Delete, restore and clean up revisions of a dataset

This page applies to customer sites. Each time you import or update a dataset, Koordinates records a new revision. You can see them on the History tab of the dataset’s data sheet (see Learn more about a dataset). This page explains what you can remove yourself and what you can’t.

What you can do yourself

I want to	Do this
Throw away an update I haven’t published	Delete a draft
Clear an update that failed	Delete an errored draft
Remove a dataset that failed its first import	Delete dataset
Stop an import that’s running	Cancel import
Remove a dataset I no longer want	Delete the dataset (permanent)
Hide a dataset for now	Change its access permissions or make it private

Delete a draft

Open the dataset in the Data Manager, select Options, then Delete draft. The published version stays as it is. Koordinates asks you to confirm, and this can’t be undone. See Manage data updates and QA using drafts.

Delete an errored draft or a failed first import

A dataset with a failed update shows Draft error, and a banner says an errored draft can block updates and stop updated data from being published. Select Delete errored draft. If the first import failed, the dataset shows Import error and the banner offers Delete dataset. Then try again with corrected data. See Troubleshoot an import or update that is slow, stuck or failed.

Cancel an import

While a dataset is importing or updating, select Cancel import. The import stops immediately and can’t be undone. If it was the first import, the dataset is deleted.

Delete a dataset

Caution

Deleting a dataset is permanent. The dataset page then says it’s no longer available. There’s no Restore option in the Data Manager.

Open the dataset, select Options, then Delete, and confirm. See Delete published data. If you’re not sure you want to delete it, hide it instead by making it private or changing its access permissions.

What you can’t do yourself

• Restore a deleted dataset. The Data Manager has no restore action. If you deleted a dataset by mistake, contact support as soon as you can, and give the dataset’s name and ID. Whether it can be recovered isn’t something this page can promise.
• Delete an earlier published revision. You can delete a draft, but not a revision that’s already published. Revisions stay on the data sheet’s History tab.
• Roll back to an earlier revision. There’s no rollback button.

Go back to earlier data

To make a dataset show earlier data again, update it with that data:

1. Find the earlier data

   Use the original files, or the source the data came from. The data sheet’s History tab shows when each revision was published and its Revision Message, which helps you find the right version of your own files.

2. Update the dataset

   Update the dataset with those files. Save it as a draft first if you want to check it.

3. Publish the update

   The result is a new revision. It doesn’t remove the revisions between.

For raster and grid datasets, the update page lets you keep or clear the existing tiles. See Prepare and update raster and grid layers.

Clean up storage

For raster and grid layers, tiles that have been replaced can stay in storage. To ask about storage on your site, contact support.

Point cloud and Kart repositories

Point clouds and other Kart data keep their history as commits in a repository. You can delete branches and repositories, but this can’t be undone. See Kart repositories.