S3 backups: versioning, retention, and restore tests
Design a version-aware object-storage workflow with a recovery manifest, deliberate cleanup rules, and tested retrieval permissions.
Read the 6-minute guideCloud & storage
Cloudflare Insta Backups
Map DNS, deployments, and stored objects.
Cloudflare-related recovery may involve several independent resources: DNS records, website deployment artifacts, application configuration, or data in R2. Identify the actual product and source of truth before choosing a backup. A cache or network delivery layer should not be assumed to contain everything needed to rebuild the origin.
What to preserve
- DNS exports and the account or domain that gives them context.
- Original deployment source, build output, and configuration.
- R2 objects, manifests, and product-specific retention requirements.
Your starting workflow
Export DNS through the supported dashboard or API route and keep the original site artifacts separately. Cloudflare documents DNS import and export, but do not treat a zone file as a copy of every account setting. Inventory other resources and preserve their supported configuration sources.
Test the recovered copy
Inspect a DNS export against the expected record inventory without applying it to production. Test the website on an isolated host. For R2, test the specific upload, retrieval, and retention operations your backup tool relies on instead of inheriting assumptions from Amazon S3.
Does exporting DNS back up my website?
No. DNS describes how names resolve; it does not contain the site’s pages, assets, database, or entire account configuration. Keep those items through their own capture methods and connect them in the recovery runbook.
From Insta Backups Lab
Put the plan into practice.