Jetson Orin Nano – Repeated EXT4/NVMe filesystem corruption and “Attempting recovery boot” on JetPack 7.2

Robotics & Edge Computing Jetson Systems Jetson Orin Nano
1

Hi,

we are using several NVIDIA Jetson Orin Nano systems with NVMe SSDs running JetPack 7.2. We are experiencing recurring filesystem corruption, sometimes leaving the system unable to boot.

The following errors appear during boot:

JBD2: Invalid checksum recovering data block
JBD2: journal recovery failed
EXT4-fs (nvme0n1p1): error loading journal
mount: /mnt: cannot mount: probably corrupted filesystem on /dev/nvme0n1p1
nvme nvme0: I/O tag timeout

This has happened more than once, so it does not appear to be an isolated corrupted image.

We also performed a cross-test with another working Jetson Orin Nano:

  • An NVMe SSD that boots correctly in another Jetson does not boot in the affected Jetson.
  • On the affected Jetson, it reaches “Attempting recovery boot” and then gets stuck.
  • The same SSD works correctly when moved back to the other Jetson.

This makes us suspect that the issue may not be caused by the NVMe SSD itself, but possibly by the Jetson hardware, firmware/UEFI, QSPI bootloader, or the NVMe/PCIe interface.

Has anyone experienced similar recurring EXT4/NVMe corruption on Jetson Orin Nano with JetPack 7.2?

Is there any good way to debug such errors or get to related debug files on the disk?

3

There is no update from you for a period, assuming this is not an issue anymore.
Hence, we are closing this topic. If need further support, please open a new one.
Thanks ~0826

  1. Is this issue reproducible on NV devkit? If so, how to do it?

  2. Which PCIe controller is in use for the NVMe? Is that C4 or C7?

  3. What would happen if you follow our FAQ to recover from recovery boot situation? Will the system make the next boot after doing that?