DataflowsError

[ArvindJha]

Hello all,

Getting the error when running Dataflow , it seems because of data size , how to resolve the error?

Error: PipelineException: The uncompressed data size specified in a packet header exceeds max limit: GatewayDecompressor - Header.UncompressedDataSize (211517212) of a compressed packet exceeds the maximum allowed uncompressed payload of 157286421..Param1 = PipelineException: The uncompressed data size specified in a packet header exceeds max limit: GatewayDecompressor - Header.UncompressedDataSize (211517212) of a compressed packet exceeds the maximum allowed uncompressed payload of 157286421 

[ArvindJha]

thanks for response , actually for extraction i give say 9 am , the power automate flow runs when extraction is successful but the trigger point for final report is when transformation is complete but both transformation and report starts running parallely even in if else condition

[Anonymous]

Hi @ArvindJha,

What you're running into is a common behavior in Power Automate, actions placed after each other don't automatically wait for the previous ones to complete unless explicitly configured to do so.

Even if you're using if/else conditions, if both branches or steps aren’t nested properly or controlled with "Configure Run After", Power Automate may still attempt to run them in parallel, especially if the actions don’t have a true dependency link.

Best Regards,
Hammad.
Community Support Team

[ArvindJha]

Thanks for the response but i dont find a way for dependency , transformation runs when extraction is successful but final dataset starts running in parallel

[Anonymous]

Power Automate doesn’t provide native dependency handling like a full orchestration tool, which makes this little tricky when you are chaining multiple dependent steps.

To enforce a true dependency, especially where the final dataset refresh must wait for the transformation to complete, you’ll need to explicitly check the status of the transformation dataflow before triggering the dataset refresh.

You can try it practically by these few steps:

* After triggering the transformation dataflow, use a loop with “Get dataflow refresh status”.

* Now inside the loop, check if the status is "Completed".

* Only after confirming completion, proceed to trigger the dataset refresh.

This polling approach ensures the dataset won’t refresh until the transformation is actually finished, even if Power Automate tries to run the steps in parallel.

Best Regards,

Hammad.

[ArvindJha]

Thanks for the response , i already have a get dataflow refresh status which triggers transformation when extraction is complete and not allowing another trigger

[Anonymous]

Since Power Automate allows only one trigger per flow and you've already built logic to check extraction completion before starting transformation, the next step is to ensure the dataset refresh is placed after the transformation status check, not as a separate branch or parallel step.

Here make sure the dataset refresh action is inside the same success path that runs after the transformation dataflow completes. And if you're using a "Do until" or status-check loop for the transformation, the dataset refresh should be outside the loop but placed immediately after it, not in a separate condition block or parallel branch.

This way, the dataset refresh won’t start until the transformation is truly done  even if there's no second trigger.

If this post helps then please mark it as a solution, so that other members find it more quickly.

Thank you,

Hammad.

[ArvindJha]

thanks for the response all start running parallely in True block

the condition is set as below for Dataset refresh still

[Anonymous]

Can you please confirm if your issue has been resolved. If yes then kindly mark the helpful reply as solution so that other community members can find the solution easily.

Best Regards,

Hammad.

[ArvindJha]

the issue is not resolved i shared the screenshot of the flow let me know if any questions

[Anonymous]

Thanks for sharing the screenshots. From what I see, the issue is happening because all three actions (two dataflows + one dataset refresh) are placed side-by-side inside the "True" branch, which means Power Automate treats them as parallel actions, even though you’ve set Run after on one of them. It’s still running in parallel because setting Run after only applies within the same action block, but parallel actions in the same branch (like your True branch) start simultaneously unless they are nested sequentially.

To fix this younwill need to sequence them by nesting one after the other like First Refresh a dataflow. Then inside that, add the next Refresh a dataflow 1 action. Then inside that, add the Refresh a dataset action, using “Add an action” below the previous one, not side by side.

This forces the flow to run them one after another in order not in parallel.

Best Regards,

Hammad.

