Select Language:
If you’re seeing a “column not found” error in QuickSight while working with your Iceberg tables, but the same query runs smoothly in Athena’s query editor, the issue is likely due to schema caching or sync problems between QuickSight and your Athena data source. Here’s a straightforward way to fix it.
First, try refreshing the dataset schema. Open your analysis in QuickSight and go to the Visualize tab. Select “Choose dataset,” then click on “Edit analysis data sets.” On the screen that appears, find your dataset and click “Edit” to refresh it. Keep in mind that QuickSight caches dataset schemas for about two minutes, so give it a little time for the update to take effect.
Next, check your custom SQL query if you’re using one. Make sure you’ve spelled the column name, like ‘probability_of_default,’ exactly as it appears in your table. Do not surround the column name with single quotes—they are used for strings, not column identifiers. Also, include the table’s schema name explicitly in your query, for example, “SELECT FROM my_schema.my_table” instead of just “SELECT FROM my_table.” Double-check for any syntax errors as well.
It’s also helpful to verify if there have been recent changes to the table structure. In the Athena console, look at the query history to see if the schema was modified recently. The column might have been renamed or the table schema changed, which can cause QuickSight to lose track of the columns.
Finally, make sure your tables and columns adhere to Athena’s naming and structural rules, especially for Iceberg tables. Incorrect naming or formats can cause issues too.
If after trying all these steps the problem still isn’t solved, it may be related to how QuickSight interacts with Iceberg table metadata. Sometimes, it takes a bit more time for QuickSight to sync properly with the latest table settings.
For more detailed help, you can check the official AWS documentation on troubleshooting for Athena and QuickSight.
ChatGPT
Perplexity
Gemini AI
Grok AI
