The SQS queue (AWS) or Event Grid subscription (Azure) was deleted or misconfigured after pipe creation. The Snowflake storage integration lost permission to read from the notification queue. The queue accumulated too many events and processing lagged behind the incoming rate. The pipe was paused and queue events expired before the pipe was resumed

Step 1: Run SELECT SYSTEM$PIPE_STATUS('pipe_name') to check the pipe state and last error.. Step 2: Verify the cloud notification queue (SQS or Event Grid) still exists and the Snowflake integration IAM role has sqs:ReceiveMessage and sqs:DeleteMessage permissions.. Step 3: Refresh the pipe with ALTER PIPE pipe_name REFRESH to manually scan the stage for unprocessed files.. Step 4: If the queue has expired messages, re-upload or re-stage the affected files.. Step 5: Set up Snowflake data loading history monitoring with COPY_HISTORY() to detect future lag.

High severityconfiguration

Power BI Refresh Error:
090114

What does this error mean?

Snowpipe's auto-ingest mechanism failed to process events from the cloud storage notification queue, stalling continuous data ingestion.

Common causes

1The SQS queue (AWS) or Event Grid subscription (Azure) was deleted or misconfigured after pipe creation
2The Snowflake storage integration lost permission to read from the notification queue
3The queue accumulated too many events and processing lagged behind the incoming rate
4The pipe was paused and queue events expired before the pipe was resumed

How to fix it

1Step 1: Run SELECT SYSTEM$PIPE_STATUS('pipe_name') to check the pipe state and last error.
2Step 2: Verify the cloud notification queue (SQS or Event Grid) still exists and the Snowflake integration IAM role has sqs:ReceiveMessage and sqs:DeleteMessage permissions.
3Step 3: Refresh the pipe with ALTER PIPE pipe_name REFRESH to manually scan the stage for unprocessed files.
4Step 4: If the queue has expired messages, re-upload or re-stage the affected files.
5Step 5: Set up Snowflake data loading history monitoring with COPY_HISTORY() to detect future lag.

Frequently asked questions

How do I check how many files are pending in a Snowpipe queue?

Run SELECT SYSTEM$PIPE_STATUS('pipe_name') — the response includes pendingFileCount and the last error message. You can also query SNOWFLAKE.ACCOUNT_USAGE.COPY_HISTORY for detailed file ingestion history.

Does ALTER PIPE REFRESH process all files in the stage?

ALTER PIPE REFRESH re-submits files from the stage that have not yet been loaded. It is idempotent — files already loaded will not be loaded again.