@Alberto-Brosich These have not been ported 22 or 23, but you can find the source in the 18.6 branch. Depending on your requirements, you may need to modify the code. I need to migrate oniondebug myself so will do that in the next day or so, since I need it for powerdock. Some of these libraries may be dependent on the OpenWrt 18 GPIO numbering which changed in OpenWrt22. I think @luz made patch for this, check his profile for posts. Personally I'm just adopting the new numbering system.

@DocHardinger Cool project! Home automation for your home on wheels I automated some of my apartment using Omega2, my existing home automation system was missing some features so I added them. I can't really answer your question since every case is different, but I prefer to be running the most current release and since there is not much difference between 22 and 23 it's fairly trivial to switch over to 23. Since 23 is the current release, all the new stuff and improvements will be on 23. For me it just came down to testing time before I could put it into production.

@mohamedkallel the wifi radio calibration stored in the factory partition is unique to each device. It's done in the factory as part of QA testing with the help of an RF calibration machine. We do not recommend changing these values. Hence we do not recommend using a flash programmer to write complete images to the flash.

@DocHardinger Both SPI and onewire are still in the pipeline. We're currently working on making device tree overlays loadable during run-time. More info on the progress of that here: https://github.com/OnionIoT/OpenWRT-Packages/issues/62 We'll need the device tree overlay functionality to make one-wire work - as a lot of the "custom" gpio modules were removed from the linux kernel. @luz posted above in this thread with more details. TL;DR: we're still working on improving the beta firmware. Expect SPI and one-wire to be completed in the coming weeks.

@Ilya-Mordasov said in No wireless: yes, that seems to be the problem, because only the wireless is not working, I tried i2c and spidev, everything is fine. Yeah, only the wifi driver relies on the data in the factory partition. So it makes sense that everything else works. Do I only need to know the factory MAC address or do I need to have access to the whole chip? My guess is that you'll need a whole new copy of the factory partition - it looks like the wifi driver in openwrt 22.03 has a workaround for invalid MAC addresses but still crashes because of the rest of the invalid configuration. Do I need to connect wires to the chip? No need to connect wires, you can do it from Linux. And how can I extract the factory partition from the live chip? To copy the factory partition from one device to another, it's actually super similar to the steps in the post linked by @crispyoz: @Lazar-Demin said in Contradiction in Omega2 Device MAC Address Allocation: install the xxd package dump the factory partition to hex: xxd /dev/mtd2 > factory.hex create a backup just in case edit the two mac addresses (can use vi) convert the edited hex file to binary: xxd -r factory-new.hex > factory.bin write the binary file to the factory partition: mtd -r write factory.bin factory Do steps 1 to 5 on a working device, and step 6 on your current device. (You'll probably need to reinstall whichever firmware was successfully booting on your current device in order to get to Linux)

@Lazar-Demin It was in the 22 Beta I was playing with on that day. As I recall I had downloaded the firmware on the day before, so whatever revision was current on 17th May. I've moved on to 23.05 now so no longer have the 22 stuff to check. Next time I'll remember to include the version details. Sorry for the newbie omission.

Yes, it worked. Thank you. I had to use different pin numbering as @luz explained in this answer for my CAN bus question.