Hope anyone can offer insight to my error here: all was working fine until I updated my hassbian venv to python 3.7 and then stretch>buster.
the error is mentioning a ‘broken pipe’ - I rolled back from 1.6 to 1.5 using HACS to see if that helped, and i got substantially more in the log.
2019-10-06 21:50:20 WARNING (MainThread) [homeassistant.components.image_processing] Updating deepstack_object image_processing took longer than the scheduled update interval 0:00:05
2019-10-06 21:50:21 ERROR (MainThread) [homeassistant.helpers.entity] Update for image_processing.kitchen_person_counter fails
Traceback (most recent call last):
File "/srv/homeassistant/lib/python3.7/site-packages/urllib3/connectionpool.py", line 672, in urlopen
chunked=chunked,
File "/srv/homeassistant/lib/python3.7/site-packages/urllib3/connectionpool.py", line 387, in _make_request
conn.request(method, url, **httplib_request_kw)
File "/usr/local/lib/python3.7/http/client.py", line 1244, in request
self._send_request(method, url, body, headers, encode_chunked)
File "/usr/local/lib/python3.7/http/client.py", line 1290, in _send_request
self.endheaders(body, encode_chunked=encode_chunked)
File "/usr/local/lib/python3.7/http/client.py", line 1239, in endheaders
self._send_output(message_body, encode_chunked=encode_chunked)
File "/usr/local/lib/python3.7/http/client.py", line 1065, in _send_output
self.send(chunk)
File "/usr/local/lib/python3.7/http/client.py", line 987, in send
self.sock.sendall(data)
BrokenPipeError: [Errno 32] Broken pipe
really looking forward to this supporting my presence detection at home. thanks @robmarkcole you always make top components!
Robin, I’ve been following your CV tutorials with great results - thanks for this!
Question - can I just install hassio (image on SD card) on RPI4 and add Coral to it over USB-C connection? Would this be possible?
I did write a healthcheck in the Dockerfile; it just tries to solve an image every so often.
What I discovered is that even when the container is declared “unhealthy”, Docker didn’t automagically restart the container for me. I ended up adapting something I found that goes and looks for unhealthy containers and then restarts them. That seems to be working.
I will have to package that all up. I think I need to see what the latest is in regards to the flask server talking to the Coral stick; I think what I have hasn’t kept up with recent changes so I probably need to merge all that back together and make a pull request for @robmarkcole.
Likewise, the Home Assistant component. Since I last looked at all this, the DOODS thing has appeared as another alternative, and I’m wondering now what the “best” solution is to implement, in regards to adoption and going with the flow, such as it is…
oh didn’t notices that they release DOODS support for HA, DOODS is completely new to me, so could be an option
Robin’s work here seems to be a nice way to do it, but didn’t managed to get it running with the TPU - only CPU and that is alot of load when you want to use all 4 camera’s each of 1 FPS
For anyone watching this thread but who doesn’t own a coral USB stick I have created a fork which doesn’t require the stick. It runs tensorflow-lite on a pi and allows object detection via a (quite basic currently) HA integration. My plan is to merge this with the stick approach, so that there will be a single integration for tensorflow-lite, with optional hardware accelleration using the stick. For anyone interested the repo is below, cheers
Curiously, I switched over to DOODS about a month ago try with the home assistant component and I don’t seem to be having the problem with the Coral USB stick hanging. Of course, so many parts of the software stack talking to have changed, hard to say what accounts for the difference?
Getting error after hours of good results. Finally got this going, everything works great, it sees people, saves a picture. But after a few hours I get the error below on the “python3 coral-app.py” run. I restart it and all looks good till a few hours again and I get the error.
Any ideas on what I can do to fix this?
Exception happened during processing of request from ('192.168.0.163', 40694)
Traceback (most recent call last):
File "/usr/lib/python3.7/socketserver.py", line 316, in _handle_request_noblock
self.process_request(request, client_address)
File "/usr/lib/python3.7/socketserver.py", line 665, in process_request
t.start()
File "/usr/lib/python3.7/threading.py", line 847, in start
_start_new_thread(self._bootstrap, ())
RuntimeError: can't start new thread
Can you create an issue on the repo plz. I did have an intermittent issue that was resolved by using a higher ampage power supply, as described in the docs
running atm. noavx 
