I started having problems with the latest version of HA where it would become completely unresponsive for about five minutes. It is running as container install on a quad-core thin client with SSD. The machine resources were not overly taxed, but Python3 executing HA would swamp out two cores during this time. There are absolutely no entries in the logs or logbook during these periods, and the entries just prior showed little suspicious.
I started disabling integrations, and discovered that when GDACS was disabled, the problem stopped. When reenabled, the problem started up again, almost immediately. I have it disabled and no problems now for about 15 hours.
Sorry to hear that. I have not come across any such issues with the GDACS integration, and I cannot think of any way the integration would prevent the whole system to work properly. The integration library us asynchronous and should not block anything.
Without any details in the logs it’s hard to investigate the root cause.
Would you be able to re-enable the GDACS integration and also enable debug logging for the integration and its libraries, to see if the integration is actually doing something and gets stuck? Thanks.
logger:
default: info
logs:
homeassistant.components.gdacs: debug
aio_georss_client: debug
aio_georss_gdacs: debug
Sorry, I won’t be able to look at this for awhile. For the moment, I just wanted to post this in case someone else was having a similar problem, which is hard to debug. I will try to come GCSI to this in a couple weeks.