Open cannfoddr opened 4 years ago
Try a refresh node command for the affected nodes. ozw-admin has a button for this or you can publish a direct MQTT message for it
I also had a lot of sensors/nodes unavailable in HA. I am using docker. After a restart of my MQTT container, everything came back to life. Happens to me twice after a restart of all containers. Hope this info helps.
I forgot to mention that I have already tried refresh node - to be sure I have run it again on multiple powered nodes and still dont see the 'node data' - which from reading the docs I believe is actually the 'Node Payload'. I still only see this data for my node 15.
I am using owntracks-recorder image with mosquitto server built in. I left openzwave running and restarted the owntracks-recorder container. I can see all sorts of event messages being passed to MQTT by openzwave but still no 'Node Payloads'.
I just shutdown openzwave and deleted node 2 and restarted - no Payload for Node 2 when it got recreated. I then deleted the whole OpenZWave topic tree in the same fashion, restart openzwave and it gets recreated with only Node 15 having Payload every other node has none.
I looked at the cache file looking for what might be special about node 15 - node 2 looks exactly the same (eyeball) it has node id, status etc... just like node 15.
Not sure where to go next?
So far I have established that the problem seems to relate to the cache file and it re-occurs on a semi-regular basis. I just fired up Home Assistant UI and noticed all my Z-Wave entities are unavailable again.
Quick check on MQTT-Explorer shows that 'Node Payloads' are missing for all nodes except one. Everything was previously working fine.
I'm having the same, or at least very similar problem, all my nodes are unavailable in Home Assistant but seem to be working fine according to the OpenZWave QT GUI. Deleting the OpenZWave tree in MQTT Explorer and then restarting the OpenZWave added resolved the problem. I'm hoping for a better solution... What is the path to the cache file your are referring to?
I am running a docker container and I remap a local volume to /opt/ozw/config. Not sure how you are running your setup but the ozwcache.... file should be In there.
Use something like MQTT explorer to connect to your MQTT instance you should see something like this:
If the = { "NodeID": is missing you have the same problem as me, sometimes one or two nodes will still have this but the nodes that don't are always unavailable in HA.
Similar issue, happens quite frequently (multiple times in a day) but running the mqtt dump always solves it:
Might just do an automation for it, but it seems quite hokey to have to do this.
Sorry if this is a bit vague but I have lost all my HomeAssistant MQTT nodes but one. 14 are unavailable and 1 is showing OK. I am using MQTT Explorer to navigate the MQTT tree and I can see that the working node 15 has what I am going to refer to as "node data" ie. :![image](https://user-images.githubusercontent.com/1035066/90340745-50dd9500-dff2-11ea-9cfd-89f55d5f609a.png)
all the other nodes seem to be missing this data:![image](https://user-images.githubusercontent.com/1035066/90340775-7ec2d980-dff2-11ea-99eb-d9fc86fd14bd.png)
I am guessing this is why HomeAssistant is reporting these as unavailable.
I am seeing status updates for all the other nodes but cant see how I can force a refresh of the 'node data'. Not sure how/why this disappeared as it was all working previously when I rebuilt the entire network.
I am running docker:latest
{ "OpenZWave_Version": "1.6.1240", "OZWDaemon_Version": "0.1.170", "QTOpenZWave_Version": "1.2.0", "QT_Version": "5.12.9", "Status": "driverAllNodesQueried", "TimeStamp": 1597563650, "ManufacturerSpecificDBReady": true, "homeID": 11111111, "getControllerNodeId": 1, "getSUCNodeId": 1, "isPrimaryController": true, "isBridgeController": false, "hasExtendedTXStatistics": false, "getControllerLibraryVersion": "Z-Wave 3.95", "getControllerLibraryType": "Static Controller", "getControllerPath": "/dev/serial/by-id/usb-0658_0200-if00" }