SDK Manager fails to download target components, block flash process

Robotics & Edge Computing Jetson & Embedded Systems Jetson Orin Nano
,
1

Host computer: Ubuntu 22.04
SDK Manager: sdkmanager_2.2.0-12028_amd64.deb

Equivalent CLI command: sdkmanager --cli --action install --login-type devzone --product Jetson --target-os Linux --version 6.1 --target JETSON_ORIN_NANO_8GB_DEVKIT --additional-sdk ‘DeepStream 7.1’ --additional-sdk ‘GXF Runtime 4.1’ --select ‘Jetson Linux’ --select ‘Jetson Linux image’ --select ‘Flash Jetson Linux’ --select ‘Jetson Runtime Components’ --select ‘Additional Setups’ --select ‘DateTime Target Setup’ --select Gstreamer --select ‘DLA Compiler’ --select ‘CUDA Runtime’ --select ‘CUDA Runtime’ --select ‘CUDA X-AI Runtime’ --select ‘CuDNN Runtime’ --select ‘TensorRT Runtime’ --select ‘Computer Vision Runtime’ --select ‘OpenCV Runtime’ --select ‘CuPVA Runtime’ --select ‘VPI Runtime’ --select ‘NVIDIA Container Runtime’ --select ‘NVIDIA Container Runtime with Docker integration (Beta)’ --select Multimedia --select ‘Multimedia API’ --select ‘Jetson SDK Components’ --select CUDA --select ‘CUDA Toolkit for L4T’ --select ‘CUDA-X AI’ --select ‘CuDNN on Target’ --select ‘TensorRT on Target’ --select ‘Computer Vision’ --select OpenCV --select ‘VPI on Target’ --select ‘Developer Tools’ --select ‘Nsight Systems’ --select ‘Nsight Graphics’ --select DeepStream --select DeepStream --select GXF --select ‘GXF Runtime’ --deselect ‘Jetson Platform Services - Coming Soon’ --deselect ‘Jetson Platform Services - Coming Soon’ --flash --license accept

Has tried following options, none of those works:

  1. re-start ubuntu
  2. re-login nvidia account
  3. sudo apt reinstall sdkmanager
  4. remove ~/.nvsdkm/ and ~/nvidia/ and ~/Download/nvidia/ and /etc/exports
  5. using sdkmanager --cli
  6. using docker solution.

image
image1400×900 198 KB

image
image1400×900 210 KB

sdkm-2024-12-30-15-25-02.log (2.5 MB)
sdkm_download.log (850.5 KB)

1 Like
2

Confirmed - I get exactly same error. I am trying to setup my new Jetson Orin Nano (Super) using Ubuntu 22.04 LTS (I can’t use my latest Debian installation - so I first had to create a virtual machine)

Screenshot from 2024-12-31 01-11-44
Screenshot from 2024-12-31 01-11-441920×1048 110 KB

4

Add me to the club as well. Physical and Virtual UB 22.04 LTS.

5

Me as well, Physical UB 22.04 LTS

6

Hi,

Press EXPORT LOGS and attach the zip file for us to review.
Also some questions to confirm:

  • Is Orin nano a devkit or custom carrier board?
  • Is Ubuntu running on a physical machine or virtual machine (e.g. WSL, VirtualBox, VMware)?

Thanks

1 Like
7

Same here, Physical UB 22.04 LTS

1 Like
8

Hello, same problem too. Find attached exported logs
SDKM_logs_JetPack_6.1_(rev._1)_Linux_for_Jetson_Orin_Nano_8GB_2024-12-31_11-22-04.zip (164.1 KB)

I tried to just download files without installation, same issue

1 Like
9

There is server issue. We are working on fixing it.

6 Likes
10

Is there an ETA?

11

Same issue (

12

same here…

13

same exact problem. is there not a way to get the sample root fs / BSP and flash from cli?

14

everyone try again. They may have fixed things

2 Likes
16

Confirm the issue of downloading failure has been resolved. It’s indeed a backend problem of Nvidia SDK Manager. Thanks everyone for sharing and confirming the experience.
#resolved

17

I am facing similar problem with CUDA runtine target in Orin Nano 8GB module. SDK Manager tool stuck at 66% and after retrying 4th time installation finish unsuccessful. Is it same developer server issue or something else ?.

SDKM Logs:
17:18:11 : CUDA Runtime - target: download https://developer.nvidia.com/assets/embedded/secure/tools/files/jetpack-sdks/jetpack-6.1/Jetson_61_b218/./l4t-cuda-tegra-repo-ubuntu2204-12-6-local_12.6.11-1_arm64.deb failed, retrying 4…
17:18:43 INFO: CUDA Runtime - target: start to download https://developer.nvidia.com/assets/embedded/secure/tools/files/jetpack-sdks/jetpack-6.1/Jetson_61_b218/./l4t-cuda-tegra-repo-ubuntu2204-12-6-local_12.6.11-1_arm64.deb to /home/jcreyes/Downloads/nvidia/sdkm_downloads/l4t-cuda-tegra-repo-ubuntu2204-12-6-local_12.6.11-1_arm64.deb
17:21:44 ERROR: CUDA Runtime - target: Download request timed out after 60 seconds. Closing connections
17:21:44 ERROR: CUDA Runtime - target: Download ‘CUDA Runtime’ failure
17:21:44 ERROR: CUDA Runtime - target: Download request timed out after 60 seconds. Closing connections
17:21:44 ERROR: CUDA Runtime - target: Download ‘CUDA Runtime’ failure
17:08:53 INFO: Gstreamer - target: verifying checksum of /home/jcreyes/Downloads/nvidia/sdkm_downloads/nvidia-l4t-gstreamer_36.4.0-20240912212859_arm64.deb
17:21:44 ERROR: CUDA Runtime - target: download failed
17:08:53 INFO: DLA Compiler - target: verifying checksum of /home/jcreyes/Downloads/nvidia/sdkm_downloads/nvidia-l4t-dla-compiler_36.4.0-20240912212859_arm64.deb
17:21:44 SUMMARY: CUDA Runtime - target: First Error: download failed
17:21:44 ERROR: CUDA Runtime - target: Download ‘CUDA Runtime’ failure
17:08:53 INFO: CuDNN Runtime - target: verifying checksum of /home/jcreyes/Downloads/nvidia/sdkm_downloads/cudnn-local-tegra-repo-ubuntu2204-9.3.0_1.0-1_arm64.deb
17:21:44 ERROR: CUDA Runtime - target: download failed

18

Keep trying. Also after it downloads. Cancel and go back in and I bet it will show its downloaded.

19

I had to run flash.sh manually - but the power mode of MAXN doesn’t show up. So I suspect this might be the root of my problem.

It downloads everything - I did select the orin nano 8GB DEV kit (that’s what I ordered) but it also gave the option of a standard 8GB kit as an option. Both did not work and failed after around 60%. Running flash.sh manually was the only way around it.

Also, I had to flash an NVMe after the flash of the nano was done with an SD card. I could not get it to work with the NVMe directly. However, flashing it did work but it doesn’t show the MAXN.

20

@krickert

Could you share this result from your board?

cat /etc/nv_boot_control.conf
and
cat /etc/nv_tegra_release

21 
un@machine:~$ cat /etc/nv_boot_control.conf
TNSPEC 3767-300-0005-R.1-1-0-jetson-orin-nano-devkit-super-
COMPATIBLE_SPEC 3767--0005--1--jetson-orin-nano-devkit-super-
TEGRA_BOOT_STORAGE nvme0n1
TEGRA_CHIPID 0x23
TEGRA_OTA_BOOT_DEVICE /dev/mtdblock0
TEGRA_OTA_GPT_DEVICE /dev/mtdblock0
un@machine:~$ cat /etc/nv_tegra_release
# R36 (release), REVISION: 4.2, GCID: 38685322, BOARD: generic, EABI: aarch64, DATE: Fri Dec 13 00:16:27 UTC 2024
# KERNEL_VARIANT: oot
TARGET_USERSPACE_LIB_DIR=nvidia
TARGET_USERSPACE_LIB_DIR_PATH=usr/lib/aarch64-linux-gnu/nvidia
un@machine:~$
22

Looks like it is correctly flashed. Please file a separately ticket and we could check your issue there.
As this topic is specific to flash issue only.