Using Liquibase with AWS Redshift

Amazon Web Services Redshift is a fully managed, petabyte-scale data warehouse service in the cloud. An AWS Redshift data warehouse is a collection of computing resources called nodes. The nodes are organized into a group called a cluster. Each cluster runs an AWS Redshift engine and contains one or more databases.

Note: For more information, see the Amazon Redshift documentation page.

Supported Versions

  • 3.0.28 – officially supported and tested with Test Harness

Prerequisites

Before using Liquibase and AWS Redshift, ensure you have completed the following:

Driver Information

To use Liquibase and AWS Redshift, you need to have two jar files – JDBC and the Liquibase Redshift extension:

  1. Ensure you have downloaded the Amazon Redshift JDBC driver. You can download the JDBC 4.2–compatible driver (without the AWS SDK). If you use the Amazon Redshift JDBC driver for database authentication, make sure that you have AWS SDK for Java 1.11.118 or later in your Java class path. If you don't have AWS SDK for Java installed, download the ZIP file with the JDBC 4.2–compatible driver (without the AWS SDK) and driver dependent libraries for the AWS SDK.
  2. Place the redshift-jdbc<version>.jar file in the liquibase/lib directory.
  3. Example: redshift-jdbc42-2.0.0.4.jar

  1. Open your liquibase.properties file and specify the driver value in it:
  2. driver: com.amazon.redshift.jdbc42.Driver

    Note: For more information about the liquibase.properties file, see Creating and configuring a liquibase.properties file.

  1. Go to the liquibase-redshift repository and download the latest released Liquibase extension liquibase-redshift-<version>.jar file.
  2. Place the liquibase-redshift-<version>.jar file in the liquibase/lib directory.
  3. If you put the redshift-jdbc<version>.jar and liquibase-redshift-<version>.jar files in any other directory, specify the path to them in the liquibase.properties file:

    Windows example:

    classpath:"..path_to_your\\drivers\\redshift-jdbc<version>.jar; ..path_to_your\\liquibase-redshift-<version>.jar"

    Linux example:

    classpath:"../path_to_your/drivers/redshift-jdbc<version>.jar; ..path_to_your/liquibase-redshift-<version>.jar"

If you use Maven, you also need to download the AWS Redshift driver jar file and put the driver in a location that your Maven build can access. Configure the Maven pom.xml file to use the local copy of the driver jar file. For example:

<dependency>
<groupId>com.amazon.redshift</groupId>
<artifactId>redshift-jdbc42</artifactId>
<version>2.0.0.4</version>
</dependency>

Additionally, you need to specify the Liquibase Redshift extension in your pom.xml file as explained in Configuring Liquibase Attributes in your Maven POM File. Make sure that the Liquibase plugin and the extension have the same version.

<dependency>
<groupId>org.liquibase.ext</groupId>
<artifactId>liquibase-redshift</artifactId>
<version>4.4.0</version>
</dependency>

Testing Your Connection

For Liquibase and AWS Redshift to work, you need to:

  1. Ensure your AWS Redshift database is configured. You can check the connection to an Amazon Redshift cluster.
  2. Specify the database URL in your liquibase.properties file:
  3. url: jdbc:redshift://endpoint:port/database

    Example: url: jdbc:redshift://<cluster-identifier>.us-east-1.redshift.amazonaws.com:5439/databasename

    Note: To get your JDBC connection, see Finding your cluster connection string.

  4. Create a text file called changelog (.xml, .sql, .json, or .yaml) in your project directory and add a changeset.
  1. Run the status command to see whether the connection is successful. Specify the name of the changelog you created in place of changelog.xml. Also, to run Liquibase commands, you need to specify your username and password in the liquibase.properties file or on the command line. However, the username and password attributes are not required for connections and systems which use alternate means of authentication.
  2. liquibase --username=test --password=test --changeLogFile=changelog.xml status

    Note: Alternatively, you can add the changeLogFile attribute to your liquibase.properties file.

