G3 Smartstick+ Choking - potential fix.

[Vit]

I am using Zooz7 stick and if the errors are specific to Zooz, I apologize for hijacking the thread.

However, after switching from the 3.0.6.0 to 3.0.5.4, I get:

Jun-03 5:22:01 PM Z-Wave Error Node Change Event for Node 2 on Network D050CF51, CC=COMMAND_CLASS_SENSOR_BINARY but no valid devices were found. (30-03-00-0C)
Jun-03 5:21:41 PM Z-Wave Error Node Change Event for Node 2 on Network D050CF51, CC=COMMAND_CLASS_SENSOR_BINARY but no valid devices were found. (30-03-FF-0C)

Node 2 is a Neo/Haozee Motion/Temp/Humidity sensor. Works fine with v.3.0.6.0


Zooz uses v7.15 library, and it's not recognized by either version of the plugin:

"Z-Wave PC Controller Library Version: Z-Wave 7.15 (Unknown)"

[rjh]

This is fixed. If some of you can run this version of the plugin to verify that would be great. It works ok in my testing here.

Here is the issue, and this probably be gibberish to most of you probably so you can ignore, its not important to understand.

The issue is that the new 700 series serial API uses the bridge controller library. The older sticks all used the static library. They no longer support the static lib. The difference is that the multicast frame has an extra byte in it so I need to check the library type and modify the packet.

The fix is in 3.0.6.1 which is in the updater now.

[rjh]

See my next post for the answer...

Interesting. So, if this has always been the case, why did this scenario work promptly with the 500-series controller, but has about an 8-second delay on the 700-series controller? It sounds like this is a bug that needs to be sent to SiLabs to investigate.

[SteveW]

Yes, this has always been true. What it does is send the multicast command (non routed), then follows up with a singlecast to each node. But right now its not even sending the multicast, only the follow up packets. Still trying to see why. We do not have access to the firmware in the USB stick, its provided by Silicon Labs so its hard to debug. I wish they would just give the developers the source to that, I guess they feel there is some special mojo in there.

Interesting. So, if this has always been the case, why did this scenario work promptly with the 500-series controller, but has about an 8-second delay on the 700-series controller? It sounds like this is a bug that needs to be sent to SiLabs to investigate.

[rjh]

Yes, this has always been true. What it does is send the multicast command (non routed), then follows up with a singlecast to each node. But right now its not even sending the multicast, only the follow up packets. Still trying to see why. We do not have access to the firmware in the USB stick, its provided by Silicon Labs so its hard to debug. I wish they would just give the developers the source to that, I guess they feel there is some special mojo in there.

Per the documentation, multicast and broadcast may only be used in direct range (1.2.1).

https://www.silabs.com/documents/pub...s/SRN13181.pdf

[ewkearns]

I just want to post a note that I am looking at this. It appears that the 700 series interface does not handle the multicast request. A multicast is only used when you try to send the same command to multiple devices. When I have more information I will post.

Per the documentation, multicast and broadcast may only be used in direct range (1.2.1).

https://www.silabs.com/documents/pub...s/SRN13181.pdf

[rjh]

I just want to post a note that I am looking at this. It appears that the 700 series interface does not handle the multicast request. A multicast is only used when you try to send the same command to multiple devices. When I have more information I will post.

[avpman]

I think merging threads is kind of difficult and confusing. The best thing is probably to just send a link to the most active thread (as I tried above) and then people move to that thread. I am actually rather happy that some of the HS team members participate in this forum. I wouldn't want them to handle stuff like policing threads, combining threads, etc. Instead, have them solve our issues... Hopefully by now they have figured out there is an issue with the new z-wave stick.

Ticketed last evening. They asked for tech access to my system. Hopefully, they're looking into it. Another few days and I'm rolling everything back to the old stick. 👍

[mulu]

I think merging threads is kind of difficult and confusing. The best thing is probably to just send a link to the most active thread (as I tried above) and then people move to that thread. I am actually rather happy that some of the HS team members participate in this forum. I wouldn't want them to handle stuff like policing threads, combining threads, etc. Instead, have them solve our issues... Hopefully by now they have figured out there is an issue with the new z-wave stick.

[avpman]

This has already been reported and is being discussed on a couple of other threads. It would be great if you'd just pick one and post there. We are all having the same problem. Please open a support ticket.

Opened a ticket last night. HS support asked me to open support assistance on MyHS. Is there a moderator here who could merge all the threads?

[SteveW]

This has already been reported and is being discussed on a couple of other threads. It would be great if you'd just pick one and post there. We are all having the same problem. Please open a support ticket.

[avpman]

What happens when you do a node connectivity test on the network? Any errors? What does the log look like when you experience delays?

Connectivity test passes and no errors in the log when I run the event. Apparently, others are having similar problems. Please see posts in this thread.

[mulu]

See this post https://forums.homeseer.com/forum/ho...80#post1476780

[mulu]

You might want to search the forums. There was another thread that also mentions delays of some 30-40s and multiple users are experiencing it. From what I recall it happens with different sticks, not just the HS one. So it appears it's something software related. Maybe there is an update in that thread about this problem.

[macromark]

I've been using the new G3 smart stick+ since it was released with V3.0.6.0 of the Zwave plugin. If I am controlling one device after another, as in a bedtime routine that shuts off one light after another, the lights no longer shut off quickly. There seems to be as long as a 40 second or more delay between when the devices actually respond and when the commands were sent. This did not happen with the previous Smartstick.

Is anyone else having this problem?

What happens when you do a node connectivity test on the network? Any errors? What does the log look like when you experience delays?