CKAN Configuration¶
Introduction¶
CKAN is often hard to install. When you do finally install it, it’s wholly functional, but capable of much more. By enabling a few features and installing a few extensions, it can be a markedly more useful product. There is a comprehensive list of available extensions on CKAN’s website that is worth browsing, to see if any are useful to your particular needs.
Installation¶
The biggest obstacle to using CKAN is installing it. Under most environments, it’s just plain hard.
If you’re a rare government agency that is able to use Amazon Web Services, simply launch an EC2 server using Link Digital’s CKAN image. Within a minute, you’ll have a CKAN server up and running.
If you can’t use AWS, then set up an Ubuntu server and install CKAN from package. Compared to installing from source, this is trivial.
If you can install it using a Docker image, that’s wonderful, but then you probably should be contributing to this How-To guide, not just reading it.
Failing those options, you can always install from source. This is hard.
Paid Hosting¶
You can always just pay somebody to install and host CKAN. This is a service that CKAN sells, and more vendors are starting to offer this service.
Recommended Additions¶
FileStore: Allows files to be uploaded to and stored within CKAN, instead of merely linking to them elsewhere.
DataStore: Opens up CSV and Excel files to store their contents, creating a lightweight API for the data and providing an on-screen preview of the data for end users.
DataPusher: When files in your repository are housed on a remote server, fetches them periodically to live within and benefit from DataStore.
data.json: Generates an inventory of all of your repository’s datasets, per the the U.S. Project Open Data standard.
ckanext-spatial: Displays a map for geodata and supports spatial queries against that data.
ckanext-googleanalytics: Provides Google Analytics data for your CKAN repository.