All Collections
Data Warehouse Sharing
Google BigQuery Connector
Google BigQuery data sharing - Troubleshooting
Google BigQuery data sharing - Troubleshooting
Niclas Bångman avatar
Written by Niclas Bångman
Updated over a week ago

Overview

Here we list some possible errors and warnings you might see in your BigQuery Share and how to solve them. Errors that are common between all Data Warehouse Shares can be found here:

Error message explanations

"View creation failed"

If you get the warning below when looking at a sharing job in Funnel then you have tables in your dataset with mismatching definitions.

view creation failed in Funnel

This error message means that all your data has been shared but no view has been created. The view creation is optional and can be selected in your Share configuration.

Create BigQuery view selection in Funnel

If you still want Funnel to create and update the view for you then this issue is easily solved by either manually editing/removing the old tables or (re-)sharing the period that causes the issue.

For example, if you add a new field to the Share and your Share only includes the last 3 months but you have 6 months of data in your dataset. Then you can temporarily reconfigure it to share 6 months, untick "Only share periods with updated data" and then press "Run job" to re-share all your data and get a working BigQuery view.

"Two or more BigQuery tables, not using date partitioning, have identical names"

This means that you have tables in your BigQuery with the same dataset and table name as in the Share but with different date partitioning settings.

For example, if you have date partition enabled in your Share but you share to an existing BigQuery table that does not have it enabled.

To solve this issue you can either

  • Delete the old tables and run the Share again

    Make sure the Share date range covers the entire range you want and "Only share periods with updated data" is turned off for the first share after deleting the tables to ensure you get data all the data you want.

  • Change the file output template so that you share your data to different tables

  • Change the table partitioning in the Funnel Share to match the existing tables in BigQuery

"This field X and maybe more fields in this export are using the wrong export format. Fix the field formats and run the export again."

This error message means that there is a difference between what type a field is configured as in Funnel (can be seen in the dimensions/metrics tab) and what type it is configured as in the Funnel Share (can be seen in the "Edit field format" on the Share edit page).

The field could have changed type in Funnel, see the example below.

For example, if you change the type of a field in Funnel (eg. in the dimensions/metrics tab if it's a custom field) the type will no longer match the type stored in the Share configuration. This is because the types are not automatically updated in the Share to avoid accidentally breaking Shares.

To solve this open the "Edit field format" modal on the edit page for the Share and find the field with the warning and press "Update field type".

"The billing account for the owning project is disabled"

You need to enable billing on your Google project, see this Google article.

If you don't see this warning the metric value could include unexpected values, please check the data explorer and/or contact support.

Data missing because of table expiration

If you manually set a table expiration or if you created your dataset in the BigQuery sandbox you will see the "Default table expiration" being set. This means that all tables in this dataset will default to deleting data older than 60 days (in this case).

To solve this make sure billing is enabled and create a new dataset without setting table expiration.

Permission issues

Make sure that you have:

  • given Funnel's service account the BigQuery User permission to the project

  • given Funnel's service account the BigQuery Data Owner permission to the dataset if you created the dataset yourself

Did this answer your question?