Hey everybody! I have been setting up a new hassio instance and I am currently integrating my Blue Iris IP Camera NVR software into home assistant. All I am using is the live stream. Blue Iris already handles all of the recording and image processing. The problem I face is that although the config checks out and HA works for a few minutes (cameras do not load), the frontend looses connection and HA hangs. The only fix is editing the camera component out with Samba or configurator (if a port is assigned and doesn’t rely on Ingress and restarting home assistant from the hassio shell.
HA is running in a VirtualBox VM on an x86 machine using a stable release of the supplied Virtualbox disk image. I have tried this with/without the stream component, and with the camera platforms “generic” and “mjpeg”. Below are the relevant configuration.yaml entries. I have tested the URLs with ONVIF device manager and a browser to confirm they work.
Having exactly the same problem ( front end becomes unresponsive) if I enable a stream_source for my cameras. Running Hassio 0.92.0 image on an core-i7 mini PC. Resources (CPU or memory) don’t seem to be the issue.
Really odd. Yeah I was still able to access my addons such as The Lounge and my Samba share was still running fine. As far as I can tell, people have had issues like this with FFMPEG since 2017. Really hope it gets fixed.
Really odd bug too. If you refresh you get the frontend with an error that it can’t connect to Home Assistant but sometimes it will just be an unresponsive page. Really strange.
Just put the config back. Force refresh and a completely different browser just gave a dead page. The camera component definitely wrecks the frontend. I’ll check tomorrow if it stops automations too.
I have the very same problem. Configured generic camera, sooner o later HA hangs and i have to reboot it. With more cameras (i have still some to put in place and configure), the problem arises sooner.
Currently i had to remove all cameras from HA yaml files.
Cameras comes from Blue Iris.
With 0.92.2 was somewhat stable and working.
With 0.93 i’m down in hell.
For me, it wasn’t the camera configs themselves causing the hassio lockups: the stream: option in my config was the culprit. Three cameras were working fine for months. But once I added the stream setting to my config the problems started. Typically within an hour or so of using a stream capability such as streaming to the TV or viewing the live stream on my phone. Didn’t matter if I had recently rebooted or not, and only a power cycle would get it going again. Drove me mad for a while and I’m not sure how I failed to notice that the issue started with the introduction of tge new stream function.
Having removed the stream setting almost 3 weeks ago my hassio install is rock solid again, no issues. I can still view all my cameras but just can’t get a live stream or stream to the TV. That’s fine for now, stability is more important.
The lockups occurred on both a tinkerboard s and in VirtualBox on a Windows 8 machine.
I also have Blue Iris and if I use the stream: component then it just doesn’t work well with my setup. It lags pretty bad if the stream ever even starts. This is with both the stream from BI and with the stream from the camera(s) directly.
The strange thing is that I have some PTZ presets on a couple of my lovelace cards as outlined here:
…and if I hit any of the PTZ buttons when stream: is in the config then HomeAssistant crashes. I have to SSH in and use the CLI to restart.
If you remove “stream:”
you will get picture-glance that use still images refreshed every like 10 seconds, and if you click on them, you will get the video. Less features, but at least it works well.
I realized that with 1 or 2 cameras, you can use stream (if it works), but if you have 10 4-5MP cameras, there is no way you can render them all in a web page without putting your client on it’s knees.
Ive got the same issue. Would really love a fix as I’ve chromecasts on all my TV’s and 2 google smart displays which can stream the camera feeds. I’m using 0.95.4
Does anyone have a workaround for this? Even an automation to reboot after crash may help, anyone?