This repository holds all of the docker-compose.yml files and Dockerfiles needed for the various KPMP applications developed at UMich.
| Directory Name | Description |
|---|---|
| ara | Ara is the log aggregator. It is intended to be a central repository for logs from the various apps hosted at UMich. See ara's README for security details |
| atlas | Atlas is the website at atlas.kpmp.org. Inside the atlas directory is a repository and a knowledge-environment directory. The repository directory is for the Apache web server. The knowledge-environment directory is to host all of the Knowledge Environment containers |
| cassiopeia | Cassiopeia is the code name for the Participant Portal (mydata.kpmp.org) |
| DEFUNCT |
|
| dataLakeProxyServer | Since we are running the datamanager tools and the DLU on the same machine, we needed a proxy server running to handle the requests. This is that |
| delphinus | Delphinus is the code name for the Digital Pathology Repository (or DPR) (currently at demo.kpmp.org/dpr) |
| eridanus | Eridanus is the code name for the notification service. It is intended to notify individuals when particular events happen (no UI) |
| images | This directory stores common docker images that we share across projects. When we build an image, we store it on DockerHub |
| libra | Libra is the code name for the datamanager tools (datamanager.kpmp.org) |
| DEFUNCT |
|
| microscopium | Microscopium is the code name for the initial Atlas Demo (a cell-type centric view) (demo.kpmp.org/atlas) |
| orion | Orion is the code name for the Data Lake Uploader (upload.kpmp.org) |
| participant-portal-registration | Intended to allow us to run the Participant Portal registration site...this may not work |
| participant-portal | Intended to allow us to run the Participant Portal admin site...this may not work |
| pictor | Pictor is the code name for the second Atlas Demo (gene search and violin plots) (demo.kpmp.org/gene-search) |
| scripts | Directory to contain a number of scripts that are used for monitoring the health of the server and applications |
| stateManager | Manages the state of packages inside of the data lake |
Instructions on how to set up each project will be in individual README files under each directory.
There are a few directories that have a docker-compose.yml file and a dev-docker-compose.yml file. The dev-docker-compose.yml file is intended for developers to use on their local machines and will simplify running some of the projects. If there is no dev-docker-compose.yml file, you can and should just use the docker-compose.yml file in the directory.
Instructions on how to get your machine set up to run Docker for these projects
- Download and install Docker (https://docs.docker.com/desktop/)
- Launch Docker
- Go to Preferences -- Go to the Advanced tab and change your memory to 4GB -- (on a mac) Go to the File Sharing tab and at leas add your base directory (/Users) -- (on a PC) Go to the Shared Drives tab and check the box next to the C drive -- (on a PC) run 'set COMPOSE_CONVERT_WINDOWS_PATHS=1'
- Apply & restart docker
- Wait until docker engine completely starts (docker whale icon in taskbar will stop moving)
WARNING: Docker containers running in Windows 10 can have a wonky date. It is a known issue.
Many of the projects are stand-alone, however there are increasingly more projects that are run in conjunction with one another.
dataLakeProxyServer contains the Apache server and configuration for Shibboleth. It also starts the dataLake network.
orion uses the dataLake network and is served from the dataLakeProxyServer Apache libra uses the dataLake network and is served from the dataLakeProxyServer Apache. It also uses the mongo db created in orion eridanus uses the dataLake network. It also uses the mongo db created in orion