To connect Looker to Amazon Aurora MySQL, follow the instructions found on the documentation page for connecting to MySQL on Amazon RDS.
In addition to the steps in the Amazon RDS instructions, Amazon Aurora may need further setup, depending on your configuration. If you have a redirected read-only endpoint for Amazon Aurora, or if you want to leverage persistent derived tables (PDTs), see the sections below.
Configuring Amazon Aurora with Redirected Read-Only Endpoints
For the typical Amazon Aurora read-only endpoint, the URL contains a
-ro element, which Looker will detect and then automatically set the connection’s
However, if your configuration tunnels to your Amazon Aurora read-only endpoint, or if you are using redirects, your URL may not include the
-ro element. If your URL doesn’t have the
-ro element, go to the Looker Connection Settings page and add the following to the Additional Params field:
Configuring Amazon Aurora MySQL for PDTs
In order to leverage persistent derived tables (PDTs) with Aurora, you must use MySQL replication, not Amazon Aurora’s default replication, which is read-only. You must set the
read_only parameter to
0 to make the MySQL replica writable, as described in our documentation on RDS and temporary tables.
If you don’t want to grant write access to the database, you can copy and paste the derived table SQL into the
sql_table_name parameter of a
view file as shown below. This creates a subquery that is used at query time:
For more details on Aurora replication, see the AWS documentation.
For Looker to support some features, your database dialect must also support them.
In Looker 21.0, Amazon Aurora supports the following Looker features:
After completing the database configuration, you can connect to the database from Looker using these directions.