s3cab guides

The full documentation, for when you want more than --help. New to s3cab? The README has the install and a quick start; start there and come back.

Each of these has a terminal-sized counterpart: run s3cab help <topic> or s3cab <command> --help for what you need mid-task without reaching for a browser. These pages are the sit-down read.

Everyday

• Backing up — how s3cab decides a file hasn't changed, the rare edits that fool it, and when to re-read everything with --rehash.
• Getting files back — recovering a deleted file, an older version, or everything onto a fresh machine. The job the backup exists for.
• Reading a compare report — what added / renamed / moved / modified / deleted mean, and why a "rotated" file reads as modified.
• Finding a file in your backups — which snapshots hold a file and what it hashed to; how a find pattern differs from an exclude pattern, and what the warnings mean.
• Exclude rules — the glob patterns that keep files out of a backup set. (Quick reference: s3cab help exclude.)
• Looking after a backup — checking backups are still restorable (verify), dropping snapshots (forget), reclaiming storage (cleanup), and why bucket versioning makes all of it safe to get wrong.

Setting up

• Setting up a cloud bucket — standing up an AWS bucket and a locked-down identity, the versioning/lifecycle model, keyless Roles Anywhere access, and non-AWS providers (R2, B2, Wasabi, …).
• Signing in to your storage — the four ways a set signs in, where s3cab looks for credentials, and what to do when the server rejects them. (Quick reference: s3cab help provider.)

Reference

• The format spec — the repository layout, the snapshot-file grammar, and recovering by hand with no s3cab at all. The stored format is a promise, and this is it written down.
• Output formats — human-readable text by default, --json for scripts.