Forum Discussion
Dataflow Timeout
- 7 years ago
Hi Mike,
Please send to our support email the screenshot with the failure you got, so we can investigate further.
Thanks,
Galia
Hi Galia
I don't get to the refresh stage as the dataflow times out when I am previewing the data when building the dataflow.
I am unable to save the dataflow.
Regards
Mike
Hi Mike,
Please send to our support email the screenshot with the failure you got, so we can investigate further.
Thanks,
Galia
- RBunting7 years agoAdvocate III
I am also having a problem like this. Do we know any more about Time Out control?
I am trying to retrieve a portion of a large table - about 10 of 88MM rows. If I load that data into its own SQL table, the Dataflows priview loads instantly. But if I try to load from the correct table, DataFlows works at it for about 10 minutes and quits.
I wouldn't expect the query to complete in SSMS in that time. How can we get Dataflows preview to be a little more patient?
- Anonymous7 years agoNot applicable
RBuntingHow are you managing to download such huge dataset to your Desktop? I still dont understand why PowerBI Desktop needs to download the entire Dataflow into the desktop application. I am talking about 10s of GBs of data. Make no sense.
v-yuta-msft
I tried to highlight some of the issues with Dataflow in y post here, but there is ablsutely no response ! My whole Org was couting on using Dataflow for enterprise level ETL.
https://community.powerbi.com/t5/Service/Dataflow-Dataset-NEED-more-Clarity/td-p/735330- RBunting7 years agoAdvocate III
AnonymousI think there's been some misunderstanding. I am loading about 10 million rows, but that's not anything like 10s of Gigs.
The problem in my case is that it simply takes time for SQL Server to pull those records out of the large table. If they are in a table by themselves they load easily. It's not a matter of size -- I just wish the Dataflow's time-out when loading the preview would be more generous, or be managed manually.
To avoid this problem, I build the M code in Desktop since it's an easier environment to work in. But I don't load any of that data into Desktop's model. I only want the code out of power query. Then I paste that into the Dataflow's advanced editor.
At this point, waiting for the preview would create a time-out error. So I don't wait. I just save and close, confident from my preview in desktop that I'm getting the data I need. Naturally, I'll do some QA on the dataflow once it's properly loaded.