- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content

effect of downtime on incremental refresh
Consider this scenario: a data source with incremental refresh policy enabled, no last-update-time field, the refresh time range is 3 days, and archive is 6 months. After prolonged normal operations (i.e. all data is in place), the refresh starts to consistently fail and after a few failed attempts it gets auto-disabled by the system. If the model remains in this disabled state for, let's say 14 days, what will happen after things get back to normal: will there be a gap in the data between day 3 and day 14, or will the service detect the gap and fill it in?
I couldn't find any documentation detailing this scenario, but just a read of the documentation seems to suggest that there will be a gap.
A pointer to some literature describing this or an account of personal experience of this situation is appreciated.
Solved! Go to Solution.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content

Hi, @beranes
If the incremental refresh policy is in place, and the refresh operation consistently fails to cause auto-disable, any data that was not refreshed during the outage will not be automatically backfilled after the refresh is resumed. This means that, depending on your scenario, if refresh has been disabled for 14 days, there will indeed be a gap in the data between days 3 and 14.
An incremental refresh policy is designed to refresh data within a specified refresh range, in this case 3 days. Once the system is back to normal and refreshes resume, it will continue to apply the policy based on the current date and will not retroactively populate data for time periods where refreshes did not occur.
To address this gap, you'll need to manually trigger a refresh for the affected time period. You can do this by adjusting the and parameters to cover the interval in Power BI Desktop, and then republishing the report to the Power BI service. After a manual refresh, you can reset the parameters to the original settings. For more information on configuring incremental refresh and handling data gaps, you can check out the link:
How to Get Your Question Answered Quickly
Best Regards
Yongkang Hua
If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content

Hi, @beranes
If the incremental refresh policy is in place, and the refresh operation consistently fails to cause auto-disable, any data that was not refreshed during the outage will not be automatically backfilled after the refresh is resumed. This means that, depending on your scenario, if refresh has been disabled for 14 days, there will indeed be a gap in the data between days 3 and 14.
An incremental refresh policy is designed to refresh data within a specified refresh range, in this case 3 days. Once the system is back to normal and refreshes resume, it will continue to apply the policy based on the current date and will not retroactively populate data for time periods where refreshes did not occur.
To address this gap, you'll need to manually trigger a refresh for the affected time period. You can do this by adjusting the and parameters to cover the interval in Power BI Desktop, and then republishing the report to the Power BI service. After a manual refresh, you can reset the parameters to the original settings. For more information on configuring incremental refresh and handling data gaps, you can check out the link:
How to Get Your Question Answered Quickly
Best Regards
Yongkang Hua
If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.

Helpful resources
Subject | Author | Posted | |
---|---|---|---|
09-23-2024 09:41 AM | |||
02-28-2025 06:59 AM | |||
02-11-2025 03:24 PM | |||
02-06-2025 04:29 AM | |||
11-13-2024 05:02 PM |
User | Count |
---|---|
109 | |
89 | |
81 | |
55 | |
46 |