Unable to boot ZED Box Orin NX | NVIDIA logo ->black screen-> continuously blinking _

Hardware ZED Box
1

Hello,

I have a ZED Box Orin NX 16GB running JetPack 6.2 / L4T 36.4.3.

The device stopped booting. The current behavior during boot is:

NVIDIA logo → black screen → continuously blinking _

During this blinking stage, the power LED remains green. I have tried to do ssh, but it did not work.

I have performed the troubleshooting/recovery procedures described in your documentation My observations are:

  • UEFI is accessible and functioning.

  • The Orin NX 16GB and 16197 MB RAM are detected in UEFI.

  • The internal NVMe SSD is detected by UEFI.

  • I created the official JetPack 6.2 recovery USB using zedbox_all_usb_flash_6.2.sh.

  • The recovery USB is detected by UEFI.

  • Running EFI\BOOT\BOOTAA64.efi manually starts and it displays the following:

    L4TLauncher: Attempting Recovery Boot

    EFI stub: Booting Linux Kernel…

    EFI stub: Using DTB from configuration table

    EFI stub: Loaded initrd from LINUX_EFI_INITRD_MEDIA_GUID device path

    EFI stub: Exiting boot services…

The system then hangs with a blinking _.

I have also tried the live-USB Force Recovery procedure using the USB-A port.

However: lsusb -d ‘0955:’ , never detects the expected Orin NX 16GB recovery device: 0955:7323.

The Micro-USB OTG port also does not enumerate as /dev/ttyACM0 on an Ubuntu 22.04 host. The same host USB port successfully detects other USB devices.

I also noticed that the UEFI RTC had reset to: 01/01/1970

I corrected the date/time, but the boot problem remained.

The device also becomes noticeably warm during these recovery attempts.

The most important point: I have important data on the internal NVMe and do not want to erase/reflash it before attempting data recovery.

Could you please advise whether this indicates a hardware/bootloader issue and provide the recommended procedure for backing up the internal NVMe data before any reflash/recovery? How should I get the ZED Box working again?

I can provide photographs of the UEFI screens, recovery boot output, serial/USB tests, and the recovery USB configuration if required.

Thank you.

shared image (3)
shared image (3)2000×1500 261 KB

Media (20)
Media (20)1080×479 20.6 KB

2

Hi,

Your data on the NVMe should be safe: nothing below erases or reflashes it. This looks like an issue we’ve seen recently. A Linux package update can break one of the drivers our board relies on, so it’s unlikely to be a hardware problem.

  1. Press Esc repeatedly at the NVIDIA splash screen to open UEFI setup, then go to Device Manager > NVIDIA Configuration > L4T Configuration. Set L4T Boot Mode to Kernel Partition, check that OS chain A status is Normal, then F10 and Y.

  1. Reboot. The box will start in a degraded mode (no camera drivers). If a first boot setup screen shows up, just create a temporary user.

  1. You should be able to read your NVMe from there, so feel free to copy anything important to a USB drive first.

  1. Run sudo dpkg --configure -a

  1. Reboot, go back to L4T Configuration and set L4T Boot Mode back to ExtLinux (F10, Y).

If it still shows a black screen after that, let me know and I’ll send you the next step.

Best,

Tristan Leduc


StereoLabs Support