Tradfri binding loses connection after Power failure / gateway reboot

Hi,

I experience the problem as well: After rebooting OH, the Tradfri hub is offline. Only entering the credentials again makes it come online until the next reboot.

After an update from stable 2.3 in the last week, I am now on 2.4 M6 (Testing or Milestone), also for the Tradfri binding - I understood this should be fixed in my software, but I still experiencing the issue.

Any hint? Or do I need to delete and reinstall the binding? Or even apply a newer software (the one from the file offered above)?

Thanks.

Hi @Tobi77

Please try the file posted above. See my last post for details on how to install and report back with your findings.

Thanks
Chris

@chriscolden Dear Chris,

thanks. I will do so tonight.

Do I have to perform these steps in advance as well, or is it only copying?

I didn’t do this, but not sure what context that was mentioned in.

Update: I decided to use the snapshot version of OpenHAB: Seems to have no issues with the Tradfri gateway… :slight_smile:

I had to restart the binding in karaf after a snapshot upgrade yesterday, so no change for me.

Also tried a power OFF/ON just now.

Here is the log from first attempt at startup:

2018-11-28 22:48:32.215 [WARN ] [iscovery.TradfriDiscoveryParticipant] - Discovered Tradfri gateway doesn't have an IP address: [ServiceInfoImpl@765501869 name: 'gw-b072bf25aee1._coap._udp.local.' address: '(null):0' status: 'DNS: JmDNS-/192.168.1.7 [/192.168.1.7] state: probing 1 task: null', has NO data
        gw-b072bf25aee1._coap._udp.local.: ]

Here is log entry after

openhab> list | grep -i fri
217  Active     80  0.10.0.201811171951     Eclipse SmartHome TRÅDFRI Binding
openhab> bundle:restart 217

that made the Gateway and all items come Online again:

2018-11-28 22:50:20.074 [WARN ] [iscovery.TradfriDiscoveryParticipant] - Discovered Tradfri gateway doesn't have an IP address: [ServiceInfoImpl@309596738 name: 'gw-b072bf25aee1._coap._udp.local.' address: '(null):5684' status: 'NO DNS state: probing 1 task: null', has NO data, empty]

Same in snapshot #1447:

openhab> list | grep -i fri
217  Active     80  0.10.0.201811281422     Eclipse SmartHome TRÅDFRI Binding
openhab> bundle:restart 217

As the snaptshot builds ahave issues with the Z-Wave binding, I had to switch back to milestone builds… :frowning:

When will the latest fixes arrive in a milestone build? I am on snaptshot now which is a bit unreliable for several bindings…
Thanks

@chriscolden: Dear Chris, maybe you can give a comment when this should be fixed in milestone builds?

I dont develop for openhab so I cannot say. I’m using snapshot builds and haven’t had the issue for a while which is odd because to my knowledge nothing has changed in the code.

After upgrade to OH 2.4 stable, same issue like before - the security code for the Ikea Tradfri bridge is empty after a reboot and can’t connect from OH. From my point of view it makes no sense to have a stable smarthome and after a reboot - for example lost power - credentials or codes must be entered manually. I hope someone have a idea how to solve this. Is it possible to create a script what saved the code after a reboot automatically ?

Same issue here. After reboot connection to Gateway lost.

I only lose connection, no settings.
It’s been like this for more than a year on Snapshots.
On my remote site I had to install a Z-Wave wall plug in order to be sure to get it online after a power outage.
More recently I have had good success reconnecting restarting the binding from Karaf.

Both my problem sites are running OH2 Snapshots on x64 Ubuntu Server 16.04 LTS.

Even in OH 2.4, latest milestone (is currently release), the binding loses connection to the gateway after a OH reboot. Maybe the ESH updates to the binding need to be included in OH?

1 Like

@Tobi77, what kind of system do you run OH2 on?
Maybe this is only a problem on certain platforms?
Will try to enable DEBUG/TRACE level debug for binding.
Only about 2 in 10 restarts goes without a hiccup.

Raspberry Pi 3B Plus, milestone…

I had the impression that snapshot builds are more stable regarding the Tradfri binding, but much more unstable for others…

I’ve been on snapshots from day one and the problem has been there since last winter.
I see the same behaviour on two sites, both running Ubuntu server on intel x64 CPU.

Hmmm. Do we have know the maintainers of the binding? Or is this an ESH business?

Note, the ESH bug report claims this to be resolved since ESH 0.10, which is already running in OH 2.4.

Still, the issue remains…

I created a bug report here:

I hope this is the expected procedure. Please inform me in case a acting wrong.

2.5.0-SNAPSHOT Build #1482

My Trådfri Gateway lost all its pairings yesterday, possibly because I had to re-boot/power-cycle it in order to get it to connect to OH2. I had problems re-pairing lights and opted to replace it with a spare I had.

This new Gateway has the same issue with failing to connect after an OH2 restart.
I tried enabling more logs, but no more is showing.

openhab> log:set DEBUG org.openhab.binding.tradfri

When restarting the binding, this is the only WARN log:

2018-12-30 15:31:44.086 [WARN ] [iscovery.TradfriDiscoveryParticipant] - Discovered Tradfri gateway doesn't have an IP address: [ServiceInfoImpl@1501157832 name: 'gw-b072bfb31baf._coap._udp.local.' address: '(null):5684' status: 'NO DNS state: probing 1 task: null', has NO data, empty]

Same log entry on both success and failure.