Bad Z-Wave route

Running Indigo 7.0.2 with Z-Wave plugin 2.0.35

Had issues with 2 z-wave switches (was 11, 12) that I excluded and re-included (now 151, 152). They still show up in various device neighbor lists. I have been unsuccessful in using a resync or optimize to correct a subset of my z-wave devices..

I've included the device settings which shows the original address of the modules I re-included - 11, 12 as the first neighbors.

Neighbors: 11, 12, 16, 49, 51, 54, 55, 56, 67, 71, 84, 111, 137, 144

When I Define and Sync I get the following messages in the log:

Z-Wave Syncing - started for "East Beds"
Z-Wave Debug SENT getNodeNeighbors: 01 06 00 80 32 01 01 4B
Z-Wave Debug RCVD getNodeNeighbors: 01 20 01 80 00 80 00 00 00 00 E5 00 44 00 08 00 00 40 00 00 00 81 00 00 00 00 00 00 00 00 00 00 00 B6
Z-Wave Debug . . getNodeNeighbors: nodeId 050, neighbors: 16, 49, 51, 54, 55, 56, 67, 71, 84, 111, 137, 144
Z-Wave Syncing - retrieved module neighbors list: 16, 49, 51, 54, 55, 56, 67, 71, 84, 111, 137, 144
Z-Wave Syncing - assigning return route to "050 - East Beds"
Z-Wave Debug SENT assignReturnRoute: 01 05 00 46 32 01 8F
Z-Wave Debug RCVD assignReturnRoute: 01 04 01 46 01 BD
Z-Wave Debug RCVD assignReturnRoute: 01 05 00 46 01 01 BC
Z-Wave Debug . . assignReturnRoute: node 050, success 0
Z-Wave Error Syncing - failed to assign return route
Z-Wave Debug SENT requestNodeInfo: 01 06 00 60 32 24 01 8E
Z-Wave Error Syncing - failed
Z-Wave Error Timeout waiting for "050 - East Beds". Module might be asleep, or is unreachable.
Z-Wave Debug RCVD requestNodeInfo: 01 06 00 49 81 00 00 31
Z-Wave Debug RCVD requestNodeInfo: failed (module asleep or doesn't support command)

Is there another way to force the removal of the old device addresses from the Neighbor list?

Can you provide sample raw Z-Wave commands to override the routing manually?

Is there also some way to limit the size of the routing tables? I have over 100 z-wave devices in my network.

[matt (support)]

Instead of re-Define/Syncing the modules have you tried the Interfaces->Z-Wave->Optimize Z-Wave Network... menu item?

If a full optimize of All Devices doesn't help (and that will take a very long time with 100+ Z-Wave devices), then I'd suggest optimizing just a couple of modules near your Mac and then try optimizing some of the module that still have the old neighbor IDs in them.

Unfortunately, there isn't much of a Z-Wave Controller API in regards to how the routing works. All Indigo can do is ask a module to try to re-discover its neighbors and then it can read out the current neighbor list.

Matt,

After numerous retries all devices with the bad routes did eventually update. Either the Define/Sync or Optimize Z-wave network command worked in different situations. I also had to reload the z-wave network after the attempts to get a clean update.

I also get numerous "rcvd incomplete interface packet (skipping)..." and "rcvd unexpected bytes (sent ACK):..." errors in the log since upgrading to 7. Wondering if my network is too large or I have a faulty device(s).

Examples:
Z-Wave Error rcvd incomplete interface packet (skipping): 01 10 00 49 84 22 0A 04 10 01 27 73 70 86 72 77 9D
Z-Wave Error rcvd incomplete interface packet (skipping): 01 07 00 13 9E 00 0C 79
Z-Wave Error rcvd incomplete interface packet (skipping): 01 1D 00 04 89 17 98 81 DF 20 E9 05 BB C6 47 A8 06 6F 4C 97 69 D9 F7 2F AF 68 4C BC C8 C5
Z-Wave Error rcvd incomplete interface packet: 01 1A 00 04 00 89 14 98 81 69 03 E1 E3 28 D6 12 C9 28 53 43 58 9A A7 71 D8 73 A9 06 (missing 2 bytes)
Z-Wave Error rcvd incomplete interface packet (skipping): 01 08 00 00 90 02 98 40 B9
Z-Wave Error rcvd incomplete interface packet (skipping): 01 07 00 13 0C 00 00 EA
Z-Wave Error rcvd unexpected bytes (sent ACK): 0A 00 04 00 54 04 86 14 32 02 03

[matt (support)]

How many of those do you see a day?

They definitely are more likely to happen if you have a lot of Z-Wave traffic. They basically mean that bad/corrupt data is being sent to Indigo from the Z-Wave Controller. I think it gets confused if too much data is coming in at the same time.

I've debated not logging them as errors, but I think it is useful to still see them in the Event Log so you can get a feel for how frequently they occur (and if they get more frequent for some reason).

12/18 135 errors, 12/19 22, 12/20 27, 12/21 51, 12/22 41, 12/23 54, 12/24 59, 12/25 42, 12/26 48, 12/27 48 -> 22-135 errors

Up until last month I was polling status every hour per device but backed all devices down to polling "Only When Activity Detected" to lower the traffic and reduce the error count.

I'm trying to cover a large area and number of devices with one Indigo system. If I could have manually manipulate the routes it would have been useful. Also, if Indigo could could scale out to support multiple systems with one view that would also help. I have to maintain two additional Vera Edge systems to cover the longer distances.

Thanks for the support.

Wow - 150 devices PLUS two Vera's?!

That's huge!


Sent from my iPhone using Tapatalk