Use Delta Live Tables pipelines with legacy Hive metastore
This article details configurations and caveats specific to Delta Live Tables pipelines configured to publish data to the legacy Hive metastore. Databricks recommends using Unity Catalog for all new pipelines. See Use Unity Catalog with your Delta Live Tables pipelines.
This article discusses functionality for the current default publishing mode for pipelines. Pipelines created before February 5, 2025, might use the legacy publishing mode and LIVE virtual schema. See LIVE schema (legacy).
How to query streaming tables and materialized views in the legacy Hive metastore
After an update is complete, you can view the schema and tables, query the data, or use the data in downstream applications.
Once published, Delta Live Tables tables can be queried from any environment with access to the target schema. This includes Databricks SQL, notebooks, and other Delta Live Tables pipelines.
When you create a target configuration, only tables and associated metadata are published. Views are not published to the metastore.
Configure a pipeline to publish to Hive metastore
When creating a new pipeline, you can specify Hive metastore under the Storage options to publish to the legacy Hive metastore. You must specify a default target schema when publishing to Hive metastore. See Configure a Delta Live Tables pipeline.
Specify a storage location
You can specify a storage location for a pipeline that publishes to the Hive metastore. The primary motivation for specifying a location is to control the object storage location for data written by your pipeline. Databricks recommends always specificying a storage location to avoid writing to the DBFS root.
Because all tables, data, checkpoints, and metadata for Delta Live Tables pipelines are fully managed by Delta Live Tables, most interaction with Delta Live Tables datasets happens through tables registered to the Hive metastore or Unity Catalog.
Cloud storage configuration
You use AWS instance profiles to configure access to S3 storage in AWS. To add an instance profile in the Delta Live Tables UI when you create or edit a pipeline:
- On the Pipeline details page for your pipeline, click the Settings button.
- In the Instance profile drop-down menu In the Compute section of the pipeline settings, select an instance profile.
To configure an AWS instance profile by editing the JSON settings for your pipeline clusters, click the JSON button and enter the instance profile configuration in the aws_attributes.instance_profile_arn field in the cluster configuration:
{
"clusters": [
{
"aws_attributes": {
"instance_profile_arn": "arn:aws:..."
}
}
]
}
You can also configure instance profiles when you create cluster policies for your Delta Live Tables pipelines. For an example, see the knowledge base.
Work with event log for Hive metastore pipelines
If your pipeline publishes tables to the Hive metastore, the event log is stored in /system/events under the storage location. For example, if you have configured your pipeline storage setting as /Users/username/data, the event log is stored in the /Users/username/data/system/events path in DBFS.
If you have not configured the storage setting, the default event log location is /pipelines/<pipeline-id>/system/events in DBFS. For example, if the ID of your pipeline is 91de5e48-35ed-11ec-8d3d-0242ac130003, the storage location is /pipelines/91de5e48-35ed-11ec-8d3d-0242ac130003/system/events.
You can create a view to simplify querying the event log. The following example creates a temporary view called event_log_raw. This view is used in the example event log queries included in this article:
CREATE OR REPLACE TEMP VIEW event_log_raw
AS SELECT * FROM delta.`<event-log-path>`;
Replace <event-log-path> with the event log location.
Each instance of a pipeline run is called an update. You often want to extract information for the most recent update. Run the following query to find the identifier for the most recent update and save it in the latest_update_id temporary view. This view is used in the example event log queries included in this article:
CREATE OR REPLACE TEMP VIEW latest_update AS
SELECT origin.update_id AS id
FROM event_log_raw
WHERE event_type = 'create_update'
ORDER BY timestamp DESC
LIMIT 1;
You can query the event log in a Databricks notebook or the SQL editor. Use a notebook or the SQL editor to run the example event log queries.
Example pipeline source code notebooks for workspaces without Unity Catalog
You can import the following notebooks into a Databricks workspace without Unity Catalog enabled and use them to deploy a Delta Live Tables pipeline. Import the notebook of your chosen language and specify the path in Source code field when configuring a pipeline with the Hive metastore storage option. See Configure a Delta Live Tables pipeline.