Purpose of the ansible script is to deploy the CD2-platform conform the Utrecht University ITS-standards of deployment.
The customized CKAN-based portal accompanied by an OAI-PMH-service allowing for harvesting the catalog (in datacite format only).
To be able to do so, the ansible script deploys 3 software modules as developed by Utrecht University.
These modules are being maintained within Github.
For the CD2-portal following repo's are required:
cd2-ckan-theme (https://github.com/UtrechtUniversity/cd2-ckan-theme).
Mainly contains the CD2-catalog's look & feel for the frontend application with CKAN as a base.
cd2-config (https://github.com/UtrechtUniversity/cd2-config).
Mainly contains configurational items for the catalog to able to work properly.
Think of the required CKAN and SOLR schema's for this particular application.
In order to add the OAI-PMH-service following repo is required:
ckanext-oaipmh-server (https://github.com/UtrechtUniversity/ckanext-oaipmh-server).
Before running the ansible playbook please check whether the correct branches or commits have been set for each github-repo.
In the playbook file:
cd2-ansible/roles/ckan/defaults/main.yml
check for and maybe adjust following entries and verify whether the correct/latest versions, branches or commits have been taken into account:
# CKAN extension versions - these are merely examples so please check the latest!
ckan_cd2_theme_version: 2873b614385444d1cb6b39425c201493049596ca # Version of 6 February 2024
ckan_cd2_config_version: aab2057e56705c0e57639f1aac6d3af36e71011a # Version of 6 February 2024
ckan_oaipmh_server_plugin_version: ab5144024cac361bfaa1d355a34920d26296f7bd # Version of 7 March 2024
Actual running of ansible is described here:
https://github.com/UtrechtUniversity/cd2-ansible/blob/development-its/README.md
After completion of the playbook the CD2-CKAN-catalog is up&running.
However, it is still an empty shell.
In order to be able to add data to the catalog following steps have to be taken first:
- Add administrator user(s)
Do this by logging in as the initial administrator (created by Ansible) via CKAN-URL /user/login
Administrator users are required as they need to perform the following steps:
-
Add organizations/studies
In this case the name of a study/organization is of importance due to the fact that the loader (external data loading application depends on these) [youth, trails, l-cid, radar, gen-r, ntr] only! -
Generate API key
The API key is required for external access through the CD2-CKAN-API. This enables adding/updating datapackages and organizations
The CD2-catalog, together with DataCite, are used as an integral publication platform for individual development data.
This requires data to be both published to the CD2-CKAN-catalog as well as DataCite; on top of that DataCite-publications will use the corresponding CD2-package publications as landingpages.
The data taxonomy loader application manages publication of datapackages/measures (https://github.com/UtrechtUniversity/cd2-taxonomy).
The loader application performs data preparation tasks required before the actual publication to
- CD2-catalog
- DataCite
Each CD2-datapackage published in DataCite gets its own DOI.
Each CD2-datapackage in DataCite has a landing page pointing to the datapackage within the CD2-catalog based on its CKAN identifier.
In order to be able to publish to DataCite the following configuration has to be taken into account:
Development/Testing purposes:
Datacite Website: https://doi.test.datacite.org
API endpoint: https://api.test.datacite.org/
Credentials as handed out by DataCite
Prefix: 10.80154
Production environment:
Datacite Website https://doi.datacite.org
API endpoint: https://api.datacite.org/dois/
Credentials as handed out by DataCite
Prefix: 10.60641
Publication to the CD2_catalog requires CKAN-API-access and the loader needs to be configured as such. This is described within the application code itself.
The API-key itself has already been created in a previous step.
After loading data, the CD2 platform is ready for use.