I’ll remove them and let you know in a couple of days how it goes.
Last night this started at 06:05 and continued until restart at 08:15. At that time 06:05 I did not access any UI. Previous time was three days ago when it started at 07:55 There are no rules or anything that fire at these times.
Normally I access the Classic UI locally from my iPhone. I also access Basic UI via the Cloud connector and Chrome web browser. I do not get any error messages when doing this but it happens spontaneously.
There are some 50 Items in my configuration. All visible to the cloud connector.
Are all of them exposed?
Note: You don’t need to expose ANY Item in the Cloud connector configuration if you simply want remote access and notifications.
Exposing Items is for integration purposes to 3rd party applications like IFTTT, Alexa, Google Assistant, etc.
Exposing Items increases the load both on your OH2 instance as well as on the myopenhab.org Cloud instance. Try to avoid it if you don’t really need this functionality
Maybe some more light on the subject:
This is the events.log right before the “crash”:
2018-12-06 06:01:25.481 [vent.ItemStateChangedEvent] - SENSORLillaFRrDet_ActualTemperature changed from 8.3 ℃ to 8.4 ℃
2018-12-06 06:05:17.266 [hingStatusInfoChangedEvent] - 'tellstick:telldus-live:d49721f1' changed from OFFLINE (COMMUNICATION_ERROR) to ONLINE
2018-12-06 06:05:17.267 [hingStatusInfoChangedEvent] - 'tellstick:switch:d49721f1:129866' changed from OFFLINE (COMMUNICATION_ERROR) to ONLINE
2018-12-06 06:05:17.269 [me.event.ThingUpdatedEvent] - Thing 'tellstick:switch:d49721f1:129866' has been updated.
2018-12-06 06:05:17.270 [hingStatusInfoChangedEvent] - 'tellstick:switch:d49721f1:291127' changed from OFFLINE (COMMUNICATION_ERROR) to ONLINE
Then things went off the rails.
I am running a Tellstick Net remotely at another location and I have a Thing “Telldus Live gateway” to connect and read the status of some sensors and switches.
This binding has caused similar problems (lockup) before but then the error messages have been:
2018-11-30 07:00:51.730 [WARN ] [nal.live.TelldusLiveDeviceController] - TimeoutException error in get
at org.asynchttpclient.netty.NettyResponseFuture.get(NettyResponseFuture.java:169) ~[?:?]
at org.openhab.binding.tellstick.internal.live.TelldusLiveDeviceController.innerCallRest(TelldusLiveDeviceController.java:308) ~[?:?]
at org.openhab.binding.tellstick.internal.live.TelldusLiveDeviceController.callRestMethod(TelldusLiveDeviceController.java:282) ~[?:?]
at org.openhab.binding.tellstick.internal.live.TelldusLiveBridgeHandler.updateDevices(TelldusLiveBridgeHandler.java:127) ~[?:?]
at org.openhab.binding.tellstick.internal.live.TelldusLiveBridgeHandler.refreshDeviceList(TelldusLiveBridgeHandler.java:114) ~[?:?]
at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511) [?:?]
at java.util.concurrent.FutureTask.runAndReset(FutureTask.java:308) [?:?]
at java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.access$301(ScheduledThreadPoolExecutor.java:180) [?:?]
at java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.run(ScheduledThreadPoolExecutor.java:294) [?:?]
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149) [?:?]
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624) [?:?]
at java.lang.Thread.run(Thread.java:748) [?:?]
Could there be any connection? If so, I could remove this binding and associated Items
Or… the internal client in the openhabcloud addon is calling your Jetty Web Server to get info from the REST API about the state of your Items and the calls fail for some reason… (system overload caused by another root-cause or miss-configuration of something?)
Thanks for the tip!
I am on a fiber connection and the OH2 is connected over cable but it is worth checking. If a bad internet connection is the case, then maybe there should be some timeout settable in the service configuration?
I think that the problem is local (see my “improved” assumption )
From the warning message, I understand that: io.openhabcloud.internal.CloudClient sends requests to Jetty and gets null responses
Maybe you can put the io.openhabcloud addon in TRACE log level to see what the heck is the io.openhabcloud.internal.CloudClient asking from Jetty during the failure?
They appear when I launch the android application or update the data in it.
This has been happening for a long time, I do not remember on which assembly I noticed, it was exactly on 1445, now it was updated on 1451, warnings are still present.
This time it happened on the local connection.
One theory about why it happened at night might be connected to how iPhone handles apps, which I know almost nothing about…
When you close an app I wonder if it really is stopped or if it continues to be connected in the background? It might be, that it stayed active in the background until iOS decided to unload it and a pipe was broken. But this is just a theory from my side. In iOS it is easy to go back to an app you have used. You double-click the button and all apps line up until you swipe them off the screen but until you do that, it seems they are somehow at least partly active.
But I might be barking up the wrong tree.
I think that you may be right here (I also don’t know how iOS handles internally the apps)
Personally, I use both Android (heavy use since it’s mine) and iOS (small use from my wife)
I have the openHAB Cloud connector running and I have never seen this in my logs
but… I don’t use myopenhab.org to access remotely my local OH2 instance from the mobile phones.
I have setup a reverse proxy with encryption (https) and authentication and I connect directly from remote.
I was avoiding myopenhab.org for months/years until recently (big fanboy of Intranet of Things concept ) , but then I decided that I want to play with Google Assistant (GA), so it was a “forced” decision I could setup my own openHAB Cloud instance but I consider it too much for my scenario (happy with the reverse proxy and satisfied with the response time of GA over myopenhab.org)
Soon we will have auth embedded in OH2 so the use of myopenhab.org should be reduced and the focus would be integrations (Alexa, IFTTT, Google, etc)
Could this be related to the fact that myOpenHAB went off-line for maintenance around the same time?
I have now uninstalled the CloudConnector binding and will wait to see what happens.
I have also recently installed a reverse proxy (nginx) with encryption (https) and authentication for remote access like you. Only issue there is that grafana graphs are not visible since they are on a different port. I might have to change to Apache2.
I have the same issue with openHAB 2.4.0 on Raspbian and an iOS device: As soon as I go a level deeper in the app (i.e., entering a group defined in the sitemap) or leave the level again, in the logfile two new lines appear:
[io.openhabcloud.internal.CloudClient] - Jetty request xxxxxx failed: null
[io.openhabcloud.internal.CloudClient] - Response Failure: null
This is independent from whether I am using the local or the remote connection.
What seems to be worse:
Every now an then I get a bunch of exceptions. This seems to be related to me leaving the local area (Wifi) and switching to remote: exceptions.txt (13.8 KB)
Being quite a novice to the openHAB world, I have no clue what to do next. But being a developer I really don’t like warnings and exceptions in log files.