Skip to content

Running the Tools

Now that we've configured our environment, let's explore the tools!

You can find all of the tools by either visiting the documentation or by typing..

Archiver: Maintaining metadata

Archiver is a tool that helps you find and clean up stale answers and liveboards in your ThoughtSpot platform.

As the number of users grow, so does the content that they create. Even though no data from your data sources are stored within ThoughtSpot, maintaining unused references to this content still requires resources. Additionally, excess metadata can slow down the backup, restore, and cluster upgrades processes.

From the available commands, we can come up with an administrative process to help maintain a clean environment. You can use the below framework with Archiver.

Run Archiver identify to isolate inactive content.

Label the tag name as clean-up.

Communicate to your ThoughtSpot User community that the admin team will be performing a cleanup activity.

Communication to Users

In an effort to keep ThoughtSpot running smoothly, Answers and Liveboards which haven't been viewed in over 120 days will be marked for deletion.

If you wish to keep stale content, please untag your Answers with the clean-up tag.

Wait a minimum of one week before performing any cleanup actions.

Maintain an active dialogue with your User community during this time, in case they have questions or concerns.

Run Archiver remove, targeting the tag name clean-up.

Strongly consider using the --export-tml so you can restore any content that users wish they had saved during the deprecation cycle.

Try it out

Let's run the command to identify all content which hasn't been viewed within the past 3 months or created in the last 4 months. We'll use the tag clean-up to communicate to our Users that this content is targeted for removal.

We'll also ignore all of our Training content, since we haven't run a new user onboarding this quarter, and our Certified datasets, even though it's unlikely that this content hasn't been viewed in the past 90 days.. but it's better to be safe than sorry!

Using --config

Since we set our non-prod configuration as the default, usage of --config non-prod is optional here. CS Tools will attempt to use the default configuration first, but if we wanted to run Archiver against another cluster (eg. Production), we can override it with the --config option.

Find the copy button to the right of the code block. 

cs_tools tools archiver identify `
--tag clean-up `
--recent-activity 90 `
--recent-modified 120 `
--ignore-tags certified `
--ignore-tags training `
--config non-prod

Find the copy button to the right of the code block. 

cs_tools tools archiver identify \
--tag clean-up \
--recent-activity 90 \
--recent-modified 120 \
--ignore-tags certified \
--ignore-tags training \
--config non-prod

Administrative Responsibility

The maintainers of CS Tools recognize when we give you lots of power.

For this reason, with tools that perform bulk management, it's important to have fail-safes before executing the commands.

With Archiver, you'll be prompted to continue with your operation before affecting potentially hundreds of Users.

Report on Archiver operations

In highly active ThoughtSpot deployments, the identification process can find more content than the admin team is comfortable investigating directly in the web user interace.

On all three Archiver commands, you'll find a --syncer option which labels information about the type of content that we'll tag or remove, as well as who owns it. This information can be incredibly helpful to inspect and will help you communicate with those Users directly.

In the next section, we'll learn about Syncers and how you can leverage them to manage all the information that Archiver produces.