Hey everyone,
sorry for the wall of text.
I acquired a Elelabs ZigBee USB stick (uses a EM358) a while ago and have been using it for a few weeks without any big issues.
However, after I was done with testing I added the rest of my ZigBee devices and suddenly after a restart all of my ZigBee nodes are offline and won’t come back. I even reverted to an older config and yet the nodes are still not coming back.
So I reset the controller, removed all nodes and added them again just for it to happen again yesterday, all nodes were freshly added and are now dead again.
Since it ran for weeks without issues I assume one of the new devices messed up the network but I’m not sure which and I’m not sure if I can recover from this without having to add all nodes every time this happens. So before readding all nodes again I thought I would ask here.
I’m was using the Docker image 2.3.0-amd64-debian when it happened. When starting with this image the controller will come online but only 1-4 random nodes will come online with it, none of these are controllable however.
I also tried 2.4.0-snapshot-amd64-debian which handles things slightly differently. Now OpenHAB is stuck in a endless discovery loop at startup and the ZigBee log just keeps flowing forever. After like 10 minutes it will have discovered all nodes one by one (even ones that aren’t even powered) but none of them will be controllable.
Adding a fresh device will work and the controller can talk to it just fine. Until this happens again that is.
Actually, I can’t add a new node until I cleanup manually.
I’m seriously confused as to why this keeps happening and since the config rollback doesn’t help I assume something at the network level gets messed up with one of the devices I added? Does the Ember stick keep its own config like the Z-wave sticks do or has the controller simply removed all nodes from the network (for whatever reason)?
I took a debug log with 2.3.0 on startup after the issue occurred and I can post it (assuming it contains no sensitive info?). Not sure if it’s of any help though since the network is already messed up at this point.
Just for reference, these are the devices I added a day before the network died:
- 1x Innr RB 185 C (RGBW E27 bulb)
- 5x Philips LCT003 (RGBW GU10 spot)
- 2x Philips LTW012 (W E14 bulb)
- 1x Philips LST002 (RGBW LED strip)
- 1x Philips RWL021 (Dimmer switch)
- 1x Philips SML001 (Motion sensor)
That makes for a total of ~24 devices on this network (including the existing).
Everything kept working for a full day, at night I noticed that the motion sensor stopped reporting its luminance, the next morning the network was unresponsive, after a restart it was dead. Controller comes up with a few random nodes, nothing responds.
I’m assuming the issue lies with one of the battery powered ones or the LED strip since I added plenty of bulbs and spots before that and they all worked fine. Still, I hope this issue can be diagnosed before continuing, it sucks that one device can take your whole network down like this.
So, anybody experienced this before? Any idea how to get the network back without pairing everything again? Any idea what could cause this?