Hello NV Teams
L4t:36.3 agx orin Devkit
When I was performing the backup and restoration of the system, I used the following two methods
1:According to the instructions in 1.txt, there were no problems at all.
2:According to the instructions in 2.txt, After successful restoration, an error will be reported “[ 50.936628] EXT4-fs error (device mmcblk0p1): __ext4_find_entry:1693: inode #3407873: comm systemd: checksumming directory block 0”
What is the reason for the above error to occur?
1.txt (1.1 KB)
2.txt (86 Bytes)
There is another situation:
I used my previous image, and then I bought a new module. When I tried to restore using the new module, it gave an error message like 3.txt. I suspect it might be caused by a material change. If it is due to the material change, is there any solution?
3.txt (68.4 KB)
There is also a picture showing an error message during startup.
Hi,
Please provide the commands for us to review.
Thanks
Hello DavidDDD
[quote=“DavidDDD, post:4, topic:365528”]
Please provide the commands for us to review.
The command I used is as follows:
$ systemctl stop udisks2.service
$ sudo tools/l4t_flash_prerequisites.sh
$ sudo service nfs-kernel-server start
backup:
sudo ./tools/backup_restore/l4t_backup_restore.sh -e mmcblk0p1-b jetson-agx-orin-devkit
restore:
sudo ./tools/backup_restore/l4t_backup_restore.sh -e mmcblk0p1-r jetson-agx-orin-devkit
Hello DavidDDD
Regarding the third situation, is there a solution for the material change? Do we have to remake the image?
Hi
Please try below commands instead
sudo ./tools/backup_restore/l4t_backup_restore.sh -b jetson-agx-orin-devkit
sudo ./tools/backup_restore/l4t_backup_restore.sh -r jetson-agx-orin-devkit
About third situation, we dont think related to material change based on the log.
Thanks
What you said I can give a try. But for the third point, what could be the reason?
I verified that this method does not produce the same error as MMC. But why does it give an error when I add the -e parameter? Moreover, the README for this version provides the documentation for adding -e.
I have another question. After we used our self-developed baseboard to back up the image and then bought a new Orin Nano to flash the system, we encountered the following error.
Hi,
It should be mmcblk0 instead of mmcblk0p1.
Seems like similar to Jetson Orin NX, backup restore not working with newly received modules, please check your new orin nano TNSPEC
Thanks
Thank you for the response
I’ll give it a try.