OH2 Z-Wave refactoring and testing... and SECURITY

So much for my theory! :roll_eyes:

I wouldn’t throw it out yet. Looking closer, I’m only seeing 3 (same second) from one other device from 6 months ago… and this is a spare Monoprice Plugin that I move around a lot. I’m saving them to a file to parse out all the nodes.

grep -r "2018" * | egrep "Receive Message = [0-9A-F][0-9A-F] [0-9A-F][0-9A-F] [0-9A-F][0-9A-F] [0-9A-F][0-9A-F] [0-9A-F][0-9A-F] [0-9A-F][0-9A-F] [0-9A-F][0-9A-F] 22 01 00" > /opt/openhab2-backup/appBusy.txt

Do you keep all your old logs? Do you keep them all in userdata/logs, or do you rotate them off to another directory? If the latter, would you mind sharing how you do that?

I use a script for upgrading that first saves a full backup. It’s a lot of data, so at times I go through and manually pull out some things, including logs, and delete the rest. So it’s a mix. I have a task to write a script to clean up the old archives somewhere in my backlog. But rotating them off to another directory is a good idea…

I have a NAS that’s shared between 2 locations that are connected via an IPSEC tunnel. I’d like to rotate the logs off to the NAS so I have them all in one place for analysis. Just haven’t gotten around to trying to figure out how to do it.

Please let me know if you have any hints about the ‘Invalid cross-device link’ error.

That sounds to me as though log4j is using a hardlink when rolling over a log for which the appender’s filePattern is set to a destination on a different partition from the log file.

Further analysis of my logs shows that all of the devices I have of this type also get APP BUSY, but nowhere near as regularly all of my Leviton devices. 791 total since the beginning of the year. Sorting all of the logs entries with APP BUSY by time does not show any pattern (daily heals, reboots, etc.), other than it looks like this happens continuously. I don’t always have logging in DEBUG, so there are gaps in time.

Since the last update, I have only seen two APP BUSY in my logs.

1 Like

Power reset and updated binding didn’t help :frowning: (had to return early because of storm)

Unfortunately I’m not sure what else to do. Both the old, and the new binding seem to be working the same with respect to the polling here. Maybe there’s something else different with the configuration, although I can’t think what that could be that would affect this.

Are you able to provide the log from the old binding from just prior to the 06:57:56.668 log that you provided earlier? I’d like to see what the device sent, and when - I don’t think it’s likely to really help, but at the moment it’s the only unanswered question I have to confirm that the old and new bindings are doing the same thing.

Thanks. I don’t believe that this is impacted by the binding, but I don’t know what else does influence it. For sure it’s seen with other systems as well (such as Vera)…

“Try again later” is a pretty useless message really - it would be nice to have defined what “later” means… Ho-hum…

I was confused then… I thought this…

…was a potential remedy. BTW, I’m now at eight in less than 24 hours.

Has anyone seen this in their logs?


21:20:07.168 [WARN ] [nding.zwave.handler.ZWaveThingHandler] - NODE 10: Invalid item type defined (DecimalType). Assuming DecimalType

Yes, it was an attempt at reducing the issue, but as I have no idea what is causing the device to report that it’s busy, and as other home automation systems are also seeing this (eg Vera - as per the link I posted earlier) I don’t think it’s directly related to the binding.

If you have any other ideas, I’m happy to give them a try, but at the moment the dev binding is doing the same thing as the master binding, so I have no idea why this would be any different - sorry :frowning: .

1 Like

I suspect you are using an old binding. This error message was changed recently so please use a newer version.

223 │ Active │ 80 │ 2.4.0.201807071112 │ ZWave Binding

Yes, its a bit older. Will update ! Thankyou

Ok, here is full log. From OH start, check node 6 and 12. https://anonfile.com/ZfP5xff0b4/openhab.log

I don’t think this is the right log…

Well I just restored to OH 2.2 with working Z-wave binding, enabled debug, confirmed that leviton devices work and uploaded logs. Why it is not right ?

Sorry - so this is a NEW log, not the ones provided earlier?

I was expecting the one that you posted earlier, but if you’ve posted a new one that’s fine as well.

My bad, yes this is new log from today.