MetricSign
EN|NLRequest Access
High severityexecution

Power BI Refresh Error:
REPLICATION_GROUP_REFRESH_FAILED

What does this error mean?

A Snowflake replication group refresh operation failed to synchronize objects from the primary account to a secondary account, leaving the secondary account out of sync.

Common causes

  • 1The primary account's replication group contains objects that have been dropped or renamed since the last successful refresh, causing a metadata conflict
  • 2Insufficient privileges on the secondary account prevented the refresh from creating or replacing replicated objects
  • 3A transient network or cloud provider connectivity issue interrupted the cross-region data transfer during the refresh

How to fix it

  1. 1Step 1: Query REPLICATION_GROUP_REFRESH_HISTORY in SNOWFLAKE.ACCOUNT_USAGE on the secondary account to see the failure details and which objects were involved.
  2. 2Step 2: Check for dropped or incompatible objects on the primary account that may be included in the replication group — SHOW REPLICATION GROUPS and SHOW DATABASES IN REPLICATION GROUP.
  3. 3Step 3: If the issue is transient, re-trigger the refresh with ALTER REPLICATION GROUP <name> REFRESH on the secondary account.
  4. 4Step 4: Verify that the secondary account's ACCOUNTADMIN has the necessary privileges to replicate all object types in the group.

Frequently asked questions

Does a failed refresh leave the secondary in a partially updated state?

No. Snowflake uses atomic refresh semantics — if the refresh fails, the secondary account retains the state from the last successful refresh and is not left in a partial state.

How frequently does replication refresh occur?

Replication refreshes can be scheduled at a defined interval (e.g., every hour) or triggered manually. The interval is set at the replication group level via ALTER REPLICATION GROUP ... SET REPLICATION_SCHEDULE.

Other execution errors