It looks like after a reboot of OH it takes a long time for the z-wave network to get meshed again. Can I force this?
Status two days after last reboot:
Syncronize network does not fix it.
maybe this will help:
Meesage after selecting syncronize network: RequestNetworkUpdate not placed on stack.
I’m not sure what you mean by this, but this command is in no way related to the mesh - it’s a command to synchronise controllers if you have multiple controllers in your system. It’s also not something that is sent during the initialisation if I remember correctly, so I don’t know why this is a problem?
Hi chris,
thank for the answer.
But how to fix my problem. After a reboot it looks like it takes very long (more the days) for the z-wave network to reorganice itself. Or is the graph not a good point ot check this?
What is my real problem?
Some of my devices are no longer reporting messages (open/close/temeprature/…) to OH. I have to wake them up manually to get messages again. I thought it is a problem of the z-wave network that the devices do not know to whom to communicate to reach the controller? And the graph shows this, a lot of stand alone devices with no connection to others.
Or do I have a different problem?
I’m not sure what the problem is? Is the problem just that the graph looks wrong, or is there actually a problem?
No - the graph is not a good indicator. For starters, in the master branch it is not updated after restart. Additionally, the graph doesn’t mean that there’s no routing since the controller is responsible for this, and especially with ZWave+ devices, there are other routing mechanisms in place to ensure delivery.
Hi, sorry for writing in this post, but I have exactly the same problem. The Z-Wave Network-Viewer shows exactly the same. This happend after a reboot. Before the reboot the whole system works correct. Now, when the Network-Viewer looks like that (see picture in first post from “homeAutomation”), I get no more ON/OFF messages from a few devices.
So - how can I force an mesh update ?
When I start “heal the device” - I get “Node xy - mesh update started” and than “mesh update finished”. But only from few devices. From the others I get no response.
I use OpenHab 2.3 release, with the Chris´s zWave 2.2.0 binding. Any advice for me ? 
#sorryBadEnglish#
Here are some more “infos”. When I restart OpenHAB (sudo reboot), and watch the log, I get some warnings.
But I´m to new in OpenHAB to understand what causes this warnings. I hope someone can give me a hint :
2018-03-27 20:56:39.343 [INFO ] [.dashboard.internal.DashboardService] - Started dashboard at http://192.168.0.125:8080
2018-03-27 20:57:10.652 [WARN ] [el.core.internal.ModelRepositoryImpl] - Configuration model ‘home.things’ is either empty or cannot be parsed correctly!
2018-03-27 20:58:34.300 [ERROR] [lmessage.RequestNodeInfoMessageClass] - Request node info not placed on stack due to error.
2018-03-27 21:00:35.179 [ERROR] [lmessage.RequestNodeInfoMessageClass] - Request node info not placed on stack due to error.
2018-03-27 21:01:14.708 [ERROR] [lmessage.RequestNodeInfoMessageClass] - Request node info not placed on stack due to error.
2018-03-27 21:01:14.767 [ERROR] [lmessage.RequestNodeInfoMessageClass] - Request node info not placed on stack due to error.
This are only the errors in the logs.
Firstly, a debug log is best for debugging 
However, I don’t think the ZWave errors will cause any problem as there should be a retry.
thanks @chris - , I will reboot openhab and change a parameter of a device that does not send me any status updates anymore. Than I post the debug log after that.
EDIT:
newbie question(!) - I set the debug in the karaf: log:set DEBUG opr.openhab.binding.zwave
But I don´t get any Debug messages in the openHAB log viewer … ?
This should work ok - assuming that opr is really org when you type the command?
Haha … that was the fault 
… now it works.
I´ll post the debug this afternoon, because I currently only have a remote connection to my smartphone (I´m at work). Copy&paste is just more than awkward at the moment
@chris , so, now heres the debug logfile. I did a reboot at 10:38:
https://drive.google.com/open?id=1uROhIi7k9naTXiXp1lZBSf6Z2D55o-8b
and this log is from changing the association groupe “lifeline” from a device from which I receive no ON/OFF status anymore…
https://drive.google.com/open?id=12pd5jqkHAFMrRmOD4j1cPRVDKZXppB1m
the device´s last heat time is days old … ???
when I try to heal the device:
2018-03-28 15:24:43.610 [DEBUG] [ding.zwave.handler.ZWaveThingHandler] - NODE 11: Configuration update received
2018-03-28 15:24:43.623 [DEBUG] [ding.zwave.handler.ZWaveThingHandler] - NODE 11: Configuration update action_heal to true
2018-03-28 15:24:43.626 [DEBUG] [ding.zwave.handler.ZWaveThingHandler] - NODE 11: Starting heal on node!
2018-03-28 15:24:43.629 [DEBUG] [ng.zwave.internal.protocol.ZWaveNode] - NODE 11: Can not start heal as initialisation is not complete (RETURN_ROUTES).
something strange here …
Can you describe the problem that I’m looking for? I see a reboot at 10:38 and it looks like there’s an error, but the binding is not in debug mode at this time, so there’s no information that I can use. As I said above, this error shouldn’t really be an error - and the request will be retried, so this won’t in itself cause any problem.
There is then another restart at around 1105 where debug is enabled and everything looks fine as far as I can see. There’s a lot in the log so I can’t look at every line - is there something I should be looking for?
The exception is fixed in the development binding.
Heal is not really supported in the current binding - only in the development binding.
Can you describe the problem that I’m looking for? I see a reboot at 10:38 and it looks like there’s an error, but the binding is not in debug mode at this time, so there’s no information that I can use. As I said above, this error shouldn’t really be an error - and the request will be retried, so this won’t in itself cause any problem.
There is then another restart at around 1105 where debug is enabled and everything looks fine as far as I can see. There’s a lot in the log so I can’t look at every line - is there something I should be looking for?
The exception is fixed in the development binding.
Heal is not really supported in the current binding - only in the development binding.
hi, I use this binding: org.openhab.binding.zwave-2.2.0-SNAPSHOT.jar
the problem I have is: node11 (fibaro single switch 2) does not send ON or OFF to openhab, when switched ON or OFF with the wallswitch. HABPanel don´t recognize it, and there is no messasge “…switched from ON to OFF…” in the Log. This is since the z wave network monitor looks like the picture here in the first post . Thats the reason why I thought there must be a problem…
This is not the development version - this is a very old snapshot. I’m not really sure where you got this as 2.2 is released, and 2.3 snapshot is the latest since a few months now.
However, this is not relevant since the development version is here -:
oh - okay … than I will give the 2.3 (17th March ´18) a try
I don´t want to waste your time with my questions to such an very old snapshot. So - sorry bro.
I did not know about the newer one… i will try. thx a lot