How to fix a corrupted PDF
A PDF that will not open is not automatically corrupted: the download, browser, file association, password or PDF reader may be the real problem. Start with the checks below, or use the broader PDF opening and viewing guide if other PDFs also fail. If the same local file fails in multiple up-to-date readers, recovery depends on how much of its original structure remains, and no reader can guarantee that a severely damaged file can be restored.
A corrupted PDF has damaged structure, not just an opening error
A healthy PDF is more than a collection of visible pages. It contains numbered objects for pages, text, images, fonts, forms and other resources, plus information that tells a reader where those objects are located and how they relate to one another.
A reader uses this structure to assemble the document. If essential data is missing, cut off or internally inconsistent, the reader may display an error instead of a page. Some readers can reconstruct a damaged cross-reference table by scanning for surviving objects, but that recovery is heuristic: it can work on one file and fail on another with a similar error message.
Damage inside the file can prevent its pages from being found
True corruption usually means that some of the bytes needed to interpret the PDF are absent or incorrect. Common causes include:
- An interrupted download or transfer. The file may have a .pdf extension even though only part of it reached your device.
- A failed save or export. An application can stop before it finishes writing the cross-reference information and trailer at the end of the file.
- Truncation during storage or copying. The beginning of the document may survive while later objects or the final file structure are missing.
- Damaged object data. A page, image, font or content stream may be incomplete or contain values the reader cannot parse.
- Incorrect PDF generation. The program that created the document may have written invalid offsets, malformed objects or an incomplete page tree.
- Physical or file-system trouble. A failing storage device, damaged archive or unsuccessful data recovery can alter the stored bytes.
The result depends on which bytes were lost. A file with a damaged cross-reference table may still contain most of its pages, allowing a tolerant parser to rebuild enough structure to open it. A truncated file may lose everything written after the cut point. If the missing portion contains the only copy of a page’s text or images, reconstructing the index cannot bring that content back.
This is why “repair” can mean different things. A program may correct references and write a structurally cleaner copy, but it cannot recreate document content that no longer exists in the file.
Viewer, browser and file problems can look like corruption
Many opening failures happen even when the PDF itself is intact. Treat these as separate possibilities before attempting recovery.
| Symptom | More likely explanation | What it means |
|---|---|---|
| The PDF opens in one reader but not another | Reader compatibility, configuration or security handling | The file is not necessarily corrupt |
| The web version fails, but the downloaded copy opens | Browser viewer, cache or incomplete web delivery | Use the complete local copy |
| Every PDF fails in the same application | Damaged or outdated reader installation | Repair, update or replace the application |
| The file requests a password | Encryption or access control | A password-protected PDF is not corrupt merely because access is denied |
| The filename ends in .pdf , but no reader recognizes it | Wrong extension or non-PDF download | Renaming a file does not convert its format |
| The file is zero bytes or much smaller than expected | Empty or incomplete transfer | Obtain another copy from the source |
| One page fails while other pages display | Damage in a page object, image, font or content stream | Partial recovery may be possible |
| The file fails in several independent readers | Structural damage, incomplete data or a mislabeled file | Investigate the source and file structure |
Test the same saved file to identify the real failure
Diagnose the file systematically before editing, converting or uploading it. Work on a copy so the original remains unchanged in case a later recovery tool needs it.
- Save the PDF to your device. If you are viewing it on a website or in an email preview, download it instead of repeatedly opening the embedded version. Wait for the transfer to finish before testing it.
- Check the file’s name, extension and size. Confirm that the name ends in .pdf and that the file is not empty. If the sender, website or backup lists an expected size, compare it with the exact byte size of your copy. A large difference indicates that the transfer may be incomplete.
- Download or copy it again. Use the original website, email attachment, shared drive or storage device rather than duplicating the suspect local file. If the second copy has a different size, the first transfer was unreliable.
- Try another known-good PDF. Open an unrelated PDF in the same application. If that file also fails, troubleshoot the reader or browser rather than attempting to repair both documents.
- Open the suspect file in another current reader. Test the same saved copy, not a new download, so you are comparing applications rather than files. A PDF that opens correctly in another reader has usable structure, even if one application rejects it.
- Try a desktop reader if the browser fails. Browser PDF viewers may behave differently from dedicated readers. Downloading the document and opening it locally separates a website or browser problem from a file problem.
- Check for a password or security message. A password prompt means the reader recognized an encrypted PDF. Obtain the correct password from the document owner; corruption repair will not supply missing credentials.
- Ask the source to open its original copy. If the sender can open the original but your downloaded attachment fails, request a new attachment, a fresh share link or a newly exported PDF. If the sender’s original also fails, the creation or save process probably produced the damage.
