Zigbee (Telegesis) coordinator won't detect my devices (Trådfri bulbs)

I won’t argue with you on that, so let’s just try it again!

  1. I stopped my OH container (Docker)
  2. Deleted the openhab.log
  3. Reset the bulb to factory reset by turning on/off six times, get a blink and then turn it off.
  4. Started the OH container again
  5. Started a new capture in wireshark (this was done during starting up of OH, so not sure what comes from that in the beginning of the session)
  6. As soon as OH is up and running i start a new search for devices in inbox
  7. Turn the bulb on
  8. Wait for the search for things to time out
  9. Stop the OH container and Wireshark sessions.

Results:
openhab.log
Wireshark session

Regards
Henrik

I’m not really sure what is showing in the log - it contains next to nothing. The coordinator is a TI coordinator, and it doesn’t tie up with the Telegesis MAC address. It also doesn’t pass the security, so it seems the network key you sent also doesn’t tie up with, and the PAN ID is also different. There are no join messages - or in fact any messages other than the occasional data request and a few broadcasts.

My guess is that this log isn’t from your network - ie the network and the openhab log don’t tie up at all.

That’s really weird! I’m 100% sure the log file was created during the capture of the wireshark session, so what could go wrong here? (I don’t have the computer in front of me right now) Can it be that I’m listening on the wrong channel? Is there anything I can test to verify my setup?

Yes, this could be possible - there are 16 channels so you might be listening to some unconnected bulbs or some other zigbee network.

Ok, I double-checked that I had no zigbee bulbes connected and unplugged the telegesis stick and still got the same data rolling in on channel 11 (it buggs me where that comes from :thinking:, but I suppose we can ignore that for now and carry on…:slightly_smiling_face:). Changing to channel 12 in wireshark removes all incoming data.

I plugged the telegesis back in and fired up OH. After that I changed to channel 12 in the configuration of the telegesis thing and saved the configuration. It took a few seconds for the thing to come back online (showing status unknown during the whait), but when back online the channel had reverted back to 11 (see log file when changing to channel 14 here). I tried again, but this time it went to status Offline - COMMUNICATION_ERROR. When going back in to edit the configuration it is back on channel 11.

Can you see what is going on in that logfile?

Regards
Henrik

It looks like the channel is being set to “automatic”, so the device is selecting channel 11. I’m not sure why you want to change that? Channel 11 should be fine - you just need to ensure that the coordinator and the sniffer are on the same channel, and then reset the bulbs so that they attempt to join the network.

1 Like

Well, I would like to change it because there is some other unit transmitting on channel 11 (as I wrote there is communication on that channel when the telegesis stick and all bulbs are disconnected. I did not set it to automatic, but 14 in the log file and I can see that it indeed tries to set it to 14 in the beginning, but for some reason it reverts back to 11 (or automatic?) without my intervention.

That doesn’t matter - you can run multiple networks on the same channel.

I know, but it makes it harder to read the wireshark logs… And it does look weird that I can’t change channel, maybe my main problem is related?

Not really :confused: . Just add a filter for the PAN.

No - you are trying to join devices to a coordinator - this is independent of channels as the device should search all channels. Using channel 11 is always better in any case as it’s the default channel for all devices.

I agree that this should be looked at, but it’s not an issue that is related to your problem, and I suggest we focus on the initial problem. You don’t need to change channel - just use channel 11.

Ok. Any suggestions for what to try now? From what I can tell there is nothing sent from the telegesis what so ever?

Based on what I see in the sniffer log, this appears to be the case.

Maybe I can try to send single line AT commands to it from within the docker container? What would be good to send in order to verify it?

I’m not really sure why sending the commands individually would make any difference, but of course you can try.

Well, if something isn’t behaving as expected it’s easier to see when you break it down. Do you have a list of the commands that needs to be sent to do a full initiation and joining sequence?

But there are no errors in the log from what I could see? Or did I miss something?

Please refer to the log.

No errors, but no data on the air either? Shouldn’t there be data transmitted as well?

If you are enabling join, then yes, as we have discussed in the past there should be data in the sniffer logs.

And I think we can agree that there isn’t, so I’m trying to figure out what to do next.

Hi Chris,
Finally, after a very long vacation and a new job etc. could I finally get myself some time to look into this again. And I finally got it to work! The solution was to unplug it from the NAS, attach it to my laptop and follow chapter 3.1 in this document found on Silicon Labs webpage. During the step “ATS0A0=0;password” I enabled my first bulb and could directly see it pop up and after a few more seconds see it join. After the complete sequence I reinserted it in my NAS and booted up OpenHAB. I could imediately find the bulb and after that I could find all other bulbs I had lying around.
Conclusion (feel free to correct me if you think I’m wrong): I believe some setting was wrong in the Telegesis stick factory default and it was not correctly initiated by the binding.

I hope this description can help someone and in any case, thank you @chris for your support in this topic and for your work on the binding!