Data flow stops ESP32

[gptech2444]

Hi, Around 12.30am last night I noticed interruptions to the data reaching HA. Around every 5-10mins the ESP32 resets. Everything has been working great for the last 2 weeks since setting this up. Nothing has changed, I'm not sure why it now has this problem. Attached is the log. logs_sumry10kw_logs.txt

[syssi]

The simplest solution to get an idea what's going on is attaching a PC to the serial line (GPIO1/GPIO3 used by the logger) of the ESP and record a log including the log messages of the bootloader + exceptions/stack traces of the framework.

Please be aware of some reboot_timeouts. One reboot_timeout of 15min is introduced by the mqtt component (See https://esphome.io/components/mqtt.html#configuration-variables -> reboot_timeout). If there is no active MQTT connection for 15 minutes the mqtt component triggers a reboot.

The same applies for the wifi component. If the ESP isn't associated to an accesspoint for 15 minutes the component triggers a reboot: https://esphome.io/components/wifi.html#configuration-variables -> reboot_timeout

This also applies for the api component if there is no native api connection. Your log looks like you aren't using the api component.

[gptech2444]

Ok, I'm not using the api as it isn't included in the pip8048 yaml and I want to be able to change the inverters settings from HA. The issue went away for a couple of hours and then it started again and was resetting itself 3 more times and then it's been ok for the last 20 hours. I've not touched or changed anything. This message is the one that stood out on the resets.

[10:01:31][C][mdns:117]: Hostname: sumry10kw [10:01:31][C][esphome.ota:073]: Over-The-Air updates: [10:01:31][C][esphome.ota:074]: Address: sumry10kw.local:3232 [10:01:31][C][esphome.ota:075]: Version: 2 [10:01:31][C][safe_mode:018]: Safe Mode: [10:01:31][C][safe_mode:020]: Boot considered successful after 60 seconds [10:01:31][C][safe_mode:021]: Invoke after 10 boot attempts [10:01:31][C][safe_mode:023]: Remain in safe mode for 300 seconds [10:01:31][W][safe_mode:029]: Last reset occurred too quickly; safe mode will be invoked in 6 restarts

[syssi]

You have to attach another device to the serial line of the ESP to get a better idea what's going on. The important information's cannot be delivered via MQTT. The message from the safe_mode handler sounds like hard crashes. The explicit exceptions can be captured on the serial line only!