Restart on failure

[tygerlord]

We face a problem with our zigate module, devices becomes Offline after a moment (from few hours to a few days) and we don't exactly know where does it come from. See here please

It seems that after a failure there are a retry and if this fail again nothing.

2021-06-13 21:04:43 WARNING (MainThread) [zigpy_zigate.api] No response to command 0x0530                                                               ┤
│2021-06-13 21:04:43 WARNING (MainThread) [zigpy_zigate.api] Retry command 0x0530                                                                        ┤
│2021-06-13 21:04:43 DEBUG (MainThread) [zigpy_zigate.uart] Send: 0x0530 b'02a6e10101000001040000050057000400'                                           ┤
│2021-06-13 21:04:43 DEBUG (MainThread) [zigpy_zigate.uart] Frame to send: b'\x050\x00\x112\x02\xa6\xe1\x01\x01\x00\x00\x01\x04\x00\x00\x05\x00W\x00\x04\┤
│2021-06-13 21:04:43 DEBUG (MainThread) [zigpy_zigate.uart] Frame escaped: b'\x02\x150\x02\x10\x112\x02\x12\xa6\xe1\x02\x11\x02\x11\x02\x10\x02\x10\x02\x┤
│2021-06-13 21:04:44 WARNING (MainThread) [zigpy_zigate.api] No response to command 0x0530                                                               ┤
│2021-06-13 21:04:44 DEBUG (MainThread) [zigpy.device] [0xa6e1] Delivery error for seq # 0x57, on endpoint id 1 cluster 0x0000: ZiGate doesn't answer to command│

Elsewhere, after a retry if the trouble still exist, the plugging is restarted, it seems not be the case in HA.

Does it here where we need to think about restarting in case of failure, or someone can point me where I should ask for this ?

Thank

[Adminiuga]

This is more of the zigoy-zigate issue. It should reconnect/re-initialize on the error conditions.

[tygerlord]

Ok I will have a look