Intermittent random devices going to unrecoverable device offline since latest release

[snigehere]

Since installing 3.40 I have notices devices going unavailable randomly.

I have two sonoff devices connected via eWeLink (I have one switch and one temp/humidity sensor) and one or both go to unavailable at random times. The only way to recover the devices is to reload the integration.

The debug log just shows the device as not responding and is then marked off line. There is no detailed response and there is no retry in the log (failing device in this case is 10008e663d).

2023-02-28 21:30:44 [D] 10008e663d => Local4 | {'cmd': 'info'} <= info: 200
2023-02-28 21:30:59 [D] 100149a892 => Local4 | {'cmd': 'info'} <= info: 200
2023-02-28 21:31:59 [D] 100149a892 => Local4 | {'cmd': 'info'} <= info: 200
2023-02-28 21:32:02 [D] 10008e663d => Local4 | {'cmd': 'info'} !! Can't connect: Cannot connect to host 192.168.0.213:8081 ssl:default [Connect call failed ('192.168.0.213', 8081)]
2023-02-28 21:32:02 [D] 10008e663d !! Local4 | Device offline
2023-02-28 21:33:14 [D] 100149a892 => Local4 | {'cmd': 'info'} <= info: 200
2023-02-28 21:34:14 [D] 100149a892 => Local4 | {'cmd': 'info'} <= info: 200
2023-02-28 21:35:29 [D] 100149a892 => Local4 | {'cmd': 'info'} <= info: 200

My router also feeds home assistant with device status and the device tracker not see any issue - eg it shows the device as "home" eg online

In the eWeLink app the device still shows as available and are controllable.

Checking my network monitor for the outage time does not expose any issues

This issue did not happen in the prior release and so I guess some part has been refactored that is now causing this issue.

[snigehere]

Just to add.... I can bring the device back online in home assistant by going into eWeLink and turning the device on/off

The debug log then shows (failing device in this case is 10008e663d)

2023-02-28 23:02:30 [D] 100149a892 => Local4 | {'cmd': 'info'} <= info: 200
2023-02-28 23:03:30 [D] 100149a892 => Local4 | {'cmd': 'info'} <= info: 200
2023-02-28 23:04:45 [D] 100149a892 => Local4 | {'cmd': 'info'} <= info: 200
2023-02-28 23:06:00 [D] 100149a892 => Local4 | {'cmd': 'info'} <= info: 200
2023-02-28 23:07:15 [D] 100149a892 => Local4 | {'cmd': 'info'} <= info: 200
2023-02-28 23:08:08 [D] 10008e663d <= Local3 | {'switch': 'on', 'startup': 'off', 'pulse': 'off', 'sledOnline': 'off', 'pulseWidth': 500, 'rssi': -67} | 1167
2023-02-28 23:08:09 [D] 10008e663d <= Local3 | {'switch': 'off', 'startup': 'off', 'pulse': 'off', 'sledOnline': 'off', 'pulseWidth': 500, 'rssi': -67} | 1168
2023-02-28 23:08:30 [D] 100149a892 => Local4 | {'cmd': 'info'} <= info: 200
2023-02-28 23:09:15 [D] 10008e663d => Local4 | {'cmd': 'info'} <= info: 200
2023-02-28 23:09:30 [D] 100149a892 => Local4 | {'cmd': 'info'} <= info: 200
2023-02-28 23:10:30 [D] 100149a892 => Local4 | {'cmd': 'info'} <= info: 200
2023-02-28 23:10:30 [D] 10008e663d => Local4 | {'cmd': 'info'} <= info: 200

[andrep2208]

not sure if this helps, i got the same issue, so i replace the sonoff folder in custom component with the previous release and i havent experienced the issue since

[Miilani]

not sure if this helps, i got the same issue, so i replace the sonoff folder in custom component with the previous release and i havent experienced the issue since

it worked for me

[snigehere]

Thank you @andrep2208 and @Miilani - yes, I have already downgraded but wanted to provide the debug info as a bug

[gabriele-latino]

same issue for me

[thefunkygibbon]

hey all, a great workaround since this issue doesn't appear to be being worked on, is to create an automation to monitor a sonoff device and restart the service automatically. its not a perfect solution but hopefully it should make the devices work for "most" of the time.
#1138 (comment)

[david-bann]

I have the same problem - downgraded to 3.3.1 and it works again.

[AlexxIT]

This is because your devices don't response on requests randomly. And I might suggest that you use mode: local.
That's how you created this problem for yourself.
In version 3.5.0 this situation is slightly improved.
And you should not use mode: local if you want everything to work well and stable.

[thefunkygibbon]

Mine is already set to 'local'

[snigehere]

@AlexxIT
This is a bug report - why did you tag it as question?

Why did you close this please? The issue still exists - "Local" is an offered configuration so why is it not working correctly ?? If local is not valid then why is it offered? is there not some possibility to retry if the device does not respond?

You seem very angry that multiple people are reporting issues ... surely that will tell you there is a problem

[AlexxIT]

I have closed this because new release, you can read release notes:
https://github.com/AlexxIT/SonoffLAN/releases/tag/v3.5.0

[snigehere]

@AlexxIT thanks for the info and update. All I saw was this tagged as question and closed saying not to use local hence my comment. I have now installed 3.5