I’m a new user of OpenHab, and use it to connect to my Nikobus.
Until now, I was testing, configuring it on my laptop.
Yesterday, I installed it on a Raspberri PI 2, and want to start using it.
Everything works fine, but after a couple of hours, the feedback from the nikobus stop working.
So if I press a switch in my home, Openhab doesn’t see it anymore.
Also the automatic refresh, in my case 300sec, doesn’t see ths status-changes anymore.
The commands from OpenHab are still working.
Is this a problem with 1.7.1? Is there a memory-leak?
This is a piece of the log file.
As you can see the connection is OK, and after 30minutes, I get an error ‘Error occurred during scheduled status refresh’
2015-09-07 17:07:33.711 [INFO ] [.o.b.n.i.core.NikobusInterface] - Connected to Nikobus :-) 2015-09-07 18:40:54.251 [WARN ] [o.u.i.items.ItemUIRegistryImpl] - Cannot find page for id 'demo'. 2015-09-07 19:16:47.525 [WARN ] [o.u.i.items.ItemUIRegistryImpl] - Cannot find page for id 'demo'. 2015-09-07 19:19:42.635 [WARN ] [osphere.cpr.DefaultBroadcaster] - Duplicate resource 2d30bf43-24e5-4cda-bc31-41d2f229e406. Could be caused by a dead connection not detected by your server. Replacing the old one with the fresh one 2015-09-07 20:08:54.289 [WARN ] [o.u.i.items.ItemUIRegistryImpl] - Cannot find page for id 'demo'. 2015-09-07 20:39:35.501 [ERROR] [.o.b.n.internal.NikobusBinding] - Error occurred during scheduled status refresh. java.util.concurrent.TimeoutException: No ACK received within timeout and retry count. at org.openhab.binding.nikobus.internal.core.NikobusAckMonitor.waitForAck(NikobusAckMonitor.java:92) ~[na:na] at org.openhab.binding.nikobus.internal.NikobusBinding.sendCommand(NikobusBinding.java:158) ~[na:na] at org.openhab.binding.nikobus.internal.NikobusBinding$2.run(NikobusBinding.java:295) ~[na:na] at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511) [na:1.8.0] at java.util.concurrent.FutureTask.runAndReset(FutureTask.java:308) [na:1.8.0] at java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.access$301(ScheduledThreadPoolExecutor.java:180) [na:1.8.0] at java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.run(ScheduledThreadPoolExecutor.java:294) [na:1.8.0] at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142) [na:1.8.0] at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617) [na:1.8.0] at java.lang.Thread.run(Thread.java:744) [na:1.8.0] 2015-09-07 21:40:35.465 [ERROR] [.o.b.n.internal.NikobusBinding] - Error occurred during scheduled status refresh. java.util.concurrent.TimeoutException: No ACK received within timeout and retry count. at org.openhab.binding.nikobus.internal.core.NikobusAckMonitor.waitForAck(NikobusAckMonitor.java:92) ~[na:na] at org.openhab.binding.nikobus.internal.NikobusBinding.sendCommand(NikobusBinding.java:158) ~[na:na] at org.openhab.binding.nikobus.internal.NikobusBinding$2.run(NikobusBinding.java:295) ~[na:na] at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511) [na:1.8.0] at java.util.concurrent.FutureTask.runAndReset(FutureTask.java:308) [na:1.8.0] at java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.access$301(ScheduledThreadPoolExecutor.java:180) [na:1.8.0] at java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.run(ScheduledThreadPoolExecutor.java:294) [na:1.8.0] at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142) [na:1.8.0] at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617) [na:1.8.0] at java.lang.Thread.run(Thread.java:744) [na:1.8.0] 2015-09-07 21:41:35.465 [ERROR] [.o.b.n.internal.NikobusBinding] - Error occurred during scheduled status refresh.
Does anyone else have the same problem?
I can’t use OpebHab because of this problem.
It’s not the connctions, because if I run the Nikobus programming software, the feedback keep on working…
Mark, any news on this. I’m not a Nikobus Openhab user yet but was planning to start installing in a few weeks. I have switch module, dimmer modules, lots of blind modules, audio module PC-logic and Link, 2 binary input modules (so overall a fairly big installation). So if there is a major issue I better have a look before getting into this.
No improvement yet. I already tried increase the auto-refresh time, but no difference.
I wll try to connect it to the PC-link module instead of the PC-logic module and will let you know the result.
Today I also asked the support team of Niko what the best way is to connect, and they let me know that you can’t use the link to a PC to control the Nikobussystem. It can only be used to program it…
I asked them thay maybe the feedback-module is an option, but the same answer…
Marc, don’t expect any help from Niko. They will do everything to prevent people from opening up Nikobus and Niko Home control. So they will tell you it doesn’t work or you shouldn’t connect. But the issue you’re experiencing is not related to the Nikobus connection or whichever module you use. PC-Link or PC-Logic should not make a difference. The connection is just a communication with your openhab, so the Nikobus is not cuasing the memory leak, it must be something in the Openhab or binding
Marc, mocht je toch twijfelen aan de Nikobus respons en gezien je toch een pclink en pclogic hebt probeer het volgende eens. Wanneer je PI2 gecrasht is probeer dan eens met de nikobus software te connecteren en via de software de lichten te schakelen (je ziet de modules rechts boven en je kan daarop klikken). Indien de nikobus reageert, wat i vermoed want anders zouden je schakelaars ook niet meer werken ben je 100% zeker dat het je Openhab/PI is die het probleem vormt.
Rohnny, het crashed zowel PI2 als onder windows. Nikobus crashed dan niet en ook de verbinding niet want ik kan nog steeds de commando’s via OpenHab sturen naar Nikobus. Enkel de terugmeldingen op Openhab komen niet meer binnen. Zowel de autorefresh (staat om de 600sec), als de refresh dat ik doe als ik een schakelaar indruk.
I tested it with version 1.3.1 with the same result.
Is the anyone where the feedback keeps on working?
Only the commands are stille working after a couple of hours.
I have tried it now also with the PC-link module, with same result…
Unfortunately I can’t use Openhab, because of the feedback that stops working after a while.
I also have issues with Nikobus feedback on rpi2 on 1.7.1
After some time the feedback of switches is not correct anymore.
Pushing a nikobus button does not reflect in openhab. all is ok after a reboot for some time
So, that’s the same problem that I have…
I first was thinking that’s it was the firmware of my PC-logic module or my PC-link module, because nobody had the same problem.
I already put my RPI2 just beside the module with a short cable, because normaly I have a cable of 10m to my router. But the result was the same.
But now you confirm that you have it also…
For me it also works again after restarting openhab.
Hello,
I’ve been reading this thread with much interest since I plan to purchase a Nikobus module especially for this.
Openhab + nikobus binding seems to be the only way to “open” my Nikobus installation to new non Niko technology.
Has anyone found a solution for this feedback issue?
Did a rollback to version 1.7.0 solved the problem?
I’m thinking of a dedicated RPI running openhab 1.7.0 + Nikobus binding conected to a main openhab server using MQTT.
I didn’t try a rollback to 1.7.0, but one to 1.3.0 , and the problem was also there…
So I suppose it is also in 1.7.0.
Just to be sure, I will try a rollback to 1.7.0 tomorrow.
