Docker containers

Containers for running ESM related tools.

Citation: Containers for running NorESM and CESM, Anne Fouilloux and Jean Iaquinta, 2019,

Approach

Create a conda recipe for your tool. We usually add it either in conda-forge or bioconda.
When using bioconda, you will get a Docker container out-of-the-box which you can use as a starting point for running a specific configuration
Publish your docker. We use dockerhub and quay.io.

How to get inputdata

You will need wget to download the corresponding inputdata for each each you will be running.

sudo yum install wget -y

Docker for Community Earth System Models (CESM)

Community Earth System Model (CESM) supports several configurations. We provide examples on how to run CESM with docker containers. Several compset and resolutions have been tested and input data stored in zenodo to facilitate the usage and deployment of these docker containers.

Tested

Installation of docker on centos 7

To test and run our docker containers, we use IaaS and create a virtual machine with centos-7. Then we install docker using the following instructions:

sudo yum check-update


curl -fsSL https://get.docker.com/ | sh

sudo usermod -aG docker centos

sudo systemctl start docker

Check status

sudo systemctl status docker

Enable docker

It allows to start docker automatically when the machine is rebooted

sudo systemctl enable docker

Change default location

The default disk on our Virtual Machine is about 20GB e.g. too small for our docker images so we create a new volume /opt/uio/ (500GB).

First create a new volume and attach it to your Virtual Machine.

Then change default folder to /opt/uio/docker:

sudo mkdir /opt/uio/docker

You should have:

ls -lrt 

drwx--x--x. 14 root   root         4096 Nov  2 10:25 docker

And then change the default location to this new folder:

sudo systemctl stop docker

Then edit (with sudo) sudo vi /etc/docker/daemon.json and add:

{
  "data-root": "/opt/uio/docker"
}

Finally restart docker:

sudo systemctl start docker

Test your installation

Logout and login again (to make sure centos username is part of docker group).

docker ps

This command should return:

CONTAINER ID        IMAGE               COMMAND             CREATED             STATUS              PORTS               NAMES
0

A few docker commands

Get info on docker service

docker info

It should return something like:

Client:
 Debug Mode: false

Server:
 Containers: 1
  Running: 1
  Paused: 0
  Stopped: 0
 Images: 21
 Server Version: 19.03.4
 Storage Driver: overlay2
  Backing Filesystem: extfs
  Supports d_type: true
  Native Overlay Diff: true
 Logging Driver: json-file
 Cgroup Driver: cgroupfs
 Plugins:
  Volume: local
  Network: bridge host ipvlan macvlan null overlay
  Log: awslogs fluentd gcplogs gelf journald json-file local logentries splunk syslog
 Swarm: inactive
 Runtimes: runc
 Default Runtime: runc
 Init Binary: docker-init
 containerd version: b34a5c8af56e510852c35414db4c1f4fa6172339
 runc version: 3e425f80a8c931f88e6d94a8c831b9d5aa481657
 init version: fec3683
 Security Options:
  seccomp
   Profile: default
 Kernel Version: 3.10.0-1062.4.1.el7.x86_64
 Operating System: CentOS Linux 7 (Core)
 OSType: linux
 Architecture: x86_64
 CPUs: 32
 Total Memory: 117.9GiB
 Name: hpc32.novalocal
 ID: XNRD:EARM:KHVU:ZH47:QS32:NBE2:FHYX:MGW7:GMLK:RNIK:KT36:2J4D
 Docker Root Dir: /opt/uio/docker
 Debug Mode: false
 Registry: https://index.docker.io/v1/
 Labels:
 Experimental: false
 Insecure Registries:
  127.0.0.0/8
 Live Restore Enabled: false

If you container exited abnormally

docker ps --filter "status=exited"

Stop and remove your docker instance

Then stop and rm your container id:

docker stop <container_id>

and

docker rm <container_id>

Remove your docker images

docker images

The command above allows you to get the image identifier. Now you can remove your image with the following command:

docker rmi <image_id>

NordicESMhub / containers