[SOLVED] Z-Wave unreliable in 2.5.0.M4

I did manage to get it to remove. I am using a Aeotec Z-Stick Gen5

2 Likes

It might take 1 or 2 network heals to get it all sorted.

Not totally correct. I see that, if your device is broken, you can remove it from the controller and also from “things”, but at the next discover you should find it again … the only way to avoid that is to check also the files, in my case there was still the old file there.

My two cents

Only if it is not removed from the network on the controller.
I think there are currently issues doing this through the binding. Research & troubleshooting are ongoing.

I usually remove the nodes from the controller without the support of the binding itself. But I agree with you, it’s a try approach

Here is the latest conversation I saw on failed nodes.

Troubles will start after network healing…

Me, too.

There have been some recent talk about the nightly heal and can on occasion (usually when a node is misbehaving) hang the network and in some cases knocking the network completely offline. Here is one thread where Chris comments on the nightly heal

Also please check the thread Bruce linked to for info about the remove node in habmin and using zensys tool. Mark and Chris have been working on getting habmin to function better at getting rid of died/ghost nodes

Robert @rrgeorge
if you consider your issue fixed please mark the thread title as [SOLVED] and put a check mark in the thread that helped you fix the issue
thanks

1 Like

In my case: I’ve checked all nodes with “zensys tool 4.7” and “Z-Wave PC Controller 5.38” by sending a NOP to all nodes. All were responding. None is failed. And it worked for 10 months without any problems. But since I made update to Snapshot 1731 and after first network heal, the troubles began.

That is a new and terrible bug. This was never the case before M3.

I do not, and I will not know for certain for at least 24 hours.

I know @5iver had some issues related to healing a while ago, before M3 IIRC.

I reported the same/similar before 2.5M2 as something that should be considered a blocker…

4 Likes

thank you for that and let us know. Do you already have you logs set to debug?

if you need to know how check here then

1 Like

Thank you. When I had discovered it was broken for me in M3, I tried unsuccessfully to find if someone had a similar issue, but didn’t have time to thoroughly look into it. I have no idea why, but I opted to roll back to M1 instead of M2, so I was unaware if the issue affected M2. So all I knew was the issue existed in M3, but not M1.

1 Like

I do now… Thank you for the reminder.

Sorry guys,

I’m not sure to understand what M1-2-3-4 means … I have 2.5.0~S1733-1 (Build #1733) on openhabian and I can say this bug is not visible (at the moment).

No issues so far with zwave

I have the same USB key … and healing at 2 AM … updated yesterday so I’ve already passed this point

There are current release versions of OpenHAB.

The stable release, normally released every 6 months, is currently 2.4 released late last year.

The Testing releases, also known ad Milestone releases are released every month. 2.5M4 or 2,5 Milestone 4 is the current version.

The Unstable, or Snapshot releases are built & released every night and are only designed for developers to test all pieces working together. Snapshot releases are not expected to totally work or be stable.

OK folks… before we all grab our pitchforks and start calling this a bug… this is common traits of a failed node. Sometimes when things are working good for a long time and we upgrade and things suddenly break, it is easy to blame the upgrade itself, when in fact, it is a low battery, node slightly out of range or whatever edge case. In a system with a lot of nodes, it could have gone unnoticed previously.

1 Like