What causes ReplicationLagExceeded?

The secondary database refresh schedule is too infrequent relative to the rate of primary changes. A large schema change or bulk DML on the primary caused the replication refresh to take longer than expected. The secondary database auto-refresh failed due to a transient network or warehouse issue and was not retried. Data retention time on the primary is shorter than the replication lag, causing snapshot unavailability. The replication group's suspension was not detected and replication silently stopped

How do I fix ReplicationLagExceeded?

Check lag time: SELECT SECONDARY_SNOWFLAKE_REGION, REPLICATION_LAG_TIME FROM SNOWFLAKE.ACCOUNT_USAGE.REPLICATION_GROUP_REFRESH_HISTORY.. Manually trigger a refresh: ALTER DATABASE REFRESH.. Increase refresh frequency in the replication schedule or switch to continuous replication where supported.. Verify the primary's DATA_RETENTION_TIME_IN_DAYS is greater than the maximum expected replication lag.. Set up Snowflake alerts on REPLICATION_GROUP_REFRESH_HISTORY to notify when refresh lag exceeds a threshold.

Medium severityexecution

Power BI Refresh Error:
ReplicationLagExceeded

What does this error mean?

A Snowflake secondary database replica has fallen too far behind the primary, causing downstream queries to read stale data. This manifests as unexpected NULL values or missing rows when querying a secondary that has not been refreshed recently enough.

Common causes

1The secondary database refresh schedule is too infrequent relative to the rate of primary changes
2A large schema change or bulk DML on the primary caused the replication refresh to take longer than expected
3The secondary database auto-refresh failed due to a transient network or warehouse issue and was not retried
4Data retention time on the primary is shorter than the replication lag, causing snapshot unavailability
5The replication group's suspension was not detected and replication silently stopped

How to fix it

1Check lag time: SELECT SECONDARY_SNOWFLAKE_REGION, REPLICATION_LAG_TIME FROM SNOWFLAKE.ACCOUNT_USAGE.REPLICATION_GROUP_REFRESH_HISTORY.
2Manually trigger a refresh: ALTER DATABASE <secondary_db> REFRESH.
3Increase refresh frequency in the replication schedule or switch to continuous replication where supported.
4Verify the primary's DATA_RETENTION_TIME_IN_DAYS is greater than the maximum expected replication lag.
5Set up Snowflake alerts on REPLICATION_GROUP_REFRESH_HISTORY to notify when refresh lag exceeds a threshold.

Frequently asked questions

How often should I refresh a Snowflake secondary database?

For near-real-time replication, use Snowflake's continuous replication (Business Critical+). For standard replication, set the refresh interval based on your RPO — every 5-15 minutes for low-RPO scenarios.

Does Power BI always query the Snowflake secondary if configured?

Power BI connects to a specific Snowflake account URL. If the URL points to the secondary, it always reads from the secondary regardless of lag. Switch the data source connection to the primary account URL during a replication lag incident.