Understanding ZWave Logs

🛎️ Get Help
1

I have a mostly central location for my C7 and all devices paired without issue from their final location. Over time, some of the further contact sensors would miss events. Besides the C7 I have one other mains-powered device that's off to one side of the home. FWIW the sensors that went flaky were at the opposite end of the home from the mains-powered unit, making the C7 actually closest. So to repeat signals, so I purchased an Aeotec Range Extender 7 and installed it near the flaky units - paired right up.

So am I supposed to just trust the unit is repeating signals? I swear at one point on the ZWave page where it lists all of the Zdevices, one showed 01->0A->13, which I suppose means repeating occurred, but that's never shown again since. The topology map shows what might mean neighboring, I guess, but how would I know?

Then the ZWave logs. How to decipher these?

Garage Window (Rear)2022-09-28 12:18:15.773 seqNo: 233, routeChanged: false, transmissionTime: 1ms, repeaters: None, speed: 100 kbs, rssi: [-90 dBm, N/A, N/A, N/A, N/A], Ack channel: 0, Transmit channel: 0
Garage Window (Front)2022-09-28 12:11:17.336 seqNo: 232, routeChanged: false, transmissionTime: 1ms, repeaters: None, speed: 100 kbs, rssi: [-89 dBm, N/A, N/A, N/A, N/A], Ack channel: 0, Transmit channel: 0
Zoe's Bedroom Window2022-09-28 12:01:28.684 seqNo: 231, routeChanged: false, transmissionTime: 1ms, repeaters: None, speed: 40 kbs, rssi: [-91 dBm, N/A, N/A, N/A, N/A], Ack channel: 1, Transmit channel: 1
Guest Bedroom Window2022-09-28 12:00:54.927 seqNo: 230, routeChanged: false, transmissionTime: 1ms, repeaters: None, speed: 100 kbs, rssi: [-83 dBm, N/A, N/A, N/A, N/A], Ack channel: 0, Transmit channel: 0
Sliding Door (Left)2022-09-28 11:55:48.169 seqNo: 229, routeChanged: false, transmissionTime: 1ms, repeaters: None, speed: 40 kbs, rssi: [-87 dBm, N/A, N/A, N/A, N/A], Ack channel: 1, Transmit channel: 1

Repeaters: None means what? It's communicating directly with the C7?

I'd sure appreciate some decoding to help me on my way. Thanks!

2

Can you post a screen shot of your z-wave details page in it's entirety?? We'll get more info from that

3

Thanks for offering to review. Here you go. For context, I ghetto mapped the layout. FWIW, I'm hoping devices around the front bedrooms will hop through 0x19. I think the issue up there is the entry area is a deep "tunnel", and its stucco so clad with chicken wire. The repeater (0x19) is inside the house. 0x15 was my problem child, yes, both the Zwave and the actual child :face_with_diagonal_mouth:

zwaveDevicesAll
zwaveDevicesAll1723×1414 204 KB

zwaveMapFull
zwaveMapFull555×866 104 KB

4

So your mesh isn't great You only have 2 mains devices and nothing is routing through them. Your edge devices are connected directly to the hub. I would add some repeaters around the house in strategic places (I recommend the Ring v2 Extender, fabulous repeater that has battery backup and can trigger rules based on switching from mains to battery)

5

Thanks - so this is what I can't understand... why would the 4 devices that surround 0x19 ignore it's proximity and refuse routing through it?

The only thing I can assume is that distance isn't an issue so they just go direct. Maybe something in the info I posted would support that? Would a device that thinks it has a decent enough connection ignore a repeater with more signal than the hub itself?

6

Why? They have pretty good signals talking directly to the controller.

Clearly 0x0B hasn't communicated to the hub since the last reboot, so there's no signal for that node, but the signals for the other three (0x0C, 0x14, 0x15) are rock solid at 13-17 dB.

Why slow things down by wanting them to go through a repeater?

The one recommendation I do have is to reset the Aeotec repeater (node 0x19) and pair it without any security.

1 Like
7

because you added 0x19 after you added the other devices. They don't know about 19 yet, z-wave is crap like that, it doesn't learn. You would need to make them fail to get them to learn a new route or re-join them

1 Like
8

Very true. However, all of these seem to be z-wave+ nodes, so in theory, they should learn other routes using explorer frames.

My guess is that the existing topology is probably the most efficient, and has been "selected" for by the mesh.

9

they should but, that's not quite how explorer frames work. "Explorer frames" sounds like they go out and look for more routes were as in fact this only happens when they fail. If the miss a beat then they send out explorer fames to find a new path. They cling like hell to the old path even if there is a much better one. I wonder if it would work to shut down the hub and trigger the devices? or you wrap the hub in tin foil :rofl:

1 Like
10

Could there ever a better path than a straight line? :smile:

(I know I'm being a wee bit facetious ....)

11

o yeah cant beat it, but its about power, what's better a direct path that you need to shout for and use more power or one extra hop but you can whisper whilst still knowing if you shout it will get there.

1 Like
12

That's an excellent point. On one of my zigbee meshes, batteries on some end-devices last till 2.6V (fresh CR2450 battery = 3.0-3.1V). It takes about 2 years for batteries to run down to 2.6V. But that mesh has a repeater within 5-10 feet of every single end-device.

13

I'll try this tonight.

Perhaps the missed events from the front room is unrelated to signal? It is sporadic.

If removing/repairing the extender without S2 does nothing I'll try this.

14

I believe Bryan has said that if you run a repair on an individual device that may result it in finding a new route.

If a full repair (entire mesh) is done, the same may happen on a larger scale. But it can take days and weeks for changes to occur.

And I see "dumb" routes all the time, but then again if things if things are working well for me (which they are), how dumb are the routes, really... :smiley:

2 Likes
15

Is this by clicking the REFRESH button in the device table?

I just removed/repaired the repeater without S2, then performed a full repair, been sitting here for ~15 minutes
image

16

Not advisable when most of your devices are battery powered. I would abort it. And run repairs on individual devices - you will have to wake the device.

Never mind this option is not available for battery powered devices.

3 Likes
17

There are multiple "Repair" buttons on the Z-Wave Details page. Just to make it more fun. :wink:

One big one at the top of the page that runs a repair on the entire Z-Wave Mesh - all devices. You usually don't need to/don't want to use this repair. It used to be a little dangerous (caused problems) but that was resolved by a Z-Wave FW update for the hub so that a full Z-Wave repair is no longer likely to cause problems. Still rarely needed, though.
image

Then (on the newer C7 hubs) there is a "Repair" button for each device, and that runs a repair only on that device.

image
image1510×145 19.8 KB

The "Refresh" button does a refresh of the z-wave detials for that device.

18

If you have a lot of devices it may take a little while, but you should start seeing it go through all the devices in your Z-Wave Details page.

19

As it turns out, only for line-powered devices ...

1 Like
20

Full repair should skip the sleepy devices, right?

2 Likes