Power BI Services Manually update works fine but scheduled not works

[luisgue]

Hi,

I have a report publish in power bi Services,  and scheduled every day to update, I used a connection to cloud with synapse, and its fine the connection and credentials. I used the same user admin to schedule the report every day and update manually. 

When Update manually the report in power bi services this works fine and this take 3 minutes.

But the schedule update everyday  starts to delay and finally have and error after 1 hour.

The capacity is "Reserved capacity for Pro Workspaces"

Aditionally the last 2 years this report scheduled update works fine, but since last month this fail.

Thanks.

[v-sshirivolu]

Hi @luisgue ,

I would also take a moment to thank @Olufemi7 , for actively participating in the community forum and for the solutions you’ve been sharing in the community forum. Your contributions make a real difference.
 

I wanted to check if you had the opportunity to review the information provided. Please feel free to contact us if you have any further questions

[v-sshirivolu]

Hi @luisgue ,

I hope the above details help you fix the issue. If you still have any questions or need more help, feel free to reach out. We’re always here to support you

[GilbertQ]

Hi @luisgue 

Can you send through a screenshot of the error message?

[luisgue]

Hi @GilbertQ 

there is details of executions

[GilbertQ]

Could you send through the error message from the Power BI Service?

[Olufemi7]

Hello @luisgue, 

This behavior is expected when scheduled refresh runs in the Power BI Service. Manual refresh completes quickly because it runs interactively, but scheduled refresh executes as a background job with stricter limits and shared capacity resources. Even with reserved capacity, if other datasets are refreshing or if queries have grown heavier, the scheduled job can be delayed and eventually fail after hitting the service time limit.

Since your dataset worked fine for years and only recently started failing, the most likely causes are:

• Increased data volume or slower queries in Synapse.
• Loss of query folding in Power BI transformations.
• Capacity overload (other datasets refreshing at the same time).
• Service-side changes in refresh behavior or limits.

 What you should do

1. Check refresh history in Power BI Service → Dataset → Refresh history. This will show duration and error details.
2. Optimize queries in Synapse and confirm query folding in Power BI Desktop.
3. Enable incremental refresh so only recent partitions are refreshed instead of the full dataset.
4. Stagger refresh schedules to avoid overlapping jobs in the same reserved capacity.
5. If needed, scale up reserved capacity or move the dataset to a less busy workspace.

Microsoft Documentation 

Configure scheduled refresh - Power BI | Microsoft Learn https://learn.microsoft.com/en-us/power-bi/connect-data/refresh-scheduled-refresh

Troubleshoot refresh scenarios - Power BI | Microsoft Learn https://learn.microsoft.com/en-us/power-bi/connect-data/refresh-troubleshooting-refresh-scenarios

How This Applies to Your Case

• Manual refresh works because it runs interactively and completes in ~3 minutes.
• Scheduled refresh fails because background jobs in Power BI Service have stricter limits (usually 2 hours max) and depend on shared capacity resources. If other datasets are refreshing or queries have grown heavier, the scheduled job can be delayed and eventually fail.
• Since this issue started recently after years of success, the most likely causes are capacity overload or loss of query folding in your Synapse connection.

Recommended Actions

1. Review refresh history in Power BI Service for error details.
2. Optimize queries in Synapse and confirm query folding in Power BI Desktop.
3. Enable incremental refresh to reduce load.
4. Stagger refresh schedules to avoid overlap in reserved capacity.
5. If needed, scale up reserved capacity or move the dataset to a less busy workspace.

Conclusion

Use the official Microsoft docs above to configure and troubleshoot scheduled refresh. The key is that scheduled refresh runs under stricter limits and shared capacity, so optimizing queries and enabling incremental refresh are the best ways to restore reliability.

[luisgue]

hi @Olufemi 

there is details of executions

[Olufemi7 · Accepted solution]

Hi @luisgue , thanks for sharing the execution details.

From your logs, the successful run shows much higher CPU time and peak memory, while the failed runs are short and light with a consistent error count. That pattern points to the scheduled refresh being cut off early due to capacity pressure, throttling, or a breakdown in query folding rather than a credentials issue. Enabling incremental refresh, confirming folding, and checking capacity health are the most effective next steps.

What to do next

• Review capacity health and throttling
• Install and open the Microsoft Fabric Capacity Metrics app to see refresh queues, throttling, and memory pressure.
• Use the Health and Compute pages to identify spikes or overlapping refreshes.
• Links:
• Install the metrics app: https://learn.microsoft.com/en-us/fabric/enterprise/metrics-app-install
• Health page overview: https://learn.microsoft.com/en-us/fabric/enterprise/metrics-app-health-page
• Compute page details: https://learn.microsoft.com/en-us/fabric/enterprise/metrics-app-compute-page
• Enable incremental refresh
• Refresh only recent partitions to reduce load and avoid long-running background jobs.
• Start with RangeStart/RangeEnd parameters, filter by date, and define the policy before publishing.
• Links:
• Overview: https://learn.microsoft.com/en-us/power-bi/connect-data/incremental-refresh-overview
• Configuration steps: https://learn.microsoft.com/en-us/power-bi/connect-data/incremental-refresh-configure
• Confirm query folding to Synapse
• In Power BI Desktop, check “View Native Query” in Power Query steps; if unavailable, folding is broken and refresh will be heavier.
• Links:
• Folding guidance: https://learn.microsoft.com/en-us/power-bi/guidance/power-query-folding
• Folding examples (including Synapse scenarios): https://learn.microsoft.com/en-us/power-query/query-folding-examples

Why this addresses your issue

• Incremental refresh reduces the amount of data processed during scheduled jobs, which improves reliability under capacity limits and background time constraints.
• Query folding ensures transformations are pushed down to Synapse, minimizing data movement and CPU/memory usage in the service.
• Capacity metrics reveal whether refreshes are overlapping or throttled, helping you adjust schedules or isolate the workspace if needed.

These steps should help stabilize your scheduled refresh.

Let me know how it goes after applying incremental refresh or checking capacity metrics.

[collinq]

Hi @luisgue ,

if this was working and now it is not, my first question would be - what changed?  Did the user have a token that might have expired?  Did the userid or password change?  Has there been a firewall change?

If you can do this locally than the data itself probably didn't change so I would inspect the items above first.