search

zigpy / bellows

A Python 3 project to implement EZSP for EmberZNet devices
GNU General Public License v3.0
177 stars 86 forks source link

[bellows.ezsp] NCP entered failed state. Requesting APP controller restart #592

Closed Danieldiazi closed 7 months ago

Danieldiazi commented 7 months ago

Hi

For the past several days, I have a lot of messages " [bellows.ezsp] NCP entered failed state. Requesting APP controller restart" and also "ERROR (MainThread) [bellows.uart] Lost serial connection: ConnectionResetError('Remote server closed connection')"

In my case, I have a Ethernet-connected gateway. Moreover, If the problem is with that device, I changed it from another new one, same model and yet I still receive those messages.

Any solution or idea about this problem?

puddly commented 7 months ago

What specific gateway? Can enable ZHA debug logging and post a log of this happening?

Danieldiazi commented 7 months ago

What specific gateway? Can enable ZHA debug logging and post a log of this happening?

Hi!

I've Silvercrest Zigbee Lidl (I followed this guide in the past: https://paulbanks.org/projects/lidl-zigbee/)

Thanks for your answer

I will try to get log file

Danieldiazi commented 7 months ago

What specific gateway? Can enable ZHA debug logging and post a log of this happening? And the debug log file: 2023-11-14 095034.695 DEBUG - NCP.txt

puddly commented 7 months ago

Are you 100% sure nothing else (addons, etc.) is trying to connect to the gateway? Any other connection attempts will disconnect the active connection.

Danieldiazi commented 7 months ago

Are you 100% sure nothing else (addons, etc.) is trying to connect to the gateway? Any other connection attempts will disconnect the active connection.

Hi, I have container version of HA, and I've ZHA integration. Maybe is a check I have into "Uptime Kuma" ¿? With "uptime kuma" I check the TCP connection on 8888 port in order to get notified If gateway is down. I'll disable it.

My check is running each 5 minutes and I've this message error each 5 minutes. I may have shot myself in the foot...

puddly commented 7 months ago

Disable the check and see what happens. The error says that the TCP server is disconnecting ZHA, which is consistent with what you describe.

Danieldiazi commented 7 months ago

Disable the check and see what happens. The error says that the TCP server is disconnecting ZHA, which is consistent with what you describe.

I changed this TCP check to a PING check, I'll wait some time and report the result. Thanks and so sorry for my mistake.

Danieldiazi commented 7 months ago

14 minutes later and I don't get any new message about that problem. Yes, I think this check was the problem. Sorry.Thanks for you great support.

MattWestb commented 7 months ago

Standard setting for the serial GW is only allowing one connection so opening the port with other program is disconnecting the first ones. Its possible overriding with some version in the config but better not having any other program making strange things with the Zigbee module.

Danieldiazi commented 7 months ago

Standard setting for the serial GW is only allowing one connection so opening the port with other program is disconnecting the first ones. Its possible overriding with some version in the config but better not having any other program making strange things with the Zigbee module.

Yes, It was my mistake, When I've created the check using"Uptime Kuma" I didn't realize it would cause this problem. I replaced this TCP check and now I have a Ping check. TCP was more specific, but no problem, The important is that GW and Zigbee network work fine!

Thanks

Danieldiazi commented 7 months ago

After two days, the message has disappeared, confirming that the problem has been solved. Thank you for your great support and again, sorry for my mistake!. I close the issue.