[ArvindJha]

thanks for the response can you share screenshot or link which tells the steps as i dont find an option told above

[Anonymous]

Hi @ArvindJha,

Thanks for reaching out to the Microsoft fabric community forum.

It looks like you are facing issue whike running your dataflow and you suggest it might be because of data size. As @rajendraongole1 already responded to your query, please go through his response and mark it as solution if it solves your issue.

I would also take a moment to thank @rajendraongole1, 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.

If I misunderstand your needs or you still have problems on it, please feel free to let us know.  

Best Regards,
Hammad.
Community Support Team

If this post helps then please mark it as a solution, so that other members find it more quickly.

Thank you.

[Anonymous]

Hi @ArvindJha,

As we haven’t heard back from you, so just following up to our previous message. I'd like to confirm if you've successfully resolved this issue or if you need further help.

If yes, you are welcome to share your workaround and mark it as a solution so that other users can benefit as well. If you find a reply particularly helpful to you, you can also mark it as a solution.

If you still have any questions or need more support, please feel free to let us know. We are more than happy to continue to help you.
Thank you for your patience and look forward to hearing from you.

[Anonymous]

Hi @ArvindJha,

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. If my response has addressed your query, please accept it as a solution so that other community members can find it easily.

Thank you.

[Anonymous]

Hi @ArvindJha,

May I ask if you have resolved this issue? If so, please mark the helpful reply and accept it as the solution. This will be helpful for other community members who have similar problems to solve it faster.

Thank you.

[rajendraongole1]

Hi @ArvindJha  - These errors are common when there is a data packet being processed in your Power BI Dataflow exceeds the maximum allowed size

Try limiting the number of rows or columns processed in your Dataflow:

Try reducing columns, filtering data, and enabling Incremental Refresh first. If the issue persists, check gateway memory and consider partitioning your data.

option 2: If possible, split large tables into smaller tables and process them separately

ref: link:

Troubleshooting Dataflows in Power BI | Power Query Tutorial Ep8 | Power BI | BI Consulting Pro | 4K

https://community.fabric.microsoft.com/t5/Service/PipelineException-when-running-dataflow-Compressed...

[ArvindJha]

Thanks for the response all of this is already taken care of still the issue persists , for incremental load to happen it has to do full load first time right ? colums are reduced to what is required , it works absoultely fine in QlikSense

[Anonymous]

Hi @ArvindJha,

Thanks for reaching out to the Microsoft fabric community forum.

You're right about Power BI requiring a full load initially when Incremental Refresh is enabled, and it sounds like you've already optimized the dataset thoughtfully.

One thing to consider is how the query steps are structured within the dataflow. If a particular transformation (like a merge, append, or expand) results in a very large in-memory object even briefly it can cause a single packet to exceed the allowed uncompressed size, leading to this error. Unlike QlikSense, Power BI doesn’t stream all data step-by-step during transformation and might materialize intermediate results in memory depending on how the query is built.

If possible, try reviewing the query steps to see if a specific operation may be generating a large intermediate dataset. Breaking that transformation into smaller steps, perhaps by materializing earlier steps into a staging Dataflow or applying filters earlier in the chain can help distribute the load more evenly.

I would also take a moment to thank @rajendraongole1, 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.

If I misunderstand your needs or you still have problems on it, please feel free to let us know.  

Best Regards,
Hammad.
Community Support Team

If this post helps then please mark it as a solution, so that other members find it more quickly.

Thank you.

[ArvindJha]

There is no transformation , it is extraction dataflow only i am reducing the columns from 150 to 20 , qliksense is even working with 50 columns , pbi fails with 20 columns , in some cases actually the problem is a column called REASON which is text field ranges from 0-255 characters , if i reduce column size to 20 it works in some cases , in other cases even that does not work where the qvd size in qlik exceeds 5 GB , Power BI is unable to handle , so does it mean that PBI cannot handle such huge data?

[ArvindJha]

therea re some other cases where dataflow keeps on running for 24 hours and fails as that is time out limit