EMMC bootloader making problems for downloading software

[palatinusmagnus]

Hello
I am trying to download software for running python scripts and OpenCV.
Although I am very inexperienced when it comes to Ubuntu and this kind of stuff, I notice two errors that I think are connected. Solving these problems might help me actually get opencv running and downloaded.

In short, I see in my errors of trying to download stuff the same phrases thrown around. Below are examples of said phrases and my hardware.

nvidia-l4t-bootloader
nvidia-l4t-xusb-firmware
nvidia-l4t-initrd

sudo apt-get update
sudo apt-get upgrade

"Errors were encountered while processing:

nvidia-l4t-bootloader
nvidia-l4t-xusb-firmware
nvidia-l4t-initrd

E: Sub-process /usr/bin/dpkg returned an error code (1)
"

Specifically
"

dpkg: error processing package nvidia-l4t-xusb-firmware (–configure):
installed nvidia-l4t-xusb-firmware package post-installation script subprocess returned error exit status 1
dpkg: dependency problems prevent configuration of nvidia-l4t-initrd:
nvidia-l4t-initrd depends on nvidia-l4t-xusb-firmware (= 32.5.2-20210709090126); however:
Package nvidia-l4t-xusb-firmware is not configured yet.

dpkg: error processing package nvidia-l4t-initrd (–configure):
dependency problems - leaving unconfigured
No apport report written because the error message indicates its a followup error from a previous failure.
Errors were encountered while processing:
nvidia-l4t-bootloader
nvidia-l4t-xusb-firmware
nvidia-l4t-initrd
E: Sub-process /usr/bin/dpkg returned an error code (1)
fatal: destination path ‘opencv’ already exists and is not an empty directory.
fatal: destination path ‘opencv_contrib’ already exists and is not an empty directory.
mkdir: cannot create directory ‘build’: File exists

"

When trying to use a github script for downloading and building openCV per a youtube video
github JetsonHacksNano/buildOpenCV

When I try running the script from the terminal as advocated, it cuts off after a couple minutes instead of completing hours later.
I see the same phrases about the bootloader and x-usb.

"Configuring incomplete, errors occurred!
See also “/home/jetson/opencv/build/CMakeFiles/CMakeOutput.log”.
See also “/home/jetson/opencv/build/CMakeFiles/CMakeErrors.log”.

real 0m10.697s
user 0m8.368s
sys 0m2.040s

CMake issues
Please check the configuration being used
"

Scrolling up in the logs, I note something

I tried again but reconfigured the script according to the video’s instructions to be better suited for SD card storage Jetsons. Changing the amount of cores being used from 4 to 1 which said would take longer but would work out. Same error happened again. I have attached a part of the verbose logs at the end of this post.

I am naïve and may be glossing over some serious detail that could help. Please let me know what I can supply to paint a better picture. I am sad that blindly following tutorials does not work, but I hope that this can be somehow resolved.
Something of note is that I am using a Jetson Nano using a 16gb EMMC storage instead of booting from a flash drive. I cant figure out how to reflash and reset said EMMC storage since there’s no card to re-burn.
amazon Yahboom-Jetson-Nano-4GB-Board

I would appreciate any tips or advice on how to approach this problem.

Thank you for your time,
Yelizar

Said verbose log from trying to build Opencv

dpkg: error processing package nvidia-l4t-xusb-firmware (–configure):
installed nvidia-l4t-xusb-firmware package post-installation script subprocess returned error exit status 1
dpkg: dependency problems prevent configuration of nvidia-l4t-initrd:
nvidia-l4t-initrd depends on nvidia-l4t-xusb-firmware (= 32.5.2-20210709090126); however:
Package nvidia-l4t-xusb-firmware is not configured yet.

dpkg: error processing package nvidia-l4t-initrd (–configure):
dependency problems - leaving unconfigured
No apport report written because the error message indicates its a followup error from a previous failure.
Errors were encountered while processing:
nvidia-l4t-bootloader
nvidia-l4t-xusb-firmware
nvidia-l4t-initrd
E: Sub-process /usr/bin/dpkg returned an error code (1)
fatal: destination path ‘opencv’ already exists and is not an empty directory.
fatal: destination path ‘opencv_contrib’ already exists and is not an empty directory.
mkdir: cannot create directory ‘build’: File exists

There it is again.
nvidia-l4t-bootloader
nvidia-l4t-xusb-firmware
nvidia-l4t-initrd
What are these things? Did I mess something up when trying to download and set up the Nano for the first time?

[DaneLLL]

Hi,
If you can boot to Ubunut desktop, these prints should be harmless. Do you install the system image through SDKManager or use sdcard image?

[Mt.Hiker]

@palatinusmagnus
Your Nano HW is not an original NVIDIA Jetson Nano devkit,
then do not run ‘sudo apt-get upgrade’.
It may crash your 3rd party carrier board, because this command treat your carrier board as an original Jetson Nano devkit.

Refer here to :
http://www.yahboom.net/study/jetson-nano

[palatinusmagnus]

Hello.

Hmm. I have a system installed but just those little errors. After looking at Mt. Hiker’s response and yours, Ill try out this SDK manager thing.

The system was already pre-installed with onboard storage from the manufacturer. No SD cards.

Thing is about the SDK manager is that it seems to have a limited amount of software packages avaliable. I may be wrong, but I am just trying to get started with basic blob tracking and image processing. Not crazy AI stuff, yolov4, or something like that. Although a neural network might be easily trained to recognize suitable blobs, I prefer a 90’s approach for now. Seems overkill. I guess I am just trying to port as much as possible from a slower raspberry pi to this new hardware.

Ill look more into the SDK manager. Maybe theres some sort of alternative to OpenCV on there.
Thanks
Yelizar

[palatinusmagnus]

Hello.

Thanks for your insight. That kind of makes sense actually. I will investigate that website and start with one of their projects to just get a feel of what’s going on. Since this is kind of different hardware, would most libraries made for regular Jetson’s not work for this 3rd party board? Maybe this is addressed on the Yahboom website, Ill look into it.

I worry that I will be forced to use the tools provided which may be a bit overkill for what I’m trying to start off with.

Thanks again,
Yelizar