Observability Permissions¶
Observability sits on top of two permission layers: the workspace-level user role that gates the endpoints, and the datastore-level team permission that gates the underlying containers and checks.
For the full reference of user roles and how they compare, see User Roles. For the canonical matrix of every team-permission-gated action across the platform, see Team Permissions Overview.
Admin bypass
Users with the Admin role bypass every team-permission check on every datastore.
Permission Matrix¶
The table below shows which team permissions can perform each action on a datastore the user has access to:
Key Details¶
- Reading is available to every team permission: the listing endpoint requires a valid workspace user, and the team permission gates only the underlying datastore, container, and check reads.
- Threshold and Maximum Age edits modify the underlying quality check and require the Author team permission (Editor is always allowed as well).
- Turning Volume Tracking or Freshness Tracking on or off changes the container settings and requires the Editor team permission on the datastore. Authors can edit the observability checks but cannot turn tracking on or off from the card.
- Favorites are a per-user preference. Any user with read access to the Observability tab can mark and unmark them, and the marks are private to that user.
Controls follow your team permission
The Edit Threshold, Edit Maximum Age, and edit check controls appear only for users with the Author or Editor team permission. Reporter, Viewer, and Drafter users see the charts and checks without them. The Enable button on an untracked container appears only for users with the Editor team permission. The Observability action in the container actions menu, which opens the Observability Settings modal, appears for users with the Editor team permission and for an Author who owns a computed container.
See Also¶
-
How It Works
When volume, freshness, and metric values are measured, where else they come from, and what each chart, color, and tooltip shows.
-
Examples
Reading the charts in real situations: a value past its limits with no anomaly, a missing day, and a container that became Inaccessible.
-
Best Practices
Guidelines for turning tracking on early, pairing charts with scheduled Scans, and reading the heatmap.