openHAB 3.1 Milestone discussion

I struggled a lot until I finally gave up. I have also the same problem with sending email using version 3.1.0.M2. It’s all fine with Stable 3.0.1.

Anyone else have bundles fail to install when upgrading? I’m on docker so just redeploy and everything is good except none of my bundles are installed. I get a message in the log. I was moving from 3.1.0M1 to 3.1.0M2

2021-03-17 19:08:14.520 [ERROR] [core.karaf.internal.FeatureInstaller] - Failed installing ‘openhab-persistence-jdbc-mysql, openhab-misc-openhabcloud, openhab-binding-pioneeravr, openhab-binding-spotify, openhab-binding-alarmdecoder, openhab-binding-amazonechocontrol, openhab-ui-habpanel, openhab-binding-ecobee, openhab-binding-harmonyhub, openhab-binding-mqtt, openhab-persistence-rrd4j, openhab-ui-basic, openhab-binding-openweathermap, openhab-binding-http, openhab-binding-astro, openhab-binding-gpstracker, openhab-binding-samsungtv’: Error:
Error downloading mvn:org.openhab.ui.bundles/org.openhab.ui.basic/3.1.0.M2

Oops, I guess this was my mistake when doing cleanups on Artifactory last night…
Should be fixed now, could you please try again?

1 Like

Thank you - it took an extra restart after the first ‘upgrade’ startup but things are better now.

Only bundle message I got this time was as follows. The onkyo bundle seems to be working though - the things are online.

2021-03-18 10:40:49.443 [WARN ] [org.apache.felix.fileinstall ] - Error while starting bundle: file:/openhab/addons/org.openhab.binding.onkyo-3.0.2-SNAPSHOT.jar
org.osgi.framework.BundleException: Could not resolve module: org.openhab.binding.onkyo [209]
Unresolved requirement: Import-Package: org.openhab.core.config.discovery.upnp

Hello @Kai,

I saw in your “openHAB 3.1 Milestone Builds” that you do not recommend to upgrade to Milestone 2. Is there already a timeline that you can share when M3 will become available?

Please apologize if I am not so familiar with the common openHAB release schedules as I am normally not chasing the latest levels. But I am planning to finally move up from OH2(.5.12) to OH3 and intended use v3.1 unless it is not recommended to upgrade directly to v3.1 for production (and maybe better using 3.0.1 instead). And my plan is to do this during the easter holidays if M3 is available in that time frame :rabbit:

Then you seem to have missed the other update: openHAB 3.1 Milestone Builds - #4 by Kai - all fine, it is safe to use.

But you are right that Milestone 3 is due right now. There was quite some activity on code this weekend and there are a few PRs that should still be merged before Milestone 3, therefore we didn’t build it yesterday. I’d hope that this Thursday or Friday might be a good moment for it - you should have it for Easter :wink: .

9 Likes

A post was split to a new topic: Problem upgrading to 3.1M2

Thanks @kai for the 3.1M3 release!

Regarding the release notes this is missing:

For binding [openwebnet]:

  • PR #10191 is also included in M3; this PR also solves #10242
  • PR #10397 (already in the rel notes) also closes #10298 but is not mentioned in the rel notes

Is possible to add these references in the release notes?

Thank you @Kai :clap: … now I have something to during the easter holidays :stuck_out_tongue_winking_eye:

@Kai many thanks for the new release. Can you please advise if PR #10119 [velux] is actually in the release or not? (As it is not mentioned in the release notes).

@massi & @AndrewFG Thanks for noticing, there were indeed a few PRs missing as they weren’t correctly tagged on Github (note: PRs turn up in the release notes if they either have “bug” or “enhancement” tag on them. Issues are never listed in the release notes, so please always make sure to have a speaking title for every PR, because that’s what users will see in the end. Thanks!

