EPUB Won't Open — How to Fix It

The fastest fix: run the file through the free EPUB repair tool — it pinpoints the exact fault (corrupt ZIP, missing mimetype, broken OPF, duplicate IDs) and auto-fixes mimetype problems and a missing language when the package can be read. Below are the manual diagnoses for everything else.

An EPUB that won't open almost always fails for a specific, diagnosable reason. Work through the file, package and reader checks below to narrow down the problem.

Step 1: Is the file actually an EPUB?

1. Check the extension — should end in .epub, not .epub.zip, .epub.part, .epub.crdownload, or similar.
2. Check that the download completed. File size alone cannot prove an EPUB is valid; compare it with the source download if that size is available.
3. Rename a copy to .zip and try to open it. An EPUB is a ZIP — if the ZIP doesn't open, the file itself is corrupt. Re-download from the source.

Step 2: Run a structural check

Upload the file to the EPUB validator. It runs a structural scan — the same error classes that most often cause store rejections — and returns the specific problem. Typical errors and what they mean:

"Mimetype file missing or invalid"

The EPUB spec requires the first file in the ZIP to be named exactly mimetype, containing only the text application/epub+zip, stored uncompressed. If someone recreated the EPUB with a generic ZIP tool, this gets broken. Auto-fix can repair missing or incorrect mimetype text and its entry position. If ZIP compression is the only mimetype defect, repack it locally as STORE; auto-fix does not repair that case alone.

"Container.xml missing or malformed"

META-INF/container.xml points the reader to the OPF package. Epublys cannot regenerate a missing container. Restore it from the original, re-export the book, or use a local EPUB editor to point it at the correct package before checking the book again.

"Package file (.opf) has invalid XML"

The OPF is the manifest of everything in the EPUB. Invalid XML (unclosed tags, bad attributes, encoding mismatch) means the reader can't parse it. Auto-fix can't repair invalid OPF XML — this one needs a manual fix. Open in Sigil, whose OPF editor catches errors on save.

"Manifest references missing files"

The OPF lists files (chapters, CSS, images) that aren't actually in the ZIP. Usually from a broken export. Either re-export from the source, or edit the OPF to remove the missing entries.

"Duplicate IDs" / "Invalid XHTML"

Common after merging or programmatic export. The OPF may have duplicate manifest IDs, or a chapter's XHTML may contain unclosed tags. Deep scan reports malformed XHTML and package checks flag duplicate manifest IDs; EPUBCheck also checks duplicate IDs inside chapter documents. These need manual repair in an EPUB editor.

Step 3: Test in multiple readers to isolate strictness

Different readers apply different strictness levels. If the file opens in some but not others, the file has a real issue that strict readers reject.

• Calibre — most lenient. If it fails here, the ZIP itself is probably corrupt.
• Adobe Digital Editions — lenient. Passes most non-strict files.
• Apple Books — middle. Rejects many structural issues.
• Thorium / Kobo / EPUBCheck — strict. Rejects anything the validator flags.

Step 4: Is it DRM?

If the EPUB came from a store (Kobo, B&N, library via OverDrive) and opens only in that store's app or a DRM-capable reader like Adobe Digital Editions, it's DRM-locked. Generic readers (Calibre without plugins, Apple Books, Thorium) will reject it. Open in the ecosystem that issued the license, or use the store's dedicated app.

If you bought from a store and the DRM is broken (license expired, device de-authorized), contact the store to re-issue the license. DRM removal is legally gray in most countries and out of scope for this guide.

Step 5: Still broken? Last resorts

• Re-download from the source. Especially if the download was interrupted. Many downloads look fine but are subtly truncated.
• Rebuild with Calibre. Import the broken EPUB, convert to EPUB (yes, EPUB-to-EPUB), and export. Calibre rebuilds the package from scratch and often repairs structural damage in the process.
• Open in Sigil. Sigil's editor tolerates more damage than end-user readers and can often show you what's wrong — missing files, malformed tags, wrong encoding.
• Accept it might be dead. Some EPUBs are genuinely unrecoverable (ZIP corruption, critical files missing from the original). If the source is replaceable, replace it.

Related

• EPUB Validator — the first stop for any broken EPUB
• How to fix EPUB errors — deeper dive on common errors
• What is an EPUB file? — the structure explained
• Accessible EPUB — write files that pass validation from the start