Jetson series device EXT4-fs (mmcblk0&&1p1): error loading journalmount: deu/mmcblk1p1: can't read superblocks

The picture below shows the state where the mmcblk0p1 file system is damaged and the system cannot start

 The picture below shows the state where the external sd card and emmc memory card are set, the external boot file system of mmcblk1p1 is damaged, and the system cannot start

 Both can be fixed with:

problem analysis

Check /dev/mmcblk0p1 and find that the block device is still there. Trying to mount the partition indicates that the file system may be damaged. Analysis: The file system is damaged.

Detection command ls /dev/mmcblk* 

problem solving

Prepare a U disk (NTFS, FAT32 is acceptable) and copy fsck to the U disk from other NX environments.

# 预设U盘根目录为/media/data/ mkdir /media/data/fsck
cp /sbin/fsck* /media/data/fsck

 After the copy of the inspection tool is completed, mount the U disk to the problem environment

# 预设U盘挂载目录为/mnt/data/
# U盘设备为/dev/sda2 (这个实际自己猜,因为boot环境没有lsblk) mount /dev/sda2 /mnt/data
cp -r /mnt/data/fsck /tmp cd /tmp/fsck/
chmod +x *
# 可以先扫描看看
./fsck.ext4 -n /dev/mmcblk0p1 # 自动修复
./fsck.ext4 -a /dev/mmcblk0p1

After repairing, see the repair report for details, and try to mount or restart. If it's just a filesystem issue, it should still be fixable. If it still cannot be repaired, consider using the SDK dump partition (if you need to back up data) and then re-flash. The previous external startup methods, such as the emmc attached to the surface of the domestic carrier board, the addition of an external sd card slot, and the emmc startup are all re-flashing and repairing sudo fsck.ext4 /dev/mmcblk1p1, using this method to skip this step.

The following is the previous external startup crash processing method:

This document uses the L4T3243(bsp) version test

In the process of using the power, it is found that the external emmc on the above Rite-Hite carrier board starts 128G, and the display cannot be used normally. The operation is as follows, according to the system you installed before, flash the core emmc again (16G on the Nvidia core module)

The flashing process is very simple and simple to demonstrate

$ Tegra186_Linux_R32.4.3_aarch64.tbz2

$ Tegra_Linux_Sample-Root-Filesystem_R32.4.3_aarch64.tbz2

$ Realtimes_L4T_3251_xavier-NX_V1.4.tar.gz

Attached download address

https://www.jianguoyun.com/p/Dd3RSWwQ5cbrChi_09YEIAA (access password: realtimes2022)

After downloading the corresponding three packages, set up the burning environment according to the burning manual, and you will find that there are only two directories as follows

Linux_for_Tegra

Realtimes_L4T_3243_xavier-NX_V1.3

$ cd Linux_for_Tegra

$ ls Check the burning parameters in case you don’t know which one to burn and report an error as shown in the figure below

$ sudo ./flash.sh rtso-6002-emmc mmcblk0p1

系统重新启动以后

$ df -h

$ sudo umount /media/nvidia/******(你挂掉的设备)
$ sudo fsck.ext4 /dev/mmcblk1p1
$ sudo resize2fs /dev/mmcblk1p1 
$ sudo vi /boot/extlinux/extlinux.conf 

修改mmcblk01p为mmcblk1p1
$  sudo reboot

 The picture below is the device after reflashing the device, the result of df -h has been restored to mmcblk0p1

Disposal result

After the repair is complete, restart and enter the system normally.