Managing SQL connections¶
In order to use a Kubernetes deployment with:
SQL query endpoints
Dataset lookup endpoints
Query enrichments in prediction endpoints
You need to setup the SQL connections that these endpoints will use. You need to declare the connection settings as seen from the Kubernetes cluster. You may need to pay special attention to firewall and authorization rules
Configuring the connection used for storage of bundled data¶
NB: this is not applicable to SQL Query endpoints
Please see Enriching prediction queries for more information
Go to Infrastructure > Settings > Connections
Fill in the “Connection for bundled” field with a DSS connection definition.
In this UI, you can select an existing connection (defined on the API Deployer node). This will copy the definition to clipboard, which you can then paste into the definition field
Note
You must replace encrypted passwords by a decrypted version. Password encryption is not supported in Kubernetes deployments at the moment. It is not currently possible to hide the passwords in this screen
Configuring the “referenced” connections¶
Please see Enriching prediction queries for more information
Go to Infrastructure > Settings > Connections
Enter the name of the connection as it is defined in the API Designer
Add and fill the definition field with a DSS connection definition.
In this UI, you can select an existing connection (defined on the API Deployer node). This will copy the definition to clipboard, which you can then paste into the definition field
Note
You must replace encrypted passwords by a decrypted version. Password encryption is not supported in Kubernetes deployments at the moment. It is not currently possible to hide the passwords in this screen