Back to Reference
Contents
Reference
422 words
Jun 25, 2026

Heratio Help Center article. Category: Digital Preservation.

Web archiving and WARC (what Heratio does and does not do)

What web archiving is

Web archiving captures websites - their pages, images, stylesheets, scripts, and embedded media - and preserves them so they can still be viewed after the live site changes or disappears. The standard container is WARC (Web ARChive format, ISO 28500), which records each captured HTTP request and response so a replay tool can rebuild the page exactly as it was served. Common capture tools are Heritrix, Browsertrix / Webrecorder, and wget's WARC mode.

What Heratio does today

Heratio implements a bounded, verifiable slice of WARC web archiving at Admin → Web archive (/admin/web-archive). One admin surface offers two capture modes, and every capture can be replayed back from its stored WARC inside Heratio:

  • Archive a URL. Snapshot any public http/https page into a WARC 1.1 file (single-page capture).
  • Capture a record page. Snapshot a published record's OWN public page on this host, plus that page's direct same-host subresources (CSS, JavaScript, images, icons), into one WARC 1.1 file.
  • Replay. Each successful capture replays back entirely from the stored WARC, never from the live site, with an "Archived snapshot" banner and a restrictive content-security policy. Captured same-host subresources are rewritten to load from the archive; nothing live is fetched.

It is SSRF-safe (record mode fetches only the record's own page + same-host assets; URL mode refuses internal / loopback / metadata / credentialed URLs), bounded (timeouts, redirect cap, size caps, subresource cap), and writes only the warc_capture register table plus the .warc files under the configured storage path. For the full user guide see the "Web archive (WARC)" help article.

Heratio's ingest and capture pipelines additionally treat any externally-produced WARC deposited as a plain file (store, checksum, PUID-identify), the way they treat any file.

Honest gap

The capture archives the record's own page plus its direct same-host subresources (record mode) or a single page (URL mode). It does not fetch off-host assets (third-party CDNs, web fonts), so those do not replay, and subresource discovery is depth-1. Neither mode is a general site crawler.

If you need full-site web archiving now

For full-site or subresource-faithful archiving, use a dedicated tool to capture the site to WARC - for example Browsertrix / Webrecorder, the Internet Archive's Archive-It, or Heritrix with pywb for replay. You may then deposit the resulting WARC file into Heratio as a preserved object (so it gets fixity and a PREMIS history).

See also

  • Digital preservation overview (help)
  • OAIS packages: SIP, AIP, DIP (help)