Difference between revisions of "IoTGateway/Docker"

Jump to: navigation, search
Line 227: Line 227:
:Update to 3.6 or later.
:Update to 3.6 or later.
  sudo apt-get install python-software-properties
  sudo apt-get install -y python-software-properties
  <span style="color:#00AA00"># Install this package on Ubuntu 14.04 or later</span>
  <span style="color:#00AA00"># Install this package on Ubuntu 14.04 or later</span>
  sudo apt-get install software-properties-common
  sudo apt-get install -y software-properties-common
  sudo add-apt-repository ppa:deadsnakes/ppa
  sudo add-apt-repository -y ppa:deadsnakes/ppa
  sudo apt-get update
  sudo apt-get update
  sudo apt-get install python3.6
  sudo apt-get install -y python3.6
  python3.6 -V
  cd /usr/bin
sudo rm python
sudo ln -s python3.6 python
python -V
== References ==
== References ==

Revision as of 04:25, 14 December 2020

Use Docker to set up build environment more safely & quickly


To deliver a clean build environment for customer development, we are used to provide a VMWare image. However, this kind of hypervisor mechanism is bigger and always comes worse performance.[1] Fortunately, we have an alternative way for this now. We put the build environment into a container. In a container, performance for building process is similar as native computer[2], and it also provides process isolation & resource limitation for security and stability.

We adopt Docker to achieve this feature.


Before you use Docker to develop, you have to install Docker on your platform. Please refer to Docker Installation Guide. You are able to install Docker on Linux, Cloud, Windows, and OS X. In general, you may choose to install on Ubuntu.[3]


If you would like to use Docker as a non-root user, you should now consider adding your user to the "docker" group with something like:

$ sudo usermod -aG docker $your_user

Note: You may need to login again to activate.

Get base image

To get the images we provided, you can use docker pull <IMAGE REPOSITORY> to get the images in the image list.

$ docker pull advrisc/u12.04-imx6lbv3:20160303

20160303: Pulling from advrisc/u12.04-imx6lbv3
6df68bc875bf: Pull complete
20f5595aa1a4: Pull complete
93b424f526d8: Pull complete
c877b3e9524f: Pull complete
29a731492f99: Pull complete
3701b127b54b: Pull complete
15247541bc49: Pull complete
729f8eff6afb: Pull complete
130b818ebc2f: Pull complete
7239f480205c: Pull complete
fbcbd43c9696: Pull complete
3f211662827e: Pull complete
65fa0f0910df: Pull complete
fda959c610b9: Pull complete
85fd678cfd54: Pull complete
0c63b8fa6a0f: Pull complete
87e1750c6c91: Pull complete
837fb16417bc: Pull complete
d72b58a6026b: Pull complete
597115359768: Pull complete
92116cadf140: Pull complete
Digest: sha256:603bff4244d016d8318c59218296dc145fd32f7e4f8f4f5c94decca11941dbdf
Status: Downloaded newer image for advrisc/u12.04-imx6lbv3:20160303

You can check the image stored locally by docker images.

$ docker images
REPOSITORY                               TAG                 IMAGE ID            CREATED             VIRTUAL SIZE
advrisc/u12.04-imx6lbv3                  20160303            92116cadf140        30 minutes ago      1.443 GB

Create a container

You can create a new container with a prompt by docker run. If you did not pull the Docker base image before, it will try to download it from the DockerHub repository first.

$ docker run -it advrisc/u12.04-imx6lbv3:20160303 /bin/bash


To leave a container, type exit to stop it or use Ctrl + P + Q to detach the container.

adv@89e9548a0d40:~$ exit

You can check the container status by docker ps.

$ docker ps -a
CONTAINER ID     IMAGE                              COMMAND        CREATED            STATUS                      PORTS     NAMES
89e9548a0d40     advrisc/u12.04-imx6lbv3:20160303   "/bin/bash"    27 seconds ago     Exited (0) 4 seconds ago              dreamy_jennings

To enter the container that you've created, you need to do docker start <CONTAINER ID> & docker attach <CONTAINER ID>.

$ docker start 89e9548a0d40

$ docker attach 89e9548a0d40

Share data with host

There are two ways to share data with host OS.

  • data volume
  • docker cp

You have to assign data volumes when creating the container. Use docker run with -v options to enable data volume. In the case below, we mount the BSP directory (/home/root/imx6LBV5080_2016-02-03) from your Docker daemon’s host into a container. So, you can access the source code at /home/adv/BSP in the container.

$ docker run -it --name docker_test -v /home/root/imx6LBV5080_2016-02-03:/home/adv/BSP:rw advrisc/u12.04-imx6lbv3:20160303 /bin/bash

// Let adv get access permission
adv@7bc2d878c870:~$ sudo chown adv:adv -R BSP

Alternatively, you can also access data in containers after the container is created. Use docker cp <CONTAINER:PATH> <LOCALPATH> to pull out data from the container, or docker cp <LOCALPATH> <CONTAINER:PATH> to push data into the container.

