Last Updated: 2023-01-31 19:00:30
Curator Best Practices: Containerizing Your Portal
Matthew Orr
Curator Engineer
August 25, 2022
Disclaimer
The topic discussed here is not covered by Curator support and is not recommended as it is difficult to predict all the ways this may cause problems as the platform evolves. This is only a guide provided as an attempt to make your life easier if you need to containerize Curator due to internal IT policies. By taking the approach discussed below, you are assuming the burden of setting it up to work in your environment, maintaining it as needed and fixing issues as they arise. You’ve been warned. Good luck!
Deploying a Container
This is mostly an implementation detail for you to figure out for your container platform. However, we highly recommend starting with one of our official Curator installers as a base. If needed, you can configure things like database connection details, persistent storage location, license key or other advanced configuration during the installation. The Linux installer includes various arguments to set these. See the section on “Custom Setup” at this link for more details.
Persistent Database
Curator uses a small database to house your portal’s configuration. This database and all of its data need to persist between rebuilds of your container. It may be better to use an external database, such as a centralized MariaDB installation or something like Amazon RDS.
Persistent Filesystem
While Curator’s code can be rebuilt within a container, Curator stores uploaded files to the
Configuration
Curator’s configuration files don’t necessarily need to be persistent, however, it is important to retain the database encryption key from the
System Upgrades
There are two parts to a Curator upgrade that happen during a normal one-click upgrade. These will need to be accounted for in your container environment to keep rebuilds from reverting to the previous version of your portal. In fact, when hosting Curator in a containerized manner, it’s recommended to disable all upgrades through the interface for this reason.
Code Updates
To update Curator’s code, use the manual update link found in your portal to download the zip archive. It will need to be extracted and everything inside of the core directory of that extract (but not the directory itself) copied over the web root directory of your installation.
Database Updates
As features are added to Curator, small tweaks to the database need to be made. These database updates are included as migration scripts in the above code updates. Once the code has been updated, the database migration scripts can be executed by running the following command from the portal’s web root folder.
Linux (Redhat-like systems):
sudo -u apache php artisan portal:migrate
Linux (Ubuntu-like systems):
sudo -u www-data php artisan portal:migrate
Windows Server:
php.exe artisan portal:migrate
Recently, Tableau has been encouraging the use of connected apps for external applications, instead of using trusted tickets. All of Tableau’s recent embedding features require connected apps. Since this only deals with behind the scenes authentication, there is no impact to the end user. In our effort to remain closely aligned with Tableau, Curator is transitioning to only using connected apps.
Curator has added the feature to be able to send mark commenting data to a webhook. With the widespread use of API integration platforms, this really opens the doorway to virtually unlimited use cases.
If you’ve got users reporting access issues, your first stop on the road to resolution should be the User Menu Access button. This feature gives you a snapshot view of any given user’s current menu structure, as well as their permissions status for all the content within that menu.