Dataset gives error for old deleted tables and columns

[stiggrr87]

Hello,

This has happened too many times. When trying to refresh dataset or use data factory pipelines then I get error that some tables are missing. Or some columns are missing. But they were some old tables and columns removed some time ago.

Is it possible to somehow clean the dataset or better to make a new one?

Sometimes its a warning and sometimes it fails with error message.

[stiggrr87]

Hi!

I have a problem with the refresh dataset activity. Cannot see the tables. About 6 months ago there were tables visible, but now its having a strange error. I cannot use this activity and need to rely on power bi own refresh schedule. 

[v-menakakota]

Hi @stiggrr87,
Thanks for reaching out to the Microsoft fabric community forum. 

Since it shows “No results found” for tables, this looks like a metadata sync problem between the pipeline and the semantic model, not just a missing column/table. The pipeline is unable to read the model schema.

You can try re-selecting the semantic model in the activity and clicking Refresh to reload the tables. Also, consider deleting and recreating the refresh activity, as pipelines sometimes cache old metadata.

If the issue still persists, the quickest fix is to re-publish the dataset and reconnect it to the pipeline, as this usually resolves any stale or broken metadata references.

If I misunderstand your needs or you still have problems on it, please feel free to let us know. 

Best Regards, 
Community Support Team

[stiggrr87]

Hi
Give me 2 days, I will try out.

[cengizhanarslan]

If you configured the refresh operation throught pipeline, bu sure you re-selected "Select all" option in settings. Because when you change something in the metadata of the model the previously created pipeline does not see the new metadata until you re-select everything.

[InsightsByV]

Hi, I have come across such similar issue before.

Can you check your model for any references to the said deleted columns/tables? If they exist in unused columns/measures/steps/anywhere - it may give you such errors or warnings. Try to remove such references and try if that works!

[GilbertQ]

Hi @stiggrr87 

If you could please double check that your data source is not having any connectivity issues when refreshing the semantic model or your pipelines which could be causing the error.