Closed njbuch closed 5 years ago
What do you mean by "loses connection". The logs show that no client connected to the ESP within 5 minutes, so it reboots itself.
@OttoWinter this might be worth a separate issue, but for folks that want to use the API without a persistent connection from Home Assistant, would it be possible to set up an alternative heartbeat (ping to default gateway, etc) so that the node isn't rebooting all the time? It'd be nice if I could take my Home Assistant containers down for maintenance without all my lights suddenly flickering every 5 minutes. I don't mind it but the wife and kids get irritated ;)
@brandond You can change the reboot timeout, or even disable it. Most of the time, it doesn't cause issues, but I've somtimes seen the ESP think it is connected to WiFi when it was really not.
Ping is also not that simple because we don't have easy access to raw sockets for ICMP pings.
Yeah, the health check is worth having, especially on esp32 - it'd just be nice if I could use something other than the control channel for it. I took a look at the arduino ICMP stuff and yeah - looks like they implemented an ICMP responder so you can ping it, but there's no way to send a ping request from the device or get a callback on a response, which is a bummer.
Hi,
I am experiencing a very similar issue.
Here are the details:
ESPHome Version: 1.11.2 Hardware: ESP32-WROOM-32 HA platform: Hass.io on RPI3b+ HA version: 0.88.1
Description:
My ESP32 modules stay "connected" in HA for days without issues until I add the "esp32_ble_tracker:" into the configuration. At which point, it becomes unreachable (for 8 to 10 minutes) many times per days.
Did many tests:
If i can be of any help to assist on this issue, let me know.
Currently trying to migrate all my Xiaomi Mijia from RPI to ESP32 using ESPHome, but need to resolve this issue first.
Really love this application and greatly appreciate the time and effort this require.
Thanks
@MichelLap please open a new issue and fill out the issue template.
Thanks,
Issue seams to have been completely fixed by latest 1.12.0 release. Did not get any disconnect in the past 48 hours since the upgrade.
Thanks for the great work.
Amazing! I will close the issue then. New issues of this type should be reported as new issues in this tracker.
Operating environment/Installation (Hass.io/Docker/pip/etc.):
Windows, platformIO, newest components as of today. ESP (ESP32/ESP8266/Board/Sonoff):
ESP32 Affected component:
api
Description of problem: It seems that the board looses connection for unknown reasons and reboots.
Problem-relevant YAML-configuration entries:
Traceback (if applicable):
Additional information: