cancel
Showing results for 
Search instead for 
Did you mean: 

PowerQuery - Validating queries when saving is taking excessively long time

I'm struggling to be able to work efficiently with Dataflows as I am experiencing excessive times to 'validate' queries after pressing the 'Save & Close' button. For entities with even a minimal amount of transformation (eg. simply changing some column data types) I've experienced wait times of over 3 hours to simply save the query to exit out back to the main entity menu/view. Should things be taking this long for an output schema validation process, regardless of the size of the underlying table data? Is this an issue being experienced by others as well?

This is making it simply impossible for me to work with the Dataflows product in this way.

 

2019-12-03_16-00-57.png

Status: New
Comments
Advocate I

I've been experiencing this today as well, quite a bit

Community Support

Hi all, 

 

Which kind of data sources does the dataflow get data from? Is there any specific features made to the entity, eg: Merge/Append queries? 

 

Please test with the same steps in Power BI desktop to check if the same issue occurs. 

 

In Power BI service, please use Blank Query, copy and paste the same Power Query then check if you are able to save the dataflow. 

 

Best Regards,
Qiuyun Yu

Advocate I

I've personally been using a blank query and copying from a Power BI Desktop model into Dataflows so I can incorporate custom queries.  It got better as I added more data sources to the Dataflow, but the first Save & Close took quite a bit of time, much longer than the same model I created through Power BI Desktop.  No crazy merging/appending going on... I tend to do most of my work in SQL prior to Power BI, and the queries there are pretty efficient.

 

EDIT: Nope, didn't get better, got far worse, and now my dataflow just won't save.  Several hours wasted trying to covnert my shared dataset over to a dataflow.

New Member

This is still a huge query.


I can prep the M in Power BI Desktop, then copy that M into a blank query in dataflows, and experience time outs 9/10 times. Doing any kind of native query seems to time out the validating the schema, which is weird considering it's not even loading the data! 

 

Why is this continually timing out? What are the contributing factors? Is it trying to run the entire query just to get an output schema? and if so why on earth would it do that for the entire query and then request you run the query again to populate the data??

 

Any help on this @v-qiuyu-msft  would be awesome.

 

Thanks


Chris Mcalpine

 

Regular Visitor

Hi 

 

Apologies for rescusitating this old thread but id be interested if anyones found the reason for this yet?

 

Ive had this problem lately when running data flows and native queries where it times out when validating the schema

 

No idea whats causing this as the very same query has been working fine last week?

 

Thanks


Richard

Regular Visitor

Facing the same issues, its very frustatting. The same query was working fine last start of the week. @v-qiuyu-msft 

Frequent Visitor

Same here. I went through the process of copy and pasting from the desktop tool and the 'Validating queries' appears to have been running for over 3 hours now.

 

New Member

Same issue here. The 'Validating queries' screen also appears to launch a lot of queries on our database.

Frequent Visitor

Same here. I have been testing by building a simple data flows with just a table date inside, which generate in 1 or 2 seconds on the desktop version, the preview in dataflows online also generates in the same time.

However, when I click on the button to save the data flows, the validation phase is still going on after 20 minutes and + 

I have no idea what is going on but this is really a big big blocker for fast development.

Frequent Visitor

Having the same issues. As mentioned above. Dataflows were working fine for months. Any edits during those times used to validate in a matter of seconds.

 

Now we are approaching it taking hours.

 

Please advise.