If we wanted to see what columns in our database are used in any metrics or visualizations. What would be the most efficient way to do that? The reason, we want to clean up some of our tables but don't want to break any dashboards.

Hello Doug, I am afraid that there is currently no straighforward way to get this info, however, you may use the AFM explain resource available for low-level inspection about queries that are being sent to database. It can be used also from browser, if you open some specific report using "Analyze" tab and replace the last URL part (

/edit

) with

/debug

. It will download a ZIP archive "explainAfm.zip" containing low-level information about report and its computation, including report.json.qt.opt.sql with actual query that was sent to your database. This may give you an idea of what tables the insights or GoodData metrics are querying in the database. As a last resort, you may also check your database logs for these details.

Hi Doug, our LDM UI has impact analysis feature. If you try to delete a field that is used on a dashboard, metric or visualisation, then the UI will inform you in advance. Important note: the analysis check only the current workspace. If the workspace is part of a hierarchy and some of the children workspaces use a field you are about to delete, the UI will not inform you. Check this documentation article for more details. https://www.gooddata.com/docs/cloud/model-data/evolve-your-model/monitor-model-changes/

Oh, yes I did see that. However, That seems to only work in our Parent level space In our child space (where we do add new tables that are specific for only those children) we can't use the same methodology

Thank you for the explanation. Let me share your feedback further to the product teams

🎉 New note created.