KNX Control Items Warning negative confirmation

I have the problem, that after a “postUpdate” to a switch-control item the event log is flooded with many warnings “negative confirmation of 8/3/100: 2e 00 9d e0 10 c9 43 64 01 00 80”. This warning comes every 100 ms.

Here my KNX thing definition:

Bridge knx:ip:bridge "ABB IPS/S 3.1.1"[ 
    type="TUNNEL",
    ipAddress="192.168.178.211",
	portNumber=3671,
    localIP="192.168.178.210",
    readingPause=100,
    autoReconnectPeriod=60,
    localSourceAddr="1.0.201"
]
{
    Thing device zentral "Zentralfunktionen" [] {
        // Alarmanlage
        Type switch-control  :  Alarmanlage_Nord_scharf  "Alarmanlage Nord scharf"  [ ga="1.001:8/2/100+8/1/100" ]
        Type switch-control  :  Alarmanlage_Nord_ausgeloest  "Alarmanlage Nord ausgelöst"  [ ga="1.001:8/3/100" ]
    }
}

Here my item definition:

Switch          NordAlarmanlageScharf               "Alarmanlage scharf"    <alarm>             (NordAlarmanlage)               {channel="knx:device:bridge:zentral:Alarmanlage_Nord_scharf"}
Switch          NordAlarmanlageAusgeloest           "Alarmanlage ausgelöst" <alarm>             (NordAlarmanlage)               {channel="knx:device:bridge:zentral:Alarmanlage_Nord_ausgeloest"}

This happens with other *-control items as well.
I’m using openHAB 4.1.1. I think this didn’t happen in the versions before.

It’s a litte bit annoying because the log is flooded with this warning until a complete restart of openHAB.
Is it a bug or a mistake in the definitions? Does anyone have an idea what’s wrong?

Don’t set a localSourceAddr. This is NOT the individual address of the interface.

I removed the setting for the localSourceAddr for the KNX thing but the same error still occurs.
Do you have other ideas what could lead to this problem?

As you set the Thing with text files: Did you restart openHAB?

Yes I did a restart. I also restart regularly openHAB to get rid of the flute of warnings.

Udo is right on the local Source Address, just remove it.
A bit more on the scenario itself would be nice, so we know what you are actually trying to achieve here.

Just thinking in my humble mind, but I might be totally wrong :
You are sending a command to the bus and somehow a read after update happens.

My understanding of the switch-control is as follows:
It models a virtual actor for knx inside of OH that represents a device not actually being a knx actor on the bus.

OH: postUpdate() on switch-control → Causes command to be sent to knx Bus and vice versa.

But the device you are addressing with your group address on the knx side does not exist (?) - so this might be the reason you do get a negative confirmation.
Did try to use sending a command? Here I’d assume the flow:
OH:SendCommand() to knx-control → knx: receives an status update. This should be fine as it is just “posting” the value to the bus.

Please have a look into the Monitor in ETS and share what is happening there, filtered on the involved Bus Addresses.

BR
Thomas

Thanks for your respones!

And shame on me - you led me in the right direction. I had an mistake in ETS so that the KNX group address 8/3/100 was not used for any device (only in a dummy device for the line coupler).
So that was the problem, that no device was acknowledging to that group address.

In general I use the signals in combination with a MDT Glass Push-button II to activate / deactivate and visualize the alarm.