Failed to apt-get install nvidia-jetpack on TX2

hello fangbin,

could you please share the boot-up logs for reference,
thanks

data0405a.txt (34.3 KB)

It looks like your last log is entirely removing items. What Iā€™ll suggest for logging is to fully boot, run your serial console, clear any console logging up until then, and start logging. Following this, for the purpose of getting logs:

# The "-s`" causes you to stay sudo until you "exit":
sudo -s
apt update
apt-get upgrade
# Keep in mind that trying to install a package already installed is not a problem. apt will just confirm
# that the package is already there.
apt-get install nvidia-cuda nvidia-opencv
apt-get install nvidia-l4t-jetson-multimedia-api
# To verify networking (only needed if packages fail to install):
ifconfig
route
# The serial console log should contain what is needed. If it reboots or hangs, then you can continue
# logging till reboot is done, and submit that as well as a single log file. If the system itself does not crash, then no need to log reboot.

Not sure Iā€™ve correctly understood your case, sorry if misunderstood.

In my case, running on Xavier NX, I donā€™t have nvidia-jetpack installed, but I see:

sudo apt search nvidia | grep installed

WARNING: apt does not have a stable CLI interface. Use with caution in scripts.

cuda-gdb-10-2/unknown,stable,now 10.2.89-1 arm64 [installed,automatic]
cuda-libraries-10-2/unknown,stable,now 10.2.89-1 arm64 [installed,automatic]
cuda-libraries-dev-10-2/unknown,stable,now 10.2.89-1 arm64 [installed,automatic]
cuda-nvtx-10-2/unknown,stable,now 10.2.89-1 arm64 [installed,automatic]
deepstream-5.0/stable,now 5.0.1-1 arm64 [installed]
libnvidia-container-tools/stable,now 0.9.0~beta.1 arm64 [installed]
libnvidia-container0/stable,now 0.9.0~beta.1 arm64 [installed]
libnvvpi1/stable,now 1.0.15 arm64 [installed]
libvdpau1/bionic,now 1.1.1-3ubuntu1 arm64 [installed,automatic]
libvisionworks/stable,unknown,now 1.6.0.501 arm64 [installed]
libvisionworks-dev/stable,unknown,now 1.6.0.501 all [installed]
libvisionworks-repo/now 1.6.0.501 arm64 [installed,local]
libvisionworks-samples/stable,unknown,now 1.6.0.501 arm64 [installed]
libvisionworks-sfm/stable,unknown,now 0.90.4.501 arm64 [installed]
libvisionworks-sfm-dev/stable,unknown,now 0.90.4.501 arm64 [installed]
libvisionworks-sfm-repo/now 0.90.4.501 arm64 [installed,local]
libvisionworks-tracking/stable,unknown,now 0.88.2.501 arm64 [installed]
libvisionworks-tracking-dev/stable,unknown,now 0.88.2.501 arm64 [installed]
libvisionworks-tracking-repo/now 0.88.2.501 arm64 [installed,local]
libxnvctrl0/bionic-updates,now 460.39-0ubuntu0.18.04.2 arm64 [installed,automatic]
nvidia-container-csv-cuda/stable,now 10.2.89-1 arm64 [installed]
nvidia-container-csv-cudnn/stable,now 8.0.0.180-1+cuda10.2 arm64 [installed]
nvidia-container-csv-tensorrt/stable,now 7.1.3.0-1+cuda10.2 arm64 [installed]
nvidia-container-csv-visionworks/stable,now 1.6.0.501 arm64 [installed]
nvidia-container-runtime/stable,now 3.1.0-1 arm64 [installed]
nvidia-container-toolkit/stable,now 1.0.1-1 arm64 [installed]
nvidia-docker2/stable,now 2.2.0-1 all [installed]
nvidia-l4t-3d-core/stable,now 32.5.1-20210219084708 arm64 [installed]
nvidia-l4t-apt-source/stable,now 32.5.1-20210219084708 arm64 [installed]
