Jetson tx1 internal memory got 100 %

My jetson tx1 internal memory of 14 gb got full and i am not able to work on tx as it is showing “[ 258.365549] tegra-i2c 7000c700.i2c: no acknowledge from address 0x50” all times and not loading. I even tried to deleted python libraries to make some free space but no luck.

nvidia@nvidia-desktop:~$ df -h
Filesystem Size Used Avail Use% Mounted on
/dev/mmcblk0p1 14G 14G 0 100% /
none 1,8G 0 1,8G 0% /dev
tmpfs 2,0G 0 2,0G 0% /dev/shm
tmpfs 2,0G 118M 1,9G 6% /run
tmpfs 5,0M 4,0K 5,0M 1% /run/lock
tmpfs 2,0G 0 2,0G 0% /sys/fs/cgroup
tmpfs 397M 0 397M 0% /run/user/1000

Please help to make it work so i can boot the jetson from sd card here after. Is re flashing the jetpack is a solution or any other solution is possible.

You could clone the rootfs, loopback mount the rootfs clone on your PC, edit or remove content, and then use that edited clone to flash. You’d still be flashing, but the result would more or less be the same disk you started with (other than whatever you’ve removed).

Instructions on cloning and flashing may have some differences depending on which release you use. If you are ok with just flashing, then you can simply flash with whichever L4T/JetPack/SDK Manager release you want to use and call it good. You cannot mix a clone from one release with the content of a different release, and so it would help to know which release is currently installed and whether or not you need to clone.

Do note that you can use many different releases to create a clone, and the clone won’t care. It is the actual reinstall of the clone where release matters. So for example, if you wanted to use JetPack4.3, but the system contains something older, you could clone using 4.3, and have a reference image to examine from the PC, but you wouldn’t be able to directly flash with that clone (you would be relegated to copying individual files one at a time to the new install).

Which release are you using now, and how valuable is your content?