From time to time (on a daily basis, sometimes multiple times a day) I see entities from supervisor being unavailable for a few minutes (fine after 3 to 5 minutes).
These incidents do not correspond to
- overall system usage (no load issue or sth. like that)
- any time pattern according to the
statestable in the recorder - Supervisor log entries (last incident: 17:35/17:37)
23-02-04 17:24:02 WARNING (MainThread) [supervisor.addons.options] Option 'interface' does not exist in the schema for Samba share (core_samba)
23-02-04 17:29:06 WARNING (MainThread) [supervisor.addons.options] Option 'interface' does not exist in the schema for Samba share (core_samba)
23-02-04 17:30:49 INFO (MainThread) [supervisor.homeassistant.api] Updated Home Assistant API token
23-02-04 17:34:20 WARNING (MainThread) [supervisor.addons.options] Option 'interface' does not exist in the schema for Samba share (core_samba)
23-02-04 17:39:24 WARNING (MainThread) [supervisor.addons.options] Option 'interface' does not exist in the schema for Samba share (core_samba)
Example for one random sensor:
SELECT *
FROM "states"
WHERE entity_id="sensor.esphome_cpu_percent"
AND state="unavailable"
AND last_updated LIKE "2023-02%"
COUNT (*)
7
I use Unavailable / Unknown Entity Monitoring - Template Sensor to monitor all unavailable entities and the Supervisor is the only integration regularly making problems - for no obvious reason. As I configured a ignore time of 2 minutes, this issue is likely much more often a thing than my entity monitoring is catching.
How can I get some more insights on the reason for this?
Any known Supervisor issues related to this?
(running supervisor-2023.01.1)