[Support]: Frigate may be causing my entire Home Assistant to randomly restart #15447
Replies: 1 comment 14 replies
-
|
is the camera set up with 1x interframe spacing? and constant framerate? |
Beta Was this translation helpful? Give feedback.
-
|
I've connected as you suggested So what do you mean with these numbers? Should I put this dimensions to my config? And to update my previous comment, my HA has restarted again recently... So no progress for now |
Beta Was this translation helpful? Give feedback.
-
|
Yes, pop them in there. detect:
enabled: true
width: 640
height: 480
fps: 5how many things do you have connected to the camera? another thread |
Beta Was this translation helpful? Give feedback.
-
|
Sorry for the late reply, but in the past 2 weeks, my entire Home Assistant is still restarting as I mentioned earlier. I tried adding the stream resolutions you suggested to my configuration, but unfortunately, it didn’t resolve the issue. The problem remains the same. To answer your question, I have:
Apologies if this image is messy, but here’s what my go2rtc Net shows: Since my last reply, I also made another change: I set the "Max Bitrate" for the high-quality stream in the Reolink interface to 3072 Kbps. |
Beta Was this translation helpful? Give feedback.
-
|
I have a similar issue with my Home Assistant completely restarting when running Frigate. Last week I started to migrate from a Raspberry Pi 4 instance of HA (with a running Frigate) that works perfectly to a HA generic-x86-64 (Intel N3700) that reboots every couple of hours (when running Frigate). I keep having to move my Coral Accelerator back from the x86 intel box to my Raspberry Pi box (and disable Frigate) to get stability back. I've tried the move 3 times now and each time HomeAssistant seems to crash/restart on an almost hourly basis. Just FYI the move has been done via a complete backup/restore, and also via a semi clean backup/restore. |
Beta Was this translation helpful? Give feedback.
-
|
It took me ages to re-find the error message, but it seems to have the same iHD_drv_video.so error. although the front door camera already has a bound for the fps: Checking further it seems this message it not present for all of the restarts of homeassistant. It only comes occasionally and I'm not sure it is related to the crashes in anyway. I cannot find anything related to the python threading issue above, so perhaps I have a totally different reason for the entire system crash? Note: I've moved my Coral accelerator from USB3 port to USB2 port and also ensured it was powered, and then unpowered. This has no affect. |
Beta Was this translation helpful? Give feedback.
-
Describe the problem you are having
For the past two months, my Home Assistant has been restarting randomly, and I haven’t been able to figure out why. I tried troubleshooting but had no luck. Eventually, I suspected some custom components might be causing the issue, so I disabled them one by one to see if anything changed.
Frigate appears to be responsible for the restarts. Could it have something to do with my CPU? I’m using an Intel(R) Core(TM) i7-2600K CPU @ 3.40GHz (a bit old, right?). I have a dashboard to monitor hardware stats, and the CPU usage has never been high. It sometimes peaks at 20% for a moment but stays very low most of the time. RAM also doesn’t seem to be an issue since it has 16 GB and is barely used. This PC was purchased as a pre-configured plug-and-play system from a local tech store.
I haven’t been able to figure out why Frigate is causing these restarts. I recently noticed that after shutting down Frigate entirely for a few days, Home Assistant stopped restarting. The restarts happen at completely random intervals—sometimes Home Assistant runs for 10 hours, and other times it restarts after just 5 minutes. There aren’t many logs to look at, aside from some Python threading errors.
Additionally, recordings occasionally stop with the following error message:
"WARNING: Unable to keep up with recording segments in cache for camera. Keeping the 6 most recent segments out of 7 and discarding the rest..."
Following this guide (link), I found that the copy times were under 1 second. This might suggest a CPU bottleneck, although CPU usage appears normal.
Should I consider setting up a separate PC just to run Frigate on my network? Or is there something else I should check?
Summary of Issues:
Version
0.14.1-f4f3cfa
Frigate config file
Relevant Frigate log output
Relevant go2rtc log output
Frigate stats
No response
Operating system
HassOS
Install method
HassOS Addon
docker-compose file or Docker CLI command
n/aObject Detector
Coral
Screenshots of the Frigate UI's System metrics pages
Any other information that may be helpful
What I’ve tried so far:
• Disabled recording temporarily.
• Switched from Frigate Card to WebRTC Card.
• Used the "Frigate (For old x86_64 CPU without AVX)" add-on.
None of above helped.
Hardware I'm using for now:
Reolink Doorbell PoE
Beta Was this translation helpful? Give feedback.
All reactions