[RELEASE] Z-Wave Firmware Updater

Ah, very cool. Thank you for that!

1 Like

anytime

Hi Bryan,
I'm attempting to update 8 Inovelli Red Series Dimmers (LZW31-SN) using the following firmware link.

https://support.inovelli.com/portal/api/kbArticles/407409000002897285/attachments/96fena8aa58942148470eb8cbcee976baa2de/content?portalId=1ce1db55e4825f16146555b24f0f1753cbf5f7d83ca0a3d8b31013b2a2d69e1f

Prior to clicking the "Update Firmware" button, I've been rebooting the dimmer using the air gap. Everything seems to work right up to the point of uploading the firmware to the switch. The below screenshot shows the end state of the update attempt.

Corel
Below is a screenshot of the log containing the entire update attempt.


Any ideas on what I can try to get this to work?

Thanks for any help you can provide and for creating a solution for a much desired capability in Hubitat.

-- Mike

1 Like

This usually happens when the signal is too weak to the end device

Hey, did you need me to try anything else to narrow this down further?

This is what i see in the device information:

  • deviceType: 12801
  • deviceId: 1
  • manufacturer: 29

It seems like firmwareTargets.FirmwareID is empty because numOfTargets is null?

Cheers

1 Like

Sorry .. spaced out on this.. I need to check and see if something happened to the extracted file.. I'll get back to you

2 Likes

Just updated 3 of my Inovelli dimmers painlessly. Working on 4th now. This is truly awesome. Thanks and great work @bcopeland

1 Like

Bryan,

I've attempted updating devices that were within 10 feet of my Hubitat Elevation after successfully performing a Z-Wave repair. I would think that the routing should have been correct and signal strength good, but I am new to Z-Wave so maybe I'm missing something.

I did just successfully update one of the 8 Inovelli Red Series Dimmers and noticed something different about the successful attempt vs. the unsuccessful attempts.

The successful attempt had firmwareFragmentSize set to 40, while all of the unsuccessful attempts had it set to 46. Could this be the problem?

-- Mike

1 Like

I live in an 2-storey townhouse. I've seen z-wave packets go from upstairs to downstairs and back upstairs rather than just go across an open room. So physical proximity doesn't mean they're taking a direct route.

I don't know what the solution is. What I would do is unpair each switch; bring the HE close to it and pair it again, and then try the firmware update.

4 Likes

It’s possible.. But the fragment size is mostly decided by the device.. depending on the command class version..

It’s also possible as @aaiyar pointed out that it’s the mesh routing. You may want to try a z-wave repair and then attempt it again..

If that doesn’t work, I’ll make a debug version that will force the fragment size smaller for you to try

4 Likes

Bryan,

I'm having some trouble with a Zen21 switch.

  • deviceType: 45329
  • inClusters: 0x5E,0x86,0x72,0x5A,0x73,0x85,0x59,0x25,0x27,0x20,0x70
  • firmwareVersion: 20.16
  • deviceId: 7708
  • manufacturer: 634

I suspect I'm trying to go a bridge too far in firmware versions. Zooz gave mew what I think i8s 2.26 : (ZEN21_V304_slave_enhanced_232_OTA_ZW050x_US_WD_027a_b111_1e1c_304_20200226.otz)

And If I read the above correctly, I'm running 2.16.

Every time I try to load it, I get the Sleepy Device message and the update just halts.

Suggestions? @aaiyar I know you updated some of these, any tips?

Thanks!
Scott

1 Like

Sorry, pretty worthless here. None of mine had that issue.

2 Likes

Going to need logs

Thanks @aaiyar. I've seen similar behavior with my mesh WiFi network. I was hoping the repair would produce an optimized route. (FYI, the only LZW31-SN to be successfully updated of the 8 I've attempted is the one that is physically farthest from the HE. Go figure.)

Thanks Bryan. All of the devices are the same model with the same current version number (1.35) reported. Shouldn't the max fragment size be the same for each of them or is it dynamically calculated by each device?

Thanks,
Mike

2 Likes

I can probably do that.

I'm flashing a Zen27 at the moment, and everything is peachy. I presume you'll want me to switch on Debug logging in the Updater?

Scott

1 Like

Thanks Ashok. Given you didn't mention any difficulties other than the Exclude thing, I figured that was the case! Thanks for confirming though!

Scott

2 Likes

The device (depending on command class version) will request a fragment size

1 Like

No need.. It should show in the normal log

Okay. The same device version with the same firmware should request the same fragment size then, correct?

I saw in your code that it looks like if the device does not return a fragment size, it is defaulted to 40. All of the failed attempts for the other Dimmers always showed a fragment size of 46. The only one that succeeded reported a size of 40. Any suggestions on what I should try next?

Thanks,
Mike

1 Like

In theory

Correct.. On earlier command class versions this was not requested by the device

1 Like