2 Likes

Thank you @Kai Installed 3.0.1.M3 and all seems to be ok, but I have these new warnings in the log:

2021-04-02 12:31:57.340 [WARN ] [ty.util.ssl.SslContextFactory.config] - Trusting all certificates configured for Client@3f704dc3[provider=null,keyStore=null,trustStore=null]
2021-04-02 12:31:57.343 [WARN ] [ty.util.ssl.SslContextFactory.config] - No Client EndPointIdentificationAlgorithm configured for Client@3f704dc3[provider=null,keyStore=null,trustStore=null]
2021-04-02 12:31:57.596 [WARN ] [ty.util.ssl.SslContextFactory.config] - Trusting all certificates configured for Client@7fa5f364[provider=null,keyStore=null,trustStore=null]
2021-04-02 12:31:57.598 [WARN ] [ty.util.ssl.SslContextFactory.config] - No Client EndPointIdentificationAlgorithm configured for Client@7fa5f364[provider=null,keyStore=null,trustStore=null]

This issue persists in 3.1M3:

Edit - forgot link:

What “this” ? Make your post a useful one please.

Since 3.1M2 (and now also on M3) this binding is not available, is that correct?

openhab-addons/bundles/org.openhab.binding.luftdateninfo at main · openhab/openhab-addons (github.com)

It seems that M3 has introduced a missing dependency problem ‘jcifs.netbios.NbtAddress’ in the ‘HDPowerView’ binding. Does anyone have an idea what I need to change in the binding code to resolve it again?

2021-04-02 13:46:51.356 [WARN ] [mmon.WrappedScheduledExecutorService] - Scheduled runnable ended with an exception: 
java.lang.NoClassDefFoundError: jcifs/netbios/NbtAddress
	at org.openhab.binding.hdpowerview.internal.discovery.HDPowerViewHubDiscoveryService.lambda$0(HDPowerViewHubDiscoveryService.java:83) ~[?:?]
	at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:515) ~[?:?]
	at java.util.concurrent.FutureTask.runAndReset(FutureTask.java:305) ~[?:?]
	at java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.run(ScheduledThreadPoolExecutor.java:305) ~[?:?]
	at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1128) [?:?]
	at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:628) [?:?]
	at java.lang.Thread.run(Thread.java:834) [?:?]
Caused by: java.lang.ClassNotFoundException: jcifs.netbios.NbtAddress cannot be found by org.openhab.binding.hdpowerview_3.1.0.M3
	at org.eclipse.osgi.internal.loader.BundleLoader.findClassInternal(BundleLoader.java:484) ~[org.eclipse.osgi-3.12.100.jar:?]
	at org.eclipse.osgi.internal.loader.BundleLoader.findClass(BundleLoader.java:395) ~[org.eclipse.osgi-3.12.100.jar:?]
	at org.eclipse.osgi.internal.loader.BundleLoader.findClass(BundleLoader.java:387) ~[org.eclipse.osgi-3.12.100.jar:?]
	at org.eclipse.osgi.internal.loader.ModuleClassLoader.loadClass(ModuleClassLoader.java:150) ~[org.eclipse.osgi-3.12.100.jar:?]
	at java.lang.ClassLoader.loadClass(ClassLoader.java:522) ~[?:?]
	... 7 more

I have the same warnings after the upgrade to M3.
Is it something to worry about?

I just tested with a plain fresh M3 install and do not see them. I assume they are related to some specific add-on?

Not correct. Tested it and the binding is there as expected:

Please create an issue and we will figure it out. It was probably me who broke that, sorry for that…

Probably. How can I check which one ?

The tried and true method is to either unload all of them and slowly add them back until the issue reappears or removing each one, one at a time until the issue disappears. Neither method is easy, painless or quick, in fact quite the opposite.
With that said, the warning appears to have something to do with secure socket layer and client certificates. Suspect any binding which is connecting securely to a cloud server

Maybe list all the bindings you use and we’ll take a shot at guessing which one is the culprit