Dennis V McEnaney

For the benefit of others; BitDefender Total Security seems to exhibit a similar problem, it deletes the debug build of 'ProperTable.exe', i.e. requires an BitDefender exception to be created.

F.Y.I. The 'DADR' value is generally a virtual MAC address, not a registered one (- with an OU assigned/included). But otherwise, proper MAC/physical/burned-in addresses should be unique; unless the unlikely case that Carrera Technology supposedly once had whereby the country (/Singapore?) that sold them bulk NICs had apparently copied the firmware verbatim - they all had the same MAC address assigned/burnt onto them; network-conflict city (!). ;P

Maybe to add to Monsieur Chaxel's answer; depending upon your location/situation, it might rely upon bog-standard network (/trunk) routing being fully configured if you are going across/into several VLANs/subnets (- as opposed to using a BBMD), and I don't believe you can target routed/gatewayed (/child) devices, i.e. ones that have a DNET/network segement # & currently DADR/virtual MAC address pairing assigned via YABE - but I could be wrong (?).

I wonder if you could (easily) use the room-simulator (with changes) as a test-bed.

But to add to Matthew's answer; the "central control system" and/or peers upon the 'internetwork' can also use BACnet Standard provided mechanisms/'services' to (actively every now & again) poll for new devices and/or information for discovered devices, e.g. fire a 'Who-Is' (unconfirmed-service) rather than (only passively) wait for the receipt of 'I-Am's. So two heads sometimes (if not generally) being better than one, sometimes the joint benefit of both a push & poll approach might prove more beneficial,...

Out of interest, I also found it hard to easily find a sample, so (as part of answering a question) I posted a modified example here: https://stackoverflow.com/questions/76543619/bacnet-inside-docker-container/76564916#76564916 Here (in isolation) is the code - but the post has some more information: // cls & java -classpath Lib\bacnet4j-3.2.2.jar;Lib\slf4j-api-1.7.9.jar;Lib\commons-lang3-3.12.0.jar Main.java // https://github.com/MangoAutomation/BACnet4J/tree/master/release/3.2.2/ // "Download"...

Just some thoughts. Some few controllers can have the concept of a proprietary/behind-the-scenes 'user', which might be lacking the right permissions to write to the schedule (by default). Schedules (I.M.H.O.) are/can be a complex thing, so it could still be possible that YABE has an edge-case/very specific bug. But maybe (more likely) the controller is exhibiting a non-standard behaviour/restriction, the kind of thing that the BTL testing warns about/tries to encourage some leniency/flexibility...

(If you also want to ensure that the non-device/child objects are included too, you'll have to click upon/scroll through all of them, each object in turn, before then saving the YabeMap file.)