[homekit] "No response" error on all openHAB items

Local HAP service is behaving as I would expect (while the service is responding) then nothing when it stops responding.

I experience the same issues that HomeKit loses the connection after while at least what I can see on my iPhone. When toggling the mDNS switch in the binding settings it immediately reactivates. I am on 3.1.0 M3 the last four weeks so might it be that other Linux updates caused the issue? No other changes in my network additionally.

please make sure

Do you think you can fix this problem with ipv6 at some point? Or do we have to permanently disable ipv6 for Homekit. Nevertheless thank you for your great support.

i dont fully understand why mDNS has an issue with IPv6 in the first place. it could even be something at JVM or java library level. in general, ipv6 create more issue than it helps, not just with openHAB… personally, i always disable it as the first step on any installations.

but yes, it would be better, if it could also work with ipv6 enable… probably i could add logic to ignore ipv6 address… let see.

I did not disable ipv6 on my Host system, where Docker is running. So on my side as wrote above, everything works fine, since I use the switch 2 times after a reboot of the container. So ipv6 shouldn’t be the problem. For me it seems, that mDNS is not able to start publishing after the container start. @yfre if I could provide you with more input, logs or configs, just let me know.

Can anybody confirm that just toggling mDNS (on to off or off to on) in the openhab HomeKit settings re-activates the binding? Isn’t that an indication that the issue is more with the binding than with the operating system? Furthermore my binding hasn’t changed four weeks and worked perfectly - on the other hand I have received several OS updates in this period. My guess is: Something has been updated within the raspberry OS that causes mDNS issues in the binding.

I can confirm that the simple act of toggling mDNS does indeed re-activate the binding. Also, simply changing the port number used will also re-activate it.

I’ve disabled IPv6 in the kernel and still the problem occurs.

1 Like

I’ve disabled IPv6 in the kernel too and still the problem occurs.
I have a rule to restart homekit binding every 2 hours (I don’t like it but it works).
Andrea

It’s an odd one, prior to version 3 it was working fine even with IPv6 enabled, would stay up and running forever.

if usage of openHAB mDNS is disabled in setting, then the source code from version 2.x is used. the only difference is the newer version of mDNS library.

Not sure whether it helps diagnose the issue / confirm anything but I was having this problem of all HomeKit devices appearing offline whereas they had previously been working. I’m on latest milestone 4. For the past week or so at least, I could get it working for a bit by switching the mDNS setting in HomeKit settings on OH - didn’t seem to matter whether it was on or off, it worked for a bit after I’d switched it but only for an hour each time at most.

This morning I went into the settings on my Apple AirPort Express which I use as a router and turned off IPv6 connection sharing. I almost don’t want to write this down in case I jinx it but it’s been working for hours since, without my having changed the mDNS setting.

1 Like

I am on 3.1.0 M4 on Synology Docker and with disconnected IPv6 and disconnected mDNS it works without interruptions. I transitioned from 3.1.0 M3 to M4 and had multiple openHAB reboots and a power outage even, but the HomeKit keeps coming up reliably. Also the Users in the homekit.json do stay.

1 Like

I’m still having the same issues, and am wondering if because I too am using airport expresses and extremes as wireless access points (IPV6 is disabled on my router and the openhab installation) that they are still using IPv6 for local link traffic as it’s impossible to switch this off completely. My HomeKit hub is an Apple TV connected to these airport devices so a problem perhaps?

Seems I did indeed jinx it. It was working all day until just 30 mins or so ago, then it stopped. And now switching the mDNS setting appears to have no effect. Since it stopped working I am getting a lot of errors in my log that look like this:

2021-05-08 17:10:15.411 [WARN ] [.server.impl.connections.HttpSession] - Exception encountered while verifying pairing
java.lang.Exception: Unknown user: C84CA8E7-0D54-4854-BAAB-A37FC027680F
	at io.github.hapjava.server.impl.pairing.PairVerificationManager.stage2(PairVerificationManager.java:113) ~[bundleFile:?]
	at io.github.hapjava.server.impl.pairing.PairVerificationManager.handle(PairVerificationManager.java:51) ~[bundleFile:?]
	at io.github.hapjava.server.impl.connections.HttpSession.handlePairVerify(HttpSession.java:127) [bundleFile:?]
	at io.github.hapjava.server.impl.connections.HttpSession.handleRequest(HttpSession.java:56) [bundleFile:?]
	at io.github.hapjava.server.impl.connections.ConnectionImpl.doHandleRequest(ConnectionImpl.java:56) [bundleFile:?]
	at io.github.hapjava.server.impl.connections.ConnectionImpl.handleRequest(ConnectionImpl.java:49) [bundleFile:?]
	at io.github.hapjava.server.impl.http.impl.AccessoryHandler.channelRead0(AccessoryHandler.java:52) [bundleFile:?]
	at io.github.hapjava.server.impl.http.impl.AccessoryHandler.channelRead0(AccessoryHandler.java:17) [bundleFile:?]
	at io.netty.channel.SimpleChannelInboundHandler.channelRead(SimpleChannelInboundHandler.java:99) [bundleFile:4.1.63.Final]
	at io.netty.channel.AbstractChannelHandlerContext.invokeChannelRead(AbstractChannelHandlerContext.java:379) [bundleFile:4.1.63.Final]
	at io.netty.channel.AbstractChannelHandlerContext.access$600(AbstractChannelHandlerContext.java:61) [bundleFile:4.1.63.Final]
	at io.netty.channel.AbstractChannelHandlerContext$7.run(AbstractChannelHandlerContext.java:370) [bundleFile:4.1.63.Final]
	at io.netty.util.concurrent.DefaultEventExecutor.run(DefaultEventExecutor.java:66) [bundleFile:4.1.63.Final]
	at io.netty.util.concurrent.SingleThreadEventExecutor$4.run(SingleThreadEventExecutor.java:989) [bundleFile:4.1.63.Final]
	at io.netty.util.internal.ThreadExecutorMap$2.run(ThreadExecutorMap.java:74) [bundleFile:4.1.63.Final]
	at io.netty.util.concurrent.FastThreadLocalRunnable.run(FastThreadLocalRunnable.java:30) [bundleFile:4.1.63.Final]
	at java.lang.Thread.run(Thread.java:834) [?:?]

On searching for information about this new error log, I found this WARNING Exception encountered while verifying pairing java.lang.Exception: Unknown user: ... every 75 sec - #3 by ndye and followed the instructions from yfre to find a backup HomeKit.json file with “user” in and use it to overwrite the current one. For now at any rate, that seems to have fixed things. But I’m not clear whether this was in any way related to the other problem in this thread. The bug referenced above looks like it was apparently fixed last year but it seems that it’s back.

please make sure you have switched private address for your home wifi off

private address generate regularly new mac address, which is used in openhab as homekit user. it is great feature for more privacy in the public wifi but not need in the home wifi.

Does this not suggest it’s not the binding that is causing the issue then, because until openhab 3.0 everything worked perfectly for me. Now, I’m trying a build of openhab 3.0.2 with HomeKit binding (no mDNS option obviously) and it’s still giving no response after a short period of time…

I switched off the private address setting on my iPhone. But same result. I think on my side, the problem must something with starting the binding. Because everything is working great till I restart the application.

Hello!

Add me to the list.
I even tried the latest snapshot.
Private Adress disabled and switching mdns heals it for me as well.
If you need any logs i would be happy to help.

1 Like