How to Repair a Corrupt PDF

“There was an error opening this document. The file is damaged and could not be repaired.” It is a bad sentence to read at 11pm the night before a deadline, and it is more often survivable than it sounds.

Many PDFs that refuse to open are not really destroyed. Their contents are intact and only the index that tells a reader where everything lives is broken. Rebuilding that index brings the document back.

What actually breaks inside a PDF

A PDF is a collection of numbered objects — pages, fonts, images, metadata — followed by a cross-reference table, the index recording the exact byte offset of every object. At the very end sits a trailer pointing at that table.

A reader opens a PDF by jumping to the end, reading the trailer, then using the cross-reference table to find whatever it needs. Efficient — and fragile. If those byte offsets stop being accurate, the reader cannot find object one, and gives up before it ever discovers that your text is sitting there perfectly intact.

Offsets go wrong in ordinary, unglamorous ways:

• a download or transfer interrupted before the last bytes arrived
• a USB stick pulled out mid-write, or a laptop that lost power while saving
• a program that crashed while generating the file
• a buggy PDF generator that wrote a table that never matched reality
• an email gateway or upload script that mangled the file as text instead of binary
• storage that has silently degraded over years in an archive

How repair works

The tool ignores the broken index entirely. It scans the file from the beginning, finds each object where it genuinely is, and writes out a new PDF with a correct cross-reference table and a valid trailer.

This is why it works so often: the repair does not need the damaged index to be trustworthy, because it rebuilds it from the actual contents. If your pages, fonts and images are present in the bytes, they come back.

Repair a PDF, step by step

1. Work on a copy. Always. Keep the damaged original untouched in case you want to try something else later.
2. Open the tool. Go to Repair PDF and drop the file in.
3. Rebuild. Our server reconstructs the structure with qpdf, a long-established PDF engine.
4. Download and open the result — and check it page by page before you trust it.

That last step matters. A file can open cleanly and still be missing an image that was genuinely lost. Verify, do not assume.

What repair can and cannot fix

Honesty is more useful than optimism here.

Usually recoverable: broken or missing cross-reference tables, damaged trailers, files a reader calls damaged but which are structurally complete, files written by a faulty generator, and files that a strict reader rejects while a lenient one opens.

Not recoverable: a truncated file whose end never arrived — the missing bytes do not exist anywhere, and no tool can invent them. Likewise a file overwritten by other data, or one whose compressed content streams are themselves corrupted. If the page data is gone, it is gone.

A rough signal: if the file size looks about right for the document, the odds are good. If a 40-page report is 3 KB, the content was never fully written and repair will not conjure it back.

Before you conclude the file is corrupt

Two quick checks save a lot of trouble:

• Try a different reader. Open it in Chrome, Firefox, and a desktop reader. Some are far stricter than others, and a file that fails in one opens in another. If a browser can display it, you can print it to a fresh PDF from there.
• Check for a password. A prompt for a password, or an error about permissions, is not damage — it is encryption. That needs Unlock PDF with the password, not repair.

Also check where you got the file. If it came off a failing drive, copy it somewhere safe before anything else. If it came from a download, re-downloading is faster and more reliable than any repair.

Preventing the next one

• Eject removable drives properly. Yanking a USB stick during a save is the classic cause.
• Do not edit documents directly on a network share or cloud folder that may drop connection mid-write. Copy locally, edit, copy back.
• For files served on the web, optimise for fast web view, which restructures the document so browsers can display page one before the whole file arrives.
• Keep archives verified. A yearly spot-check of important PDFs catches silent degradation while a backup still exists.

Is it private and free?

Rebuilding a PDF’s internal structure needs a real PDF engine, so this tool runs on our secure server rather than in your browser. Your file is uploaded over an encrypted HTTPS connection, processed in an isolated workspace, and deleted automatically the moment the repaired PDF is sent back— nothing stored, logged or shared. Free: no watermark, no sign-up, no daily limit, files up to 100 MB.

For a confidential document, desktop repair tools keep the file on your own machine, and that is the more cautious choice.

Frequently asked questions

Will I lose any content? Repair does not remove anything that is present. Anything already missing from the file stays missing — that is the honest limit.

The tool ran but the PDF still looks wrong. Then the damage is in the page content itself, not the index. Try a different reader, and go back to whoever or whatever produced the file for a fresh copy.

Can it fix a PDF that opens but shows blank pages? Sometimes — blank pages can come from a broken structure. Often it means the page content streams are damaged, which repair cannot rebuild.

Does it work on password-protected files? No. Remove the protection first with Unlock PDF.

My file was fine yesterday. Then suspect the storage or the transfer rather than the document. Copy it off that drive now, and check whether a backup or cloud version history has an earlier copy — that is usually the fastest recovery of all.

Try repairing it now

Open Repair PDF, upload a copy of the damaged file, and find out in seconds whether it can be rebuilt — free, watermark-free, and deleted right after.