harleyqmc
Shame to get hung up on the most basic of steps here but I am struggling to find the solution last few nights
Started off with
$ export MIX_TARGET=rpi0
$ mix deps.get
$ mix firmware
$ mix burn
Which all complete without error but the Pi does not boot up. That lead me to this thread
Where @fhunleth suggested rpi3a
But trying export MIX_TARGET=rpi3a results in same hang with the green LED goes off once then solid green
Then tried the rpi0_2 option also in the same reply
And that fails with a different LED output where it flashes once and then goes DED
I think that is a “card read error” but I do not think it can be that as I did manage to get the Pi to boot with Raspberry Pi OS and on two separate SD cards
So I am not sure if I am still burning wrong MIX_TARGET or missing some simple step or what
I also tried these but they just hung with solid green LED also
https://github.com/elixir-circuits/circuits_quickstart/releases/download/v0.7.1/circuits_quickstart_rpi3a.fw
https://github.com/elixir-circuits/circuits_quickstart/releases/download/v0.7.1/circuits_quickstart_rpi0.fw
Any suggestions would be great
Thanks
Trending in Questions
Other Trending Topics
Categories:
Sub Categories:
Forums
Popular Tags
- #ecto
- #liveview
- #troubleshooting
- #learning-elixir
- #deployment
- #library
- #erlang
- #testing
- #genserver
- #mix
- #absinthe
- #remote-other
- #otp
- #plug
- #how-to-question
- #macros
- #postgres
- #elixirconf
- #channels
- #exunit
- #discussion
- #code-sync
- #podcasts
- #javascript
- #onsite
- #dialyzer
- #docker
- #authentication
- #umbrella
- #full-time-contract
- #podcasts-by-brainlid
- #ecto-query
- #blog-post
- #elixirconf-us
- #elixir-ls
- #ai
- #phoenix_html
- #iex
- #graphql
- #genstage
- #websockets
- #supervisor
- #advent-of-code
- #distillery
- #processes
- #api
- #forms
- #hex
- #security
- #metaprogramming










Showing Posts 1 to 10- Show Best Posts
- Show All (oldest first)
- Show All (newest first)
jjcarstens
What have you tried after seeing the green LED? I think there might be some misconception here:
nerves.localor see that the device is connected over network if you are using the middle USB port of the rpi0 or USB A port of rpi3aCan you take a look over Connecting to a Nerves target and try some of the suggestions there to see if maybe the device is actually working properly?
Or if needed, you can configure WiFi when burning the circuits quickstart image to an SD card. You could try that as well and see if the new device joins your network
harleyqmc
Dam thanks for quick reply.
So doing the circuits quickstart image just now
popped it in the Pi and the green LED did exaclty this
So this probably is not expected to work
It is not listed in the WiFi devices dashboard
Other notes I do not really have access to a USB to TTL cable for serial connection at the moment and it is definitely plugged into middle USB port and not “pwr in” one
Am I missing something glaringly obvious here?
jjcarstens
Are you able to confirm that your USB cable supports data and power? It’s really common now to have power only USB cables (especially if from a cheap bin or off the shelf grab) - so much so that I invested in a USB tester just for this purpose.
Trying several different USB cables might be the next step. You can also enable the HDMI output temporarily if you have an adapter
harleyqmc
Can not really confirm I guess… Though I definitely agree with the second part because I simply can not find a way to easily identify one from the other visually or by sense of smell
However, I have tried three separate cables ( pretty much all I could find around the house ) and one of which actually transfers images from an old android phone so I have to believe that one is at least data sensitive
In fact… I am basically only working through
And they recommend a ( seemingly ) data USB capable cable in the notes that I bought specifically to avoid lame ass problems like this
Because I just want to tool around with nerves and IoT.
