Problem with SPICE in Quicksight - Data was not imported - Please Help

Good morning. I’m having trouble importing data into a dashboard in QuickSight.

My process works like this: the data is saved in Athena, an engine is run to modify the data, it’s saved back to Athena, and a dataset is created from those tables.

It worked perfectly for over a month (no changes were made, and it’s updated daily). But today, QuickSight is telling me that 0 rows were imported from Spice.

However, if I check the dataset, the refresh tab shows that 3400 rows were imported (which is normal; it’s usually 3100-3500). But the summary tab shows “no rows imported.” The thing is, sometimes it says yes, and then after 5 seconds it reverts to saying no rows were imported.

I’ve tried duplicating it, or editing a calculated field, and when I try to save or duplicate it, I get an error message. It’s strange because I own the dataset.

If anyone knows how I can fix this, I’d appreciate it, as the AI ​​suggests I have to recreate it (which I really don’t want to do).

UPDATE: I create a new dataset and also a new dashboard. But i get the same problem, data not imported. How can i check if my account is the problem?

Hi @Alejandro_fdez and welcome to the Quick Community!

This is an interesting scenario as you mentioned that everything worked perfectly for a time period and then stopped all of a sudden with no changes made.

A couple things to check:

  • Have you tried switching it to Direct Query to see if that type of connection works?
  • Have you confirmed no issues on the Athena side?
  • You mentioned that you’re receiving an error when trying to save or duplicate, what was the error message you were receiving?
  • Have you tried importing with both the newer data prep experience as well as trying to revert back to the legacy experience?
  • On your dataset’s refresh tab, are there any ingestions that are listed as ‘Failed’; if so, and you click on it, what is the error message you’re receiving?
  • Lastly, often times ingestions fail or skip rows if they encounter field values that do not match the field type, is it possible that a value that got updated is in an invalid format?

Additionally, it may be worthwhile to create a support ticket so that you can report this behavior to the AWS support team and get more direct account assistance.

Hi Bret, thanks for the response!

I’ve tried switching to DirectQuery and it does work properly. It’s slow, but it works. In Athena there don’t seem to be any errors, and there have been no changes to the source table format.

The error message simply says that an error has occurred and that I should try again later, but it doesn’t specify the issue. It only shows the error ID and the date.

I’ve tried using the new one in several different ways and I get the same error every time. I also tried with just a single table, without changing anything, and the same thing happens.

The ingestions seem to be correct, it reads the rows, but then in the Summary it says that none of them have been imported.

Do you think this could be a permissions issue with my account?

Hi @Brett, I am facong similar problem, the quick soght dataset which completed the full refresh within 3 hours is failing since 2 weeks. Even the incremental refresh of 24 hours fails after 10 hrs duration.

Hi @Tanu_13 - 3 hours is a big time for refresh, can you please check the sql and test it by running in source database directly. Possibly a query tuning is require in this case.

hi @Alejandro_fdez - Is it possible to create a new dataset and import the data in spice mode, just to check whether the dataset is corrupted or not.

Hi @Sanjeeb2022 @brett ! Thanks again for your reply!

I tried creating a new dataset with both source tables (without calculated fields and without filters). And another dataset with only one table.

I get the same result in both: the refresh tab shows that the rows have been ingested, but then the summary tab says they haven’t been imported.

I checked Spice and I do have available space.

UPDATE: Now I get an error when I try to edit the dataset; it won’t let me save the changes. And the same thing happens with the new one I created.

Hello! Any comments please :sob:

Hi @Alejandro_fdez,

If DirectQuery does work properly, I agree that there’s nothing wrong with the connection to Athena. As this seems to be an issue with how Quick is ingesting, your best option may be to create a support ticket. The support team should be able to access additional information regarding the error being received

I create the dataset again, but with the legacy mode. It seems it works now. Really strange error. Because i also create the same in new mode (not legacy) and is not working.

1 Like

@vignessh.b, adding for visibility

@Alejandro_fdez can you share the datasetId created in new mode and also the region ? If you dont want to share these details in public then please create a ticket ?