When a PDF refuses to open, the message on screen usually reads like an apology rather than an explanation. "The file is damaged and could not be repaired" tells you almost nothing about what is damaged, or whether the document inside is still recoverable.
It is more informative than it looks. Each reader validates a PDF in a different order and gives up at a different point, so the wording of the failure narrows down which part of the file is at fault — the header, the index, an individual object, or the encryption layer. This guide maps the messages you are likely to see back to the fault that produced them.
Why the same file gives different errors
A PDF stores its pages as numbered objects, and records the byte position of every object in a lookup table at the end of the file called the cross-reference table, or xref. Opening a PDF means jumping to the end, reading that table, and using it to locate page one. Almost everything that goes wrong goes wrong in that sequence.
Readers differ in how strictly they follow it. Adobe Acrobat validates the structure properly and refuses a file that fails; Chrome and Firefox scan the file for anything page-shaped and quietly work around a broken index. That difference is why one file can produce four different messages on four different machines, and it is why the reader you happen to be using changes the wording but not the underlying fault.
Adobe Acrobat and Reader errors
Acrobat produces the most specific messages of any reader, because it is the only one that checks the file against the specification rather than trying to be helpful. These are the ones that come up most often:
| Message | What has actually failed | What to do |
|---|---|---|
| "There was an error opening this document. The file is damaged and could not be repaired." | Acrobat's own recovery pass ran on a broken xref and failed | Repair with a tool that rebuilds the page tree from scratch |
| "The file is either not a supported file type or has been damaged." | The file does not begin with %PDF-, or is not a PDF at all | Check the first bytes before repairing (see below) |
| "Expected a dict object." / "Invalid object number." | One or more object streams inside the file are malformed | Repairable, but damaