Before you toast everything, let’s do a little “horsy-doggy-ducky”.
So (being overly explicit)…when you say the “nodes load up in the log again” , do you mean they show in the zwave log ? (configured in logback.xml)…you should have this set to a DEBUG level until this gets straightened out.
Again being overly explicit, in my own logback.xml (or logback_debug.xml ), (hit by ./start.sh and ./start_debug.sh, respectively) I have the following entries:
logs/zwave.log
logs/zwave-%d{yyyy-ww}.log.zip
30
%d{yyyy-MM-dd HH:mm:ss.SSS} %-5level %logger{30}[:%line]- %msg%n%ex{5}
and
IF the nodes show in the zwave log, you’ll see the nodes getting acquired by the controller and (pretty voluminous) chatter along the network. At the near beginning of the log you’ll find a listing of the node numbers that are known to the controller (this is before all the “is it dead/alive/Schodinger’s node” chatter.
If you do not see the nodes listed at the near-beginning of the zwave log, then you should look at your node inclusion/exclusion discipline. (Been bit by this myself). Are the nodes included in your controller ? I don’t mean included by habmin, I mean have you walked around to your devices, z-stick in hand, put your z-stick into inclusion mode and seen the lights flashing with the “I’ve got you” signalling. (word of advice, given the number of times your been round the barn here, you’ll like be tempted to try an re-include your devices. I would recommend that for each device, you first do an “exclude” cycle before any “include”.)
If you do see the nodes at the beginning of the zwave log, then you have to look at the detailed messages in the DEBUG log to see what might be going on here. I’ll (of course) defer to @chris on the details, but deep in the log weeds you can find useful info sometimes.