Purpose
Simple compose scripts to bring up various open source or 3rd party on local docker.
License
MIT License. See LICENSE.md
TODO: Add license text to each file
Available Containers/Images
| Product | Distribution | GUI container | Min Memory | Local Mounts |
|---|---|---|---|---|
| Anaconda | continuum.io | Jupyter notebook | ?? | ~/Documents/GitHub |
| Datastax | dse | DSE Studio | 2GB/4GB ? | -none |
| Dask | https://dask.org/ | Dask Dashboard + Jupyter | ?? | ~/Documents/GitHub |
| DynamoDB | dyanmodb-local | with web shell | ?? | -none- |
| Elasticsearch | Elastic.co | Kibana | ?? | -none- |
| [Kaggle Docker](docker-python-kaggle/README.m | ) | Kaggle | Jupyter notebook | 4GB |
| Kafka | Confluent.io CE with KSQL | Confluent GUI | 8GB | -none- |
| mongodb | Mongo | Mongo Express | ?? | -none- |
| neo4j | Neo4j.com | built in console | ?? | -none- |
| portainer | Portainer.io | Portainer UI | ?? | -none- |
| redis | Redis | Redis-Commander | ?? | -none- |
| Tensorflow | tensorflow.org | Jupyter notebook | ?? | ~/Documents/GitHub |
| Tika | https://tika.apache.org/ | Text extraction | ?? | -none- |
Using Docker
Open a command prompt. Change pwd to the docker container directiory in this repository and execute one of the following commands.
Starting and stopping docker with docker-compose
| command | cwd | Purpose |
|---|---|---|
docker-compose up |
compose file dir | bring up service attached to logs |
docker-compse -f <proj_dir>/<compos yml> up |
project root | bring up service attached to logs |
docker-compose up -d --build |
compose file dir | bring up service detached building new imsages |
docker-compose down |
compose file dir | bring down service |
docker-compse -f <proj_dir>/<compose_yml> down |
project root | bring down service - run from repo root |
Docker volume related commands
| Command | Purpose |
|---|---|
docker stats |
show memory and other usage |
docker volume ls |
list persistent volumes |
docker volume rm <volume> |
remove the persistent volume |
Opening a command line prompt on the VM hosting docker
Windows and Mac Docker desktop runs in a VM. The named volumes are inside that VM. You can look at the named volumes in the MobyLinux VM that hosts the docker containers with the following steps.
docker container run --rm -it -v /:/host alpinechroot /hostls /var/lib/docker/volumes/This from docker tips blog. Works for Windows and Mac.
Named Volumes
You will see some redundancy in the actual names of the named volumes declared in docker-compose.yml files. This is because docker-compose synthesizes the actual named volumen name. Docker prepends the directory docker-compose.yml directory name to the volume names specified in the docker-compose.yml files. This means that a named volume, dog_data run from inside the mongodb directory takes on the name mongdb_dog_data.
IDE Integration
Writing and deploying code for Jupyter Notebooks (anaconda) and Tensorflow
You can share a working / development directory using the file share.
- Enable file share for the correct drive using the Docker --> Settings --> Shared Drive panel
- Look in the docker-compose.yml files
- Uncomment the home directory Volumes line in docker-compose.yml
- Customize the line. Add any subdirectory path to the just un-commented Volumes entryf
- Start the container using
docker-compose - Open the IDE to a directory inside the directory you put in the docker-compose.yml file. From Docker container tips
Docker Management
The Docker plugin for Visual Studio Code supports container, volume and network management
Shell prompts in a docker container
The Docker plugin for Visual Studio Code supports opening a shell into the container from the docker control pane.