Forum Discussion
Gateway Issues (personal mode)
Hello,
I have an on-premises data gateway (personal mode) and when I try to update my data either by "refresh scheduling" or "refresh now" and I am always returned an error:
"Before the data import is XXXXXXXXX finished, its data source timed out. Double-check whether that data source can process import queries, and if it can, try again."
but when i do the refresh in Power BI Desktop the upgrade is done without any kind of problems.
I also read here: https://docs.microsoft.com/en-US/power-bi/refresh-data
in the part of the note says "Data updates have to be completed in less than 2 hours."
it is also important to indicate that when I refresh on the Desktop the update always takes me between 5/6 hours because I get the data by import mode from an Oracle database.
Is there any way i can resolved this problem?
I already tried to put the latest version of the gateway but the data update failed and returned the same error as above. and now I am testing to clear the cache on the desktop, re-publish to the PBI service and try to update again. But I think it will again return the same error derived from what I read in the article above.
Could someone help me or indicate a better way to do this process?
Best Regards,
JO
3 Replies
- AnonymousNot applicable
Anonymous The least of the challenges here in my mind is the gateway... 5 hours to refresh? How large is the model, and why is it taking so long?
Something processing that long implies a really large model, but you are still able to publish which means that you are still under 1 GB. Can you provide some more context of your load, refresh process?
Alternatively, is there a reason you aren't using an enterprise gateway on a standalone server or machine?
- AnonymousNot applicable
I link to an oracle database, and I have to do import from my tables, my tables have all over 100 columns and all of them are string. (each table with more or less 50k / 80k of rows.
but I came up with an idea that is to have a script write a CSV of those tables and then load that CSV into the PBI Desktop. (which makes the process much faster.)
I'm using a personal just to do some testing. but I already have an enterprise.
thankss for your help.
- TeigeGaoSolution Sage
Hi Anonymous ,
It looks like that it is a performance problem, you can convert it to csv then import csv to PowerBI as a workaround, you can also consider reducing the size of data which used for importing.
By the way, if you still meet this problem, I would suggest you opening a ticket at https://powerbi.microsoft.com/en-us/support/
Best Regards,
Teige