eMMC storage is getting filled and not able to run other applications by using docker containers

[samjith888]

I have installedJetson OS BSP into Jetson TX2 NX device , along with that installed the Jetson SDK components too by using SDK manager.

Then tried pulling the Deepstream docker image, but its not pulled since out of memory issues.

Note: Already 128 GB SD card is attached in this device.

Command used while flashing the device:

./flash.sh <target_board> mmcblk0p1 - boot <target_board> from eMMC

I have to run another heavy deep learning models in this device. Any help to tackle the out of memory issue will be highly appreciated.

[AastaLLL]

Hi,

Do you run out of memory or storage?

Jetson’s memory is fixed and not able to extend.
But for larger storage, you can try to add an extra NVMe SSD:
https://docs.nvidia.com/jetson/archives/l4t-archived/l4t-3273/index.html#page/Tegra%20Linux%20Driver%20Package%20Development%20Guide/flashing.html#wwpID0E0VM0HA

Thanks.

[samjith888]

I have 128 GB sd card which is attached to this device. If i change the boot from sd card, does it decrease the performance ? or can i mount the sd card with the Jetson memory(eMMC).

Installing new packages (DL frameworks, docker, etc) will use the boot storage, which is 16GB. This boot storage its getting filled when i installed the Jetson OS + Jetson SDK components. So i couldn’t pull a docker image even. Is there any way to extend the boot storage by using the SD Card storage

[AastaLLL]

Hi,

An alternative is to change the docker folder to the sd card.

$ service docker stop
$ mv /var/lib/docker /<new location>/docker
$ ln -s /<new location>/docker /var/lib/docker
$ service docker start

Thanks.