For the last few weeks my home assistant crashes every Monday morning around 1:30AM. Automations seem to stop working, Samba and SSH access are down. The front-end is still available most of the times (not always), but the history is not working anymore. The supervisor tab also shows a blank screen. Trying to shutdown or reboot via server configuration doesn’t do anything. The only way to get things up and running again is by power-cycling the Rpi. Sometimes more than one power cycle is needed. After a successful reboot I see that the last history recordings have happened around 1:30AM. After reboot I cannot check the logs anymore what caused the crash. Before reboot I cannot see the supervisor or host logs since they are on the blank supervisor page. Before reboot I can see the core log, but so far I do not see the cause there. All I see is recorder errors, because the database is down. What would be the best way to find the root-cause? I tried disabling certain automations, configurations and integrations on Sunday evening and see if it crashes overnight. No luck so far and I feel that this trail and error route might take ages with only one trail a week. Is there a way to read back the logs before they are erased? (power-off hte Rpi after the crash, take out the SD-card, read the logs on my windows PC, before restarting the Rpi?) I read a lot off issues can occur due to corrupted SD cards. How can I check if the SD is OK or not? Any other Ideas on how to find the root-cause on these crashes?
Running Home Assistant OS on Rpi 3B+, regular updates of core and OS are done. Always running the latest or previous version. None of the updates so far solved the issue.
Hi, check out this thread, seems like exactly the same issue I’m having, I’m also running Home Assistant on a Raspberry Pi 4, this happens every Monday morning about the time you state, I’m thinking it occurs due to something happening based on start of week, I have garbage collection custom component as did the person who created the other post, do you use this too?
Thanks @captain_daveman for pointing me to that other thread. I did search for similar issues before opening this tread, but I apparently overlooked that one. To keep things in one place I’ll post in that tread from now on.
I have the same problem