Research Database

The Research Database is a digital repository designed for public discovery of the research conducted by economists affiliated with the Federal Reserve Bank of Minneapolis Research Division. Publications, data, and conference proceedings constitute the majority of materials. The Research Database uses open source software, Hyrax, developed by the Samvera community.

BagIt Functionality

BagIt archives can be created from the command line using the bag rake task:

USER_ID=1 BAG_PATH=/tmp/bags rake bag:create[publication1,publication2,publication3]

USER_ID is the id of the user that will be notified.

The bag will be located in the directory set by the ENV['BAG_PATH'] environment variable or you can specify it when running the rake task. It will be named mpls_fed_research_<time-stamp>.tar (time stamp being a unix time stamp). The contents of the bag will consist of all the attached files in the specified publications. The created bag will use sha256 to create checksums for the files.

To download a bag that has been created you can visit the route: /bag/mpls_fed_research_<time-stamp>.tar.

Local Development with Docker

Using docker is the most straightforward way of setting up a local development environment.

1. Install docker for your local environment
2. Check out the cypripedium code base and navigate to that code base: git clone git@github.com:MPLSFedResearch/cypripedium.git; cd cypripedium

3. Bring up the docker container: docker-compose up Once the container has started, you should be able to visit in your browser:

   1. http://localhost:8983/solr/#/hydra-development (dev instance of solr)
   2. http://localhost:8983/solr/#/hydra-test (test instance of solr)
   3. http://localhost:8080/rest/ (dev and test instance of fedora)
   4. http://localhost:3000/ (Your locally running rails application)

Running the tests

The best way to know that your local development environment is running as expected is to run the automated test suite:

1. In a new terminal window, navigate to your code base
2. Connect to your running rails container: docker-compose exec web bash
3. Run the test suite: bundle exec rspec spec

Local Development without Docker

1. Install postgresql. Because we use the postgresql database in production, we follow the rails recommendation and also use it in development. The easiest way to set up a local development instance is to set it up to allow local password-less connections (follow this guide)
2. Run bundle exec rake db:setup to create expected databases and run database migrations
3. Run solr_wrapper; check in your browser at localhost:8983/solr to see it running.
4. Run fcrepo_wrapper; check in your browser at localhost:8984/rest to see it running.
5. Run sidekiq. You can see it running at localhost:3000/sidekiq
6. Run redis-server; Redis will run, you can check redis running by running redis-cli ping in a new window

Data setup

Whether you use docker or not, there are a few steps necessary to let you interact with your local instance.

1. Generate collection types and admin set:

   bundle exec rails hyrax:default_collection_types:create
   bundle exec rails hyrax:default_admin_set:create

2. Generate a new user and make them an admin on the console

   bundle exec rails c
   u = User.new
   u.display_name = "User Name"
   u.email = "email@testdomain.com"
   u.password = "password"
   u.password_confirmation = "password"
   u.save
   admin = Role.create(name: "admin")
   admin.users << u
   admin.save

   If you then type u.admin? it should return True if the previous steps were successful.

3. bundle exec rails s. In your browser, check to see the application running at localhost:3000.

4. In the brower session, log in as the admin user you just created.

Capistrano deployment

There are currently 2 environments for cypripedium: the legacy ec2 installation & and the future SSM environment.

To deploy to the legacy environment, the following command will continue to work

cap <environmentname> deploy

To deploy to the new SSM environment, you must have a configured .aws/credentials file, with a profile named "frbm-ssm".

After this is in place,

bundle exec cap ssm deploy

Will deploy to the SSM environment. The deployment system uses tags to determine the instance ids of the relevant target hosts (SSM does not use ip addresses or the like), based on the value of the host_env environment variable

For instance, a host with the tags Environment=stage and Project=rdb will be the target of

host_env=stage bundle exec cap ssm deploy

The system defaults host_env to "prod"

Ansible

When building a new system via Ansible, the "localhost.rb" configuration will be used, and requires no further steps (other than having frbm-ssm profile set up)