't think it is a processor bottleneck. Here is the system resources while the page is still loading
I also noticed that when I initially went to the (legacy) plug-in management page, I was able to disable and enable a plug-in. When I went back, they all showed "Loaded Remotely".
I will continue to play and provide more info as I test it. There are problems with some plug-ins and scripts, but I will have to start over so that I can start each of them one at a time.
Current Date/Time: 12/10/2019 4:16:04 PM
HomeSeer Version: HS3 Pro Edition 3.0.0.548
Operating System: Microsoft Windows 10 Pro - Work Station
System Uptime: 0 Days 0 Hours 2 Minutes 57 Seconds
IP Address: 192.168.2.36
Number of Devices: 3208
Number of Events: 1625
Available Threads: 798
HS4 is running on a different but identical machine. I installed all HS3 plug-ins and registered them under HS4. That all worked. Then I copied my scripts, config files, html directory and database from HS3. HS4 converted it from .hsd to .json and moved it to a new folder
After allowing Chrome to time out on the page 3-5 times, I was finally able to apply some filters so that the page would load normally. I noticed that the "All" choice on both locations is located alphabetically within the list, instead of at the top where it should be.
I also noticed that the filters are not persistent. When I restarted HS4, I was back to an unfiltered display, which caused it to come to a dead halt again while loading all of my devices. While it is in this state, the web browser is totally locked, I cannot launch setup, tools, events, etc. Running HS3, my large number of Devices does slow it down a bit. The initial unfiltered loading of the Device Manager can take just over a minute, but HS4 can take 12-13 minutes or more before I can apply a filter to reduce the number of that are displayed. Since the filters are not persistent, I will be faced with this every time I relaunch a browser view of HS4. This is on a reasonably powerful 4-core hyperthreaded Xeon powered server. The CPU usage remains below 20%, so I don't think it is a processor bottleneck. Here is the system resources while the page is still loading
I also noticed that when I initially went to the (legacy) plug-in management page, I was able to disable and enable a plug-in. When I went back, they all showed "Loaded Remotely".
I will continue to play and provide more info as I test it. There are problems with some plug-ins and scripts, but I will have to start over so that I can start each of them one at a time.