Hi,
I’m having issues in configuring my Weinzierl 762 Linemaster in ROUTER mode. If I try the TUNNEL setup it works flawlessly.
Openhab 2.4 runs on a FreeBSD 12.0-RELEASE-p3. The /16 network has no routers between the devices. Openhab is on machine 10.11.1.10. The dhcp server is setup to assign 10.11.1.20 to the Weinzierl. I’m using ETS inside and I programmed the interface to “route”, “acknowledge” and “repeat” everything and it’s address is 12.0.0. I checked my network router that it was not filtering any multicast traffic. I removed any such filter on the FreeBSD’s own firewall as well.
My .things config:
Bridge knx:ip:bridge "Weinzierl KNX IP LineMaster 762" [
type="ROUTER"
] {
Thing device bin5 "DATEC 4x Out 8x In" @ "H2 Dining" [
address="12.0.7"
] {
Type contact : H2_Living_Window1 "H2 Living Window 1" [ ga="1.009:<0/5/2" ]
Type contact : H2_Living_Window2 "H2 Living Window 2" [ ga="1.009:<0/5/0" ]
Type contact : H2_Living_Window3 "H2 Living Window 3" [ ga="1.009:<0/5/1" ]
Type contact : H2_Living_Door "H2 Living Door" [ ga="1.009:<0/5/3" ]
Type contact : H2_Living_Motion "H2 Living Motion" [ ga="1.009:<0/5/11" ]
Type contact : H2_Dining_Motion "H2 Dining Motion" [ ga="1.009:<0/5/8" ]
}
}
When I turn everything on PaperUI shows the bridge immediately online but not the “bin5” device of the configuration.
Telegrams from the KNX bus seem to work fine. To test it I activated a cyclical sending on the binary interface and as expected, every 60 seconds I see 6 packets with tcpdump in the form of:
IP (tos 0x0, ttl 16, id 214, offset 0, flags [none], proto UDP (17), length 45)
10.11.1.20.3671 > 224.0.23.12.3671: UDP, length 17
That is from the KNX router to the multicast address as I assume it is supposed to happen.
I can also see on the karaf console the six matching entries similar to:
[DEBUG] [x.internal.handler.DeviceThingHandler] - onGroupWrite Thing 'knx:device:bridge:bin5' received a GroupValueWrite telegram from '12.0.7' for destination '0/5/11'
The problem occurs in the opposite direction. No polling from openhab gets any reply from the KNX bus. My karaf log is full of:
[DEBUG] [knx.internal.client.AbstractKNXClient] - Could not read value for datapoint 0/5/11: timeout waiting for group read response from 0/5/11. Going to retry.
Looking at the tcpdump though I can see the packets running the other way in the form of:
IP (tos 0x0, ttl 1, id 39976, offset 0, flags [none], proto UDP (17), length 45)
10.11.1.10.3671 > 224.0.23.12.3671: UDP, length 17
(that is from openhab’s ip to the multicast address).
I’ve seen a lot of similar “router mode” related discussions but all seem to have a bunch of errors to work with. Mine simply seems to ignore the traffic from the net to the bus.
Am I missing something obvious? It’s my first setup so I’m pretty much clueless about a lot of things here.
Any help is kindly appreciated,
Riccardo