nvidia-l4t-bootloader/stable,now 32.5.1-20210302195856 arm64 [installed]
nvidia-l4t-camera/stable,now 32.5.1-20210219084708 arm64 [installed]
nvidia-l4t-configs/stable,now 32.5.1-20210219084708 arm64 [installed]
nvidia-l4t-core/stable,now 32.5.1-20210219084708 arm64 [installed]
nvidia-l4t-cuda/stable,now 32.5.1-20210219084708 arm64 [installed]
nvidia-l4t-firmware/stable,now 32.5.1-20210219084708 arm64 [installed]
nvidia-l4t-graphics-demos/stable,now 32.5.1-20210219084708 arm64 [installed]
nvidia-l4t-gstreamer/stable,now 32.5.1-20210219084708 arm64 [installed]
nvidia-l4t-init/stable,now 32.5.1-20210219084708 arm64 [installed]
nvidia-l4t-initrd/stable,now 32.5.1-20210302195856 arm64 [installed]
nvidia-l4t-jetson-io/stable,now 32.5.1-20210219084708 arm64 [installed]
nvidia-l4t-jetson-multimedia-api/stable,now 32.5.1-20210219084708 arm64 [installed]
nvidia-l4t-kernel/stable,now 4.9.201-tegra-32.5.1-20210302195856 arm64 [installed]
nvidia-l4t-kernel-dtbs/stable,now 4.9.201-tegra-32.5.1-20210302195856 arm64 [installed]
nvidia-l4t-kernel-headers/stable,now 4.9.201-tegra-32.5.1-20210302195856 arm64 [installed]
nvidia-l4t-multimedia/stable,now 32.5.1-20210219084708 arm64 [installed]
nvidia-l4t-multimedia-utils/stable,now 32.5.1-20210219084708 arm64 [installed]
nvidia-l4t-oem-config/stable,now 32.5.1-20210219084708 arm64 [installed]
nvidia-l4t-tools/stable,now 32.5.1-20210302195856 arm64 [installed]
nvidia-l4t-wayland/stable,now 32.5.1-20210219084708 arm64 [installed]
nvidia-l4t-weston/stable,now 32.5.1-20210219084708 arm64 [installed]
nvidia-l4t-x11/stable,now 32.5.1-20210219084708 arm64 [installed]
nvidia-l4t-xusb-firmware/stable,now 32.5.1-20210302195856 arm64 [installed]
nvidia-settings/bionic-updates,now 460.39-0ubuntu0.18.04.2 arm64 [installed]
ubuntu-drivers-common/bionic-updates,now 1:0.8.6.3~0.18.04.1 arm64 [installed]
vdpau-driver-all/bionic,now 1.1.1-3ubuntu1 arm64 [installed,automatic]
vpi1-demos/stable,now 1.0.15 arm64 [installed]

So it seems you may first get l4t versions rather than desktop versions of nvidia and CUDA packages.
For opencv, the version shipped in jetpack provides gstreamer support but not CUDA. For this you may build your own using scripts such as JEP/install_opencv4.5.0_Jetson.sh at master Ā· AastaNV/JEP Ā· GitHub or GitHub - mdegans/nano_build_opencv: Build OpenCV on Nvidia Jetson Nano.

Hi JerryChang,

Iā€™m working on R32.5.1 TX2 and have met the same issue after running ā€œsudo apt install nvidia-jetpackā€.
I canā€™t login to ubuntu after install nvidia-jetpack.
Do you have any solution for this issue?

Thx
Yen

Did you run that package install directly from the Jetson? My thought is that if you did, then failure to log in would likely be from the filesystem filling up. In that case, if you can log in via serial console, then you could delete some files. If not, then if you donā€™t want to lose anything, you might clone, remove some files from a host PC with a loopback mount of the clone, and then flash the repaired clone back in (assuming it really is just a full disk and not some other issue). Can you log in via serial console, or at least get a boot log from serial?

Hi,

Sorry, I used the wrong version of rootfs, so it canā€™t login.
My problem is solved.

Thx
Yen