// Copy sdcard image from container to host
$ docker cp 89e9548a0d40:/home/adv/imx6LBV5080_2016-02-03/build_all/tmp/deploy/images/imx6qrsb4410/fsl-image-x11-imx6qrsb4410.sdcard .

// Copy BSP source code into container
$ docker cp imx6LBV5080_2016-02-03.tgz 89e9548a0d40:/home/adv

Remove image or container

Once you do not need the container, you can use docker rm <CONTAINER ID> to remove it.

$ docker rm 89e9548a0d40

If you want to remove Docker base image, use docker rmi <IMAGE ID>

$ docker rmi advrisc/u12.04-imx6lbv3:20160303

BSP Building

It's very easy to build our BSP in Docker environment.

  1. Get BSP source code
  2. Run a Docker container with data volume sharing for the BSP, or you can use docker cp to put BSP into the container.
  3. In the container, go to BSP folder and follow the same introductions as you do on normal computer.
  4. Finally, copy your images from container to host.

Image List

All our images are stored in DockerHub. You can explore and find the image as listed below.

Image Repository Image Tag Base OS Supported Platforms
advrisc/u12.04-imx6lbv3 20170103 Ubuntu 12.04 Freescale i.MX6 Yocto 1.5 (fsl-yocto-3.10.17_1.0.x)
Freescale i.MX6 Yocto 1.7 (fsl-yocto-3.14.28_1.0.0)
Freescale i.MX6 Yocto 1.8 (fsl-yocto-3.14.52_1.1.0)
Freescale i.MX6 Yocto 2.0 (fsl-yocto-4.1.15_1.1.0)
Freescale i.MX6 Yocto 2.1 (fsl-yocto-4.1.15_2.0.0)
advrisc/u12.04-quarklbv2 20160621 Ubuntu 12.04 Intel Quark Yocto 1.4.2 (Intel Quark BSP 1.0.4)
Intel Quark Yocto 1.7.2 (Intel Quark BSP 1.2.0)
advrisc/u12.04-imx6abv1 20160713 Ubuntu 12.04 Freescale i.MX6 Android 4.2 (android_JB4.2.2_1.1.0)
advrisc/u12.04-imx6abv2 20160809 Ubuntu 12.04 Freescale i.MX6 Android 4.3 (android_JB4.3_1.1.0)
Freescale i.MX6 Android 4.4.2 (android_KK4.4.2_1.0.0)
advrisc/u14.04-imx6abv4 20170310 Ubuntu 14.04 Freescale i.MX6 Android 5.0.0 (android_L5.0.0_1.0.0)
Freescale i.MX6 Android 5.1.1 (android_L5.1.1_2.1.0)
Freescale i.MX6 Android 6.0.1 (android_M6.0.1_1.0.0)
advrisc/u16.04-imx6abv5 20170523 Ubuntu 16.04 Freescale i.MX6 Android 7.0.0 (android_N7.1.1_1.0.0)
Rockchip RK3399 Android 7.1.2
advrisc/u14.04-410clbv1 20170717 Ubuntu 14.04 Qualcomm db410c Yocto 2.0 (oe-rpb-manifest jethro)
Qualcomm db410c Yocto 2.1 (oe-rpb-manifest krogoth)
Qualcomm db410c Yocto 2.2 (oe-rpb-manifest morty)
advrisc/u14.04-am57lbv1 20170313 Ubuntu 14.04 TI AM57xx Yocto 2.1 (Processor SDK Linux


You can use Dockerfile to create your own base image. Your Dockerfile may looks like this.

FROM advrisc/u12.04-imx6lbv3

MAINTAINER Vendor's Company

USER root

# The packages you want to install
RUN apt-get install -y NewPackages

WORKDIR /home/adv
USER adv

Then, you can use docker build to generate a new base image with tag.

$ cd <the-folder-which-has-Dockerfile>
$ docker build -t vendor/new-base-image .

Bug Fix


If your repo command is upgraded to v2.0 and above, you have to upgrade git & python3.

  • Git
Update to 2.10.2 or later.
sudo apt-get install -y libcurl4-openssl-dev
sudo apt-get install -y tk gettext

wget https://github.com/git/git/archive/v2.25.0.tar.gz
tar zxvf v2.25.0.tar.gz
cd git-2.25.0/

make configure
./configure --prefix=/usr
make all doc
sudo make install install-doc install-html

git --version
  • Python3
Update to 3.6 or later.
sudo apt-get install -y python-software-properties
# Install this package on Ubuntu 14.04 or later
sudo apt-get install -y software-properties-common

sudo add-apt-repository -y ppa:deadsnakes/ppa
sudo apt-get update
sudo apt-get install -y python3.6

cd /usr/bin
sudo rm python
sudo ln -s python3.6 python

python -V


  1. Differences between Docker, LXC & Hypervisor (Chinese)
  2. Docker Performance
  3. Docker Installation on Ubuntu
  • External Link
  1. Docker Introduction (Chinese)