I am even questioning whether or not I bought or received a Pi zero 2 W at this point but it looks like it is. It is printed on the board
I did do this with
rpi0and it hung at the rainbow splash screen. Obviously the wrong MIX_TARGETHow would you recommend going about this HDMI output otherwise?
Still use the circuit quickstart image?
harleyqmc
An update…
In config.txt I set
disable_splash=0
enable_uart=0
For both
export MIX_TARGET=rpi3aand$ sudo fwup circuits_quickstart_rpi3a.fwAnd both just keep rebooting over and over.
Yea so no idea
There a debug mode?
jjcarstens
What host OS are you using to build and burn firmware?
I just tried the
rpi3afirmware from your original post and it successfully booted. Once running, the green LED turns completely off.I also did not have success using
exportfor the WiFi settings andsudoburn task. It needed to be within the same command and I needed to force the WiFi settingsTrying that again with
NERVES_WIFI_FORCEis worth a try. Depending on your host, you might need a driver installed to see a device over OTG.Investing in a simple USB TTY serial cable is also very useful for these smaller devices. If you want the output on the HDMI, then you’ll need to enable it in erlinit.config instead of the UART pins
Chris660
Not saying this is the case, but that could be a power supply issue.
How are you powering the board?
The “official” rpi zero-2 W PSUs are rated at 5V 2.5A.
RemyXRenard
I was having a similar problem with connecting to a raspberry pi with a Nerves project.
Config:
Raspberry Pi Zero 2 W
Windows 11 running WSL2
MIX_TARGET = rpi3a
Using the instructions from the “Build a Weather Station with Elixir and Nerves”
The main issue was that I was unable to ping or ssh to the device over USB. I tried multiple USB cables, confirmed I was using the gadget port (not the power only port), tried another firmware file (circuits_quickstart_rpi3a.fw) from GitHub - elixir-circuits/circuits_quickstart: Try out Elixir Circuits on Nerves! · GitHub, all unsuccessfully.
I also tried several attempts at getting mDNS/zeroconf/Bonjour working on both WSL2 and Windows 11, which I thought might be part of the problem. I learned some things, but didn’t solve the problem.
One solution proposed here seemed promising, which was to set the NERVES_WIFI_SSID and _PASSWORD as env variables or pass them as arguments to fwup. However, due to how WSL2 works, with limited access to MicroSD cards,
mix burnpasses this off to fwup.exe on the Windows side, so the arguments and env vars are not honored, as far as I can tell.My solution was to configure WiFi in advance, before creating the firmware file, by updating config/target.exs in the project. By adding my wireless ssid and psk to the wlan0 section, the RPi pulled an address through dhcp and I could ssh/ping it over WiFi.
This is basically the instructions from the book mentioned above and the VintagetNetDirect site (GitHub - nerves-networking/vintage_net_wifi: WiFi networking for VintageNet · GitHub), doing it before you burn the firmware.
To be more explicit, assuming you have the “sensor_hub” project from the book all set up, edit the config/target.exs file ensuring it has a section that looks like this (there might be other interfaces in there, like usb0 and eth0, that’s fine:
Then run
mix firmwareandmix burnas usual. Swap the SD card to the pi. In my case it has a steady green LED. I checked my router for devices pulling addresses and after a tense few moments, “nerves-8821” showed up on my network, and was able to ssh to the device.I hope that helps someone else.
tyoc213
Hi there, I’m trying to do the same but without the sd card as instructed here GitHub - nerves-livebook/nerves_livebook: Develop on embedded devices with Livebook and Nerves · GitHub the last line of the log is
fwup: Expecting platform=rpi3a and architecture=armI have tried with
rpi3aandrpi0same result.This is what I get from running
ssh nerves.localand executingdmesgfhunleth
Could you
sshinto the device and either look at the message of the day or runNerves.Runtime.KV.get_active("nerves_platform"):That will tell you which image matches what’s running on your device.
If this doesn’t point out why the firmware update platform check failed, then please post what you see and I’ll try to think about how this could have happened.