To add a database connection, click on the gear icon in the top right, and navigate to Admin settings > Databases > Add a database.
You can edit these settings at any time. Just remember to save your changes.
The display name for the database in the Metabase interface.
Your database’s IP address, or its domain name (e.g., esc.mydatabase.com).
The database port. E.g., 3306.
The database username for the account that you want to use to connect to your database. You can set up multiple connections to the same database using different user accounts to connect to the same database, each with different sets of privileges.
The password for the username that you use to connect to the database.
Use a secure connection (SSL)
You can paste your server’s SSL certification chain.
Use an SSH tunnel
See our guide to SSH tunneling.
Unfold JSON Columns
In some databases, Metabase can unfold JSON columns into component fields to yield a table where each JSON key becomes a column. JSON unfolding is on by default, but you can turn off JSON folding if performance is slow.
Additional JDBC connection string options
You can append options to the connection string that Metabase uses to connect to your database.
Re-run queries for simple explorations
Turn this option OFF if people want to click Run (the play button) before applying any Summarize or filter selections.
By default, Metabase will execute a query as soon as you choose an grouping option from the Summarize menu or a filter condition from the action menu. If your database is slow, you may want to disable re-running to avoid loading data on each click.
Choose when Metabase syncs and scans
Turn this option ON to manage the queries that Metabase uses to stay up to date with your database. For more information, see Syncing and scanning databases.
If you’ve selected Choose when syncs and scans happen > ON, you’ll see the following options under Database syncing:
- Scan sets the frequency of the sync query to hourly (default) or daily.
- at sets the time when your sync query will run against your database (in the timezone of the server where your Metabase app is running).
Scanning for filter values
Metabase can scan the values present in each field in this database to enable checkbox filters in dashboards and questions. This can be a somewhat resource-intensive process, particularly if you have a very large database.
If you’ve selected Choose when syncs and scans happen > ON, you’ll see the following options under Scanning for filter values:
- Regularly, on a schedule allows you to run scan queries at a frequency that matches the rate of change to your database. The time is set in the timezone of the server where your Metabase app is running. This is the best option for a small database, or tables with distinct values that get updated often.
- Only when adding a new filter widget is a great option if you want scan queries to run on demand. Turning this option ON means that Metabase will only scan and cache the values of the field(s) that are used when a new filter is added to a dashboard or SQL question.
- Never, I’ll do this manually if I need to is an option for databases that are either prohibitively large, or which never really have new values added. Use the Re-scan field values now button to run a manual scan and bring your filter values up to date.
Periodically refingerprint tables
Turn this option ON to scan a sample of values every time Metabase runs a sync.
A fingerprinting query examines the first 10,000 rows from each column and uses that data to guesstimate how many unique values each column has, what the minimum and maximum values are for numeric and timestamp columns, and so on. If you turn this option OFF, Metabase will only fingerprint your columns once during setup.
Default result cache duration
How long to keep question results. By default, Metabase will use the value you supply on the cache settings page, but if this database has other factors that influence the freshness of data, it could make sense to set a custom duration. You can also choose custom durations on individual questions or dashboards to help improve performance.
- Use instance default (TTL). TTL is time to live, meaning how long the cache remains valid before Metabase should run the query again.
If you are on a paid plan, you can also set cache duration per questions. See Advanced caching controls.
Connecting to MySQL 8+ servers
Metabase uses the MariaDB connector to connect to MariaDB and MySQL servers. The MariaDB connector does not currently support MySQL 8’s default authentication plugin, so in order to connect, you’ll need to change the plugin used by the Metabase user to
mysql_native_password`: `ALTER USER 'metabase'@'%' IDENTIFIED WITH mysql_native_password BY 'thepassword';
Unable to log in with correct credentials
How to detect this: Metabase fails to connect to your MySQL server with the error message “Looks like the username or password is incorrect”, but you’re sure that the username and password is correct. You may have created the MySQL user with an allowed host other than the host you’re connecting from.
For example, if the MySQL server is running in a Docker container, and your
metabase user was created with
CREATE USER 'metabase'@'localhost' IDENTIFIED BY 'thepassword';, the
localhost will be resolved to the Docker container, and not the host machine, causing access to be denied.
You can identify this issue by looking in the Metabase server logs for the error message:
Access denied for user 'metabase'@'172.17.0.1' (using password: YES).
Note the host name
172.17.0.1 (in this case a Docker network IP address), and
using password: YES at the end.
You’ll see the same error message when attempting to connect to the MySQL server with the command-line client:
mysql -h 127.0.0.1 -u metabase -p.
How to fix this: Recreate the MySQL user with the correct host name:
CREATE USER 'metabase'@'172.17.0.1' IDENTIFIED BY 'thepassword';
Otherwise, if necessary, a wildcard may be used for the host name:
CREATE USER 'metabase'@'%' IDENTIFIED BY 'thepassword';
That user’s permissions will need to be set:
GRANT SELECT ON targetdb.* TO 'metabase'@'172.17.0.1'; FLUSH PRIVILEGES;
Remember to drop the old user:
DROP USER 'metabase'@'localhost';
Syncing records that include JSON
Metabase will infer the JSON “schema” based on the keys in the first five hundred rows of a table. MySQL JSON fields lack schema, so Metabase can’t rely on table metadata to define which keys a JSON field has. To work around the lack of schema, Metabase will get the first five hundred records and parse the JSON in those records to infer the JSON’s “schema”. The reason Metabase limits itself to five hundred records is so that syncing metadata doesn’t put unnecessary strain on your database.
The problem is that, if the keys in the JSON vary record to record, the first five hundred rows may not capture all the keys used by JSON objects in that JSON field. To get Metabase to infer all the JSON keys, you’ll need to add the additional keys to the JSON objects in the first five hundred rows.
This JSON “schema” inference doesn’t work with MariaDB, due to implementation differences between MySQL and MariaDB.
Raising a MySQL Docker container of MySQL 8+
If you are spinning up a new MySQL container, and:
- you want Metabase to connect to the container without having to manually create the user or change the authentication mechanism,
- or you’re facing a
RSA public key is not available client side (option serverRsaPublicKeyFile not set)error,
['--default-authentication-plugin=mysql_native_password'] modifiers when you run the container, like so:
a simple docker run:
docker run -p 3306:3306 -e MYSQL_ROOT_PASSWORD=xxxxxx mysql:8.xx.xx --default-authentication-plugin=mysql_native_password
or in docker-compose:
mysql: image: mysql:8.xx.xx container_name: mysql hostname: mysql ports: - 3306:3306 environment: - "MYSQL_ROOT_PASSWORD=xxxxxx" - "MYSQL_USER=metabase" - "MYSQL_PASSWORD=xxxxxx" - "MYSQL_DATABASE=metabase" volumes: - $PWD/mysql:/var/lib/mysql command: ['--default-authentication-plugin=mysql_native_password']