Unable to Form Network with znp_CC1352P_2_LAUNCHXL_tirtos_ticlang_HFC.hex about zigbee-3.0-usb-dongle-plus-firmware HOT 8 OPEN

This firmware has an issue that I and others have experienced. When z2m is restarted, it is often necessary to unplug and re-plug the USB dongle, before restarting z2m. Otherwise, all these weird errors appear. I don't know if that's related to the HFC modification or not. So, it could be that you didn't have to rebuild the firmware and a re-plug would have done the trick. See if your version works better and doesn't require the re-plug, I might have overlooked something.

from zigbee-3.0-usb-dongle-plus-firmware.

I flashed your file and put it right into the rPi to spin up a new instance. I flashed mine and followed the same steps. Mine came right up. I don't think its anything you overlooked. The only thing I can think of, is that I had to manually edit the znp.syscfg because I couldn't apply koenkk's patch.

* @versions {"tool":"1.12.0+2376"}

The patch is looking for

* @versions {"tool":"1.12.0+2406"}

That was the only deviation from the steps

I'm going to do some more testing and see how things go. I'm having adapter crashing every couple days on a large network, so I am hopeful that flow control may help.

from zigbee-3.0-usb-dongle-plus-firmware.

I successfully tested my build for 24 hours on my small 3 device test network without any issues. I proceeded to flash both of my production adapters with the build and both have been running for 16 hours without any issue so far.

I did not have any issue with the adapters not coming up after restarting Z2M several times. I did see similar phenomenon using 20220219 where I would have to stop and restart the container. Interestingly, I ran a checksum from my build and yours and both have different MD5 signatures.

My build is posted here along with a couple steps I had to deviate from, if you wish to test.

from zigbee-3.0-usb-dongle-plus-firmware.

Sounds good. I definitely will, since having to unplug the USB is a pain. BTW, are you sure we're talking about the same development version of Koenk's firmware? The one I used is referencing @versions {"tool":"1.14.0+2667"} and not * @versions {"tool":"1.12.0+2376"}

from zigbee-3.0-usb-dongle-plus-firmware.

That might be the difference. I'm building 6.10.01 for Hardware Flow Control.. It seems to be quite stable.
The one you are referencing is 6.30.00.84, and it uses the new UART2 driver and rtos7. I suspect that might be one of issues with the build.

FWIW.. There's a newer 6.40 SDK out there. The patch doesn't seem to want to apply to it. I may have to fiddle.

from zigbee-3.0-usb-dongle-plus-firmware.

I just built and tested SDK 6.40.00.13 both with and without hardware flow control. You can find it in the dev branch in my repo. I was able to flash my dev stick from 6.10 to 6.40 and have it startup without any need to unplug the stick, and was able to use the existing network. I was also able to form a new network without any NIB errors with this build. Going to run it for a day or to in my lab mesh then will try to deploy on one of my production meshes.

from zigbee-3.0-usb-dongle-plus-firmware.

Perfect. Hopefully you can take over future builds from me as I have a very old PC and limited time.

from zigbee-3.0-usb-dongle-plus-firmware.

That composer studio is a pig. It's slow as molasses on my 2021 MacBook Pro. I can only imagine. Time is not on my side either, although I'm happy too push a new build out now and then. Let me know how your testing goes.

from zigbee-3.0-usb-dongle-plus-firmware.