  1. Run your first update with the Liquibase update command, which will make changes to your database. You can also run the updateSQL command if you want to inspect the SQL before running the update command.
  2. liquibase --changeLogFile=changelog.xml updateSQL
    liquibase --changeLogFile=changelog.xml update

After your first update, you will see a new table along with the DATABASECHANGELOG and DATABASECHANGELOGLOCK tables added to the database:

  • DATABASECHANGELOG table. This table keeps a record of all the changesets that were deployed. When you deploy, the changesets in the changelog are compared with the DATABASECHANGELOG tracking table, and only the new changesets that were not found in the DATABASECHANGELOG will be deployed.
  • DATABASECHANGELOGLOCK table. This table is used internally by Liquibase to manage access to the DATABASECHANGELOG table during deployment and ensure only one instance of Liquibase is updating the database at a time, whether that is creating, updating, or deleting changes.

Supported Commands and Change Types

Change Type Supported Command Supported
addAutoIncrement Supported calculateCheckSum <id> Supported
addCheckConstraint Supported changelogSync Supported
addColumn Supported changelogSyncSQL Supported
addDefaultValue Supported changelogSyncToTag Supported
addForeignKeyConstraint Supported changelogSyncToTagSQL Supported
addLookupTable Supported clearCheckSums Supported
addNotNullConstraint Supported dbDoc Supported
addPrimaryKey Supported deactivateChangeLog Supported
addUniqueConstraint Supported diff Supported
alterSequence Supported diff JSON Supported
createFunction Supported diffChangeLog Supported
createIndex Supported dropAll Supported
createPackage Supported futureRollbackCountSQL <value> Supported
createPackageBody Supported future-rollback-from-tag-sql <tag> Supported
createProcedure Supported futureRollbackSQL Supported
createSequence Supported generateChangeLog Supported
createSynonym Supported help Supported
createTable Supported history Supported
createTrigger Supported listLocks Supported
createView Supported markNextChangeSetRan Supported
customChange Supported markNextChangeSetRanSQL Supported
delete Supported registerChangeLog Supported
disableCheckConstraint Supported releaseLocks Supported
disableTrigger Supported rollback <tag> Supported
dropAllForeignKeyConstraints Supported rollbackCount <value> Supported
dropCheckConstraint Supported rollbackCountSQL <value> Supported
dropColumn Supported rollbackOneChangeSet Supported
dropDefaultValue Supported rollbackOneChangeSetSQL Supported
dropForeignKeyConstraint Supported rollbackOneUpdate Supported
dropFunction Supported rollbackOneUpdateSQL Supported
dropIndex Supported rollbackSQL <tag> Supported
dropNotNullConstraint Supported rollbackToDate Supported
dropPackage Supported rollbackToDateSQL Supported
dropPackageBody Supported snapshot Supported
dropPrimaryKey Supported snapshotReference Supported
dropProcedure Supported status --verbose Supported
dropSequence Supported syncHub Supported
dropSynonym Supported tag <tag string> Supported
dropTable Supported tagExists <tag string> Supported
dropTrigger Supported unexpectedChangeSets Supported
dropUniqueConstraint Supported update Supported
dropView Supported updateSQL Supported
empty Supported updateCount <value> Supported
enableCheckConstraint Supported updateCountSQL <value> Supported
enableTrigger Supported updateTestingRollback Supported
executeCommand Supported updateToTag <tag> Supported
insert Supported updateToTagSQL <tag> Supported
loadData Supported validate Supported
loadUpdateData Supported    
markUnused Supported    
mergeColumns Supported    
modifyDataType Supported    
output Supported    
renameColumn Supported    
renameSequence Supported    
renameTable Supported    
renameTrigger Supported    
renameView Supported    
setColumnRemarks Supported    
setTableRemarks Supported    
sql Supported    
sqlFile Supported    
stop Supported    
tagDatabase Supported    
update Supported    

Related Links