Database backups and recovery

Check method

With a read-only key: presence and age of the last restore point and whether point-in-time recovery is on.

Possible evidence type

Code or configuration review, without exploiting the risk.

Access and limitations

Connected repository. Actual results and checks not run appear in your report. A check with no finding does not guarantee the absence of risk.

Concrete scenario

The project assumes backups exist but has no evidence of a successful recovery.

What to inspect

Distinguish configured backups, retained copies and a tested restore. Automated inspection may not have provider access to verify them.

Remediation approach

Document retention and recovery ownership and schedule a safe restore exercise outside production.

Understand grants, policies and identity

Database grants determine whether a role can reach a table or function; row-level security determines which rows it can access. Enabling RLS without a usable policy can block legitimate users, while a broad policy can expose every tenant. A privileged server credential may bypass these protections and must never be shipped to the browser.

1. Identify the role making the request and the identity used by the policy.
2. Review both table grants and policies for each read or write operation.
3. Check owner access to private, draft and archived rows separately from public access.

Validate without changing production data

Catalog-only inspection can show policy definitions and privilege configuration, but does not prove every application-level access path. Two-account isolation exercises provide stronger behavioral evidence when separately authorized. Backup availability is also different from a successful restore test.

1. Review definitions and grants in the supported read-only audit.
2. Use isolated test records for any owner-approved behavioral exercise.
3. Retest the expected owner access and denied cross-account access after changes.

References

OWASP A05