Ezlo Plus hub went offline, orange blinking light, does not react

Since about 12 hours, my controller is offline and no longer responds to anything. Please help…

If I power cycle the hub (SN 90000437) it seems to boot ok. After various blinking lights, it ends with a solid green, at which moment I can ping the hub and the hub seems to show online in the vera app. However, after about 60 seconds, the hub transitions to a blinking orange led (rapid blinking / 2Hz). After this moment, the hub no longer responds to anything. Local meshbots do not run and the controller shows offline in the app.

The hub seems to be running firmware 2.0.32.2097.3, as can be witnessed from mdns traffic during startup of the hub.

Strangely, a continues ping still shows responses originating from the hub. However, starting a new ping once the orange light blinks, does result in ping timeouts.

I lost control of various lights and shades throughout the house, as well as the doorbell functionality. So it’s quite a nuisance at the moment.

Any help would be much appreciated!

Only enough, if I time it correctly, I can initiate a ssh session to the hub. That session is not terminated. This is what I just found in the ha-network.log:

2022-11-02 19:17:19.891272 INFO : Tunnels monitoring…

2022-11-02 19:17:19.891620 INFO : Tunnels monitoring…

2022-11-02 19:17:42.065032 INFO : Upload logs to cloud: /home/logs/postponed/90000437_ver-2.0.32.2097.3_date-2022.11.02_17.17_rotated_logs.tgz

2022-11-02 19:17:47.382217 ERROR: R: 6 - Error cloudStorage:6362b446129e0f1c3e5bf510, http code: 0

2022-11-02 19:17:47.382499 ERROR: Connection error: cloudStorage:6362b446129e0f1c3e5bf510: ‘Couldn’t resolve host ‘upload-cloud.ezlo.com’’

2022-11-02 19:17:47.382719 ERROR: emitConnectionClosed: event.http_connection_closed.httpStatus not set, some other Http status: 1

2022-11-02 19:17:47.383155 ERROR: clearQueue: No connection to remove found in a plugin queue: (cloudStorage, 6362b446129e0f1c3e5bf510)

2022-11-02 19:17:47.395903 ERROR: Exception (N11HubPlatform6LogCmd14LogUploadErrorE): File was not uploaded: {“error”:“Cannot resolve DNS name”,“fileStorageInfo”:{}}

2022-11-02 19:17:47.398511 WARN : Failed to send postponed logs

root@HUB90000437:/tmp/log/firmware#

Hello @jouked ,

Your controller’s uptime is 45 min, which means it’s reporting to the servers for at least that time.
Is the controller still experiencing the same behavior?

Well, through ssh I was able to get into the hub. I witnessed connectivity problems to my local router, even though I have internet connectivity. Ping to 192.168.100.1 (local router) fails. At the same time, pinging 8.8.8.8 (google dns) works. Weird, but hey ho.

I manually edited /etc/resolv.conf to replace the the local router as a name server with google’s dns server, as the logfiles clearly showed dns errors.

This got the controller back online, though only partially. The fact that you can see the uptime of 45m is a result of changing /etc/resolv.conf.

At the same time, z-wave functionality is still unavailable:

root@HUB90000437:/tmp/log/firmware# date
Wed Nov 2 20:22:12 CET 2022
root@HUB90000437:/tmp/log/firmware# tail ha-zwaved.log
**** zwaved Restarted at Wed Nov 2 20:22:10 CET 2022
**** zwaved Restarted at Wed Nov 2 20:22:12 CET 2022
**** zwaved Restarted at Wed Nov 2 20:22:14 CET 2022

root@HUB90000437:/tmp/log/firmware#

It looks like the lan port and wifi are both active, this might cause these weird connectivity issues. Was there any change is networking setup with the latest firmware release for the Ezlo hub?

I checked with the vera app, where indeed Wifi and LAN were enabled. I disabled wifi and just rebooted the hub, lets see if this helps.

This seems to be an (unusual) firmware issue. We will need to access your controller remotely to troubleshoot it.
I sent you a PM asking for further info.

Not much luck. Disabling wifi and rebooting the hub leads to a stable ‘up’ state of the hub, with a solid green light. Networking is stable, as is the cloud connection.

However, z-wave, zigbee etc are not functioning, giving constant restart messages in the logs.

I sent an OTA update to the controller. Please power cycle the controller and let me know the results.

Problem has been solved by Ezlo staff. That was some amazing quick customer service, thanks for that!

1 Like