Hi all, sorry for such a noob post, but I really hope any help I get can help others you might need it too.
I’m following the web frame guide but, after the unicorn, I’m still just seeing “check your balenaCloud dashboard”. I’m guessing I’m supposed to be seeing a rickroll video on youtube.
I’ve tried two different Raspi’s (an original B, and a zero with a USB WiFi stick), reflashing the SD card, and multiple attempts, and I just don’t understand why I’m not even getting error messages (unless there’s something in the boot sequence I’m not seeing). I’ve run the balena cli from a windows machine and a macbook, no difference.
I’ve granted support access, so if anyone can help me it’d be amazing, and I will happily explain the problem/solution(s)!
I had a look at your device, it seems that the user container never manages to start up correctly and is stuck in restart loop. Could you try setting the RESIN_HOST_CONFIG_gpu_mem to 396 from the dashboard as mentioned in the Configuration section?
We’ve got this device booting again for you. The issue seems to be that because you’re using a RPi 1 and/or a Zero, there’s just not enough memory. In the guide it says to use a Pi 3B, so when it says allocate 396MB to the GPU there’s still 628MB left for the OS and everything else.
The issue with your device was that because 396MB was allocated to the GPU on a device with 512MB RAM, only 116MB was left for the OS and other services - they then couldn’t start up meaning your subsequent setting of 256MB could not apply.
We’ve resolved the issue by manually editing /mnt/boot/config.txt with another memory value that then allowed the supervisor to start and apply the setting of 256MB. Unfortunately though this isn’t enough to allow WPE to start up. You really need a device with 1GB RAM for this project. The Pi 2 might work, but we’ve not tested it. Pi 3B and 3B+ are definitely OK.