A few details would help narrow this down.
When the scheduled run executes:
Does the pipeline update show Succeeded or Failed?
In the pipeline Event Log, do you see rows being processed/written?
Is your manual run a normal update or a Full Refresh?
Is the schedule configured directly on the DLT pipeline or triggered through a Workflow/Job?
Common causes are:
No new data available at the scheduled time (the update succeeds but processes 0 rows).
Incremental processing vs Full Refresh (manual full refresh reprocesses historical data, scheduled updates only process new data).
Different configuration/parameters between scheduled and manual runs.
Permission differences if the pipeline is being triggered through another service or workflow.
The first thing I'd check is the Event Log for a scheduled run and compare it to a successful manual run. If both succeed but the scheduled run shows 0 records processed, the issue is usually related to source data availability or incremental processing rather than the pipeline logic itself.
Data Engineer | Apache Spark | Delta Lake | Databricks