Skip to main content
automatically move data from operational databases to lakehouses. When source schemas change, pipelines adapt automatically to keep data flowing without breaking downstream processes.

Configure a pipeline

When you create a pipeline, configure the following settings:

Select destination

Choose the destination for your data. You have to configure at least one destination before you can create a pipeline.

Handle schema changes

How Orka handles changes

When schemas change in the source database, intelligently:
  1. Keep unaffected data flowing - other fields continue working normally
  2. Track changes - metadata records alterations
  3. Apply transformations as needed:
    • Dropped columns: safe defaults or nulls
    • Renamed columns: automatic mapping
    • New columns: appropriate handling

Impact on your pipeline

Orka isolates changes to affected columns. Other columns continue working reliably.

Pipeline lifecycle

Create a pipeline

  1. Discover data in the data catalog
  2. Select tables and columns needed
  3. Configure destination and settings
  4. Orka provisions infrastructure

Monitor a pipeline

Orka provides visibility into:
  • Data flow rates and volumes
  • Schema change events
  • Error rates and types
  • Pipeline health status

Pipeline statuses

Orka tracks pipeline health through status indicators:
StatusDescription
RUNNINGPipeline is active and processing data normally
STARTINGPipeline is initializing and preparing to process data
PAUSEDPipeline is manually paused and does not process data
PARTIALLY_PAUSEDSome components of the pipeline are paused while others continue running
FAILEDPipeline has encountered an error and stopped processing
UNKNOWNPipeline status can’t be determined (usually a temporary state)
Check the pipeline details page for specific error messages and recommended actions when a pipeline shows FAILED status.

Maintain a pipeline

Check pipeline status regularly and address errors when they occur.

Delete a pipeline

Always check for active pipelines before you delete connections.

Our recommendations

  • Start with non-sensitive data - create your first few pipelines with non-sensitive data to understand the workflow
  • Monitor schema changes - monitor schema change notifications to understand how your source systems evolve
  • Test in development first - test pipelines in development environments before deploying to production

Troubleshoot

If you delete a connection that’s being used in an active pipeline, the pipeline will break.To fix:
  • Recreate the deleted connection with the same configuration, or
  • Create a new pipeline with a different connection
Always check for active pipelines before you delete a connection.
Check these common causes:Source connection issues
  • Test the source connection
  • Verify database credentials are still valid
  • Check network connectivity
Destination connection issues
  • Test the destination connection
  • Verify write permissions
  • Check storage capacity
Creating new data protection rules after you publish tables triggers a re-scan of all tables, which can temporarily affect active sync pipelines.
Create and test all necessary data protection rules before you publish tables to the catalog.