- Compare available copies and backups. Check email history, cloud version history, synchronized folders, previous exports and backup systems. An older complete version is usually more valuable than an uncertain automated repair.
- Record how far the file opens. Note whether you can see the first page, thumbnails, page count, document properties or only an error. Partial access shows that at least some document objects remain readable and may be extracted.
Technical users can also inspect a copy with qpdf:
qpdf --check suspect.pdf
Do not rely on an error message alone to declare a PDF corrupt. The strongest practical evidence is that a newly obtained local copy fails in multiple independent, up-to-date readers while other PDFs open normally in those same readers.
Choose a recovery path based on what still works
Preserve the original file before trying any fix. Create a duplicate with a clear name such as report-damaged-original.pdf, then direct every recovery attempt to a different output file. Repeatedly overwriting the only copy can remove evidence or surviving objects that another tool might have used.
Fix the reader or browser when another application opens the PDF
If the document opens correctly in one reader, its core structure is usable. Focus on the application that fails:
- Update the failing reader to its current version.
- Close the application completely and reopen it.
- Repair or reinstall the desktop reader if other PDFs also fail.
- Download web-hosted PDFs instead of using the browser preview.
- Clear the browser cache if the website repeatedly serves an incomplete copy.
- Check that .pdf files are associated with a PDF reader rather than an unrelated application.
- Remove conflicting or obsolete PDF plug-ins if the failure occurs only inside a browser.
Opening the file in the Soda PDF Reader is a useful comparison test. If it displays there, use that result to distinguish a viewer-specific failure from severe structural damage. Successful viewing does not mean a different application’s installation has been repaired, and a failed upload or opening attempt does not by itself prove that Soda can restore missing file structure.
Replace an incomplete copy rather than repairing it
If a repeated download produces a larger file, or the sender’s original opens while your copy does not, discard the incomplete copy after preserving it for comparison. Obtain the document again through a more reliable route.
Ask the sender to attach the original file again rather than forwarding the already damaged attachment. For files produced by office, design or accounting software, ask for a fresh PDF export from the editable source. If the file came from a web portal, sign in again and generate a new download instead of reusing an old temporary link.
A fresh export is usually the best recovery because it recreates every object, page and reference from the source document. It avoids guessing which parts of the damaged PDF were supposed to be present.
Changing the filename or replacing the .pdf extension does not repair an incomplete file. Likewise, changing an unknown file’s extension to .pdf does not turn it into a PDF.
Salvage readable pages when the PDF opens only partly
If the PDF opens far enough to show some pages, recover those pages before experimenting with deeper structural repairs. Depending on the software and the damage, you may be able to:
- Save a new copy under another filename.
- Extract the pages that still display correctly.
- Export readable pages to another format.
- Copy essential text and images into a new document.
- Print the visible page range to a new PDF as a last-resort reconstruction.
Inspect the recovered copy carefully. Confirm the page count, order, text, images, forms and links rather than assuming that a successful save preserved everything.
Printing to a new PDF can capture visible pages, but it creates a new rendered document rather than restoring the original structure. Interactive forms, embedded files, bookmarks, accessibility tags, layers, comments, links or digital-signature validity may not survive. Keep both the damaged original and the salvaged output.
Rebuild or recreate a structurally damaged PDF
When no reader can open the file, the safest path is to return to an intact source:
- Re-export the PDF from the original Word document, spreadsheet, presentation, design file or publishing application.
- Restore an earlier copy from cloud version history, email, backups or document management.
- Ask the creator or records owner for the original.
- Use a specialist structural-recovery tool only on a duplicate.
- Compare every recovered page with an authoritative source before relying on it.
Advanced users may allow qpdf to parse a damaged input and write a new output:
qpdf suspect.pdf recovered.pdf
This can help when enough objects survive for qpdf to reconstruct the cross-reference information and serialize a new file. It is not guaranteed to recover every document, and a generated output must be checked for missing pages, images, fonts and content.
A PDF is unrecoverable when the required content bytes are gone and no original, backup or alternate copy exists. No reader can infer an entire missing page from a broken reference, and no repair command can reproduce images or text that were never preserved. In that situation, recreating the document from its source material is the only dependable fix.
Source: PDF file structure, cross-reference and trailer details from the PDF Reference, Third Edition, Adobe; opening-error, download and damaged-file guidance from Why can’t I open a PDF and security guidance from Protected Mode troubleshooting, Adobe; structural checking and heuristic recovery behavior from Running qpdf, qpdf documentation; and reader capabilities from PDF Reader, Soda PDF. Page last checked 2 October 2026.