Java Exception "upnp-async" after upgrade from 4.0.3 to 4.1.0

Setup, Configuration and Use Migration
1

I tried to do an update from 4.0.3 to 4.1.

I’m running OH in a Unraid/Docker environment (stable for many years).
After upgrade to 4.1 I get the following output in the logs:

Exception in thread "upnp-async-11" java.lang.IllegalArgumentException: Invalid URI host: null (authority: :30080)
        at org.eclipse.jetty.client.HttpClient.checkHost(HttpClient.java:541)
        at org.eclipse.jetty.client.HttpClient.newHttpRequest(HttpClient.java:526)
        at org.eclipse.jetty.client.HttpClient.newRequest(HttpClient.java:484)
        at org.jupnp.transport.impl.jetty.JettyStreamClientImpl.createRequest(JettyStreamClientImpl.java:113)
        at org.jupnp.transport.impl.jetty.JettyStreamClientImpl.createRequest(JettyStreamClientImpl.java:1)
        at org.jupnp.transport.spi.AbstractStreamClient.sendRequest(AbstractStreamClient.java:87)
        at org.jupnp.transport.RouterImpl.send(RouterImpl.java:327)
        at org.jupnp.protocol.RetrieveRemoteDescriptors.describeService(RetrieveRemoteDescriptors.java:325)
        at org.jupnp.protocol.RetrieveRemoteDescriptors.describeServices(RetrieveRemoteDescriptors.java:263)
        at org.jupnp.protocol.RetrieveRemoteDescriptors.describe(RetrieveRemoteDescriptors.java:205)
        at org.jupnp.protocol.RetrieveRemoteDescriptors.describe(RetrieveRemoteDescriptors.java:184)
        at org.jupnp.protocol.RetrieveRemoteDescriptors.run(RetrieveRemoteDescriptors.java:104)
        at java.base/java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1136)
        at java.base/java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:635)
        at java.base/java.lang.Thread.run(Thread.java:833)

I have no idea where I can start my research for this issue :frowning:

Has anyone a hint for me what coud be the reason for this excemtion?

After rollback to 4.0.3, OH is working in the same environment stable.

2

It looks similar to this issue but it is supposed to be fixed:

https://github.com/openhab/openhab-core/issues/2494

1 Like
3

I think in your case the device communicates an empty host with port :30080 and not null. So probably an additional check (or catch) in the jupnp library needs to be added for it to be suppressed.

1 Like
4

If you didn’t get this exception on older openHAB versions, it may be due to the “UPnP Suggestion Finder” which can be disabled in the “Advanced” settings of “Add-on Management”:

image
image832×487 39.3 KB

5

I set up a new container.
And, after 20 rounds the OH runs. So, this message comes with the start of OH and after X rounds, it stops an OH is running.

I activate the debug log (log:set debug org.jupnp).
But I don’t see any message with the port 30080 ..

6

I have created this issue to track it:

https://github.com/jupnp/jupnp/issues/159

2 Likes
7

There’s now also a PR to fix it :slight_smile:

https://github.com/jupnp/jupnp/pull/161

1 Like