Jetson Orin Nano Detection Issue

I have a Jetson orin nano which I can successfully flash using SDK Manager running on a Ubuntu 20 install (I can do this either to SDCard or NVME drive).

SDK manager detects the Jetson instantly when it is in forced recovery. Unfortunately I am having problems flashing the device from the command line as some of the script based checks don’t detect the device.

I have looked at the ‘lsusb’ results which correctly show the device connected on the host machine.

Bus 004 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
Bus 003 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 002 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
Bus 001 Device 004: ID 045e:00cb Microsoft Corp. Basic Optical Mouse v2.0
Bus 001 Device 007: ID 0955:7523 NVIDIA Corp. APX
Bus 001 Device 002: ID 1c4f:0015 SiGma Micro USB Keyboard
Bus 001 Device 005: ID 8087:0032 Intel Corp.
Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub

I am using the second to latest firmware 5.1.3 as opposed to 6. When I execute nvautoflash.sh along with the flag to show the board ID it states 0 devices connected.

*** Checking ONLINE mode … OK.
*** Checking target board connection … 0 connections found.
*** Error: No Jetson device found.

I am slightly confused why the Jetson is detected by SDK manager but not via the Nvidia scripts.

Can someone please advise what else I could try to progress with command line based flashing please.

What exact command did you run?
Put the full log here. Don’t crop it.

Hi Dave,

So the script I am running is basically doing the equivalent of downloading the two zip files and extracting them. So you end up with a folder called Linux_for_tegra which inside has all the config files and the example rootfs folder. I am navigating to the extracted /Linux_for_tegra folder and am executing

./nvautoflash.sh —print_boardid
I have also tried
sudo ./nvautoflash.sh —print_boardid
I have also tried
./nvautoflash.sh

They all produce the same output

dan@dan-System-Product-Name://install/Linux_for_Tegra$ sudo ./nvautoflash.sh --print_boardid
*** Checking ONLINE mode … OK.
*** Checking target board connection … 0 connections found.
*** Error: No Jetson device found.

The board has been started in forced recovery mode and is connected via a USB C connection using the same cable used for the SDK Manager approach.

Follow all the steps here:
https://docs.nvidia.com/jetson/archives/r36.3/DeveloperGuide/IN/QuickStart.html

Yep so following those steps it worked and the device flashed successfully via the command line. Does that mean using nvautoflash.sh is an unreliable means of testing connections ?

A friend of mine has the same device (although they were using virtualisation for the Ubuntu host side vs a native Ubuntu box like I am) and it was seemingly reporting a connection.

Is there any strong reason you have to rely on nvautoflash.sh?
I don’t think it’s documented anywhere in the devloper guide.
lsusb should be enough.

No I don’t think so, I think someone had included it as a check. However if it’s been superseded or is old code no longer referenced I’ll suggest they remove it.

Thanks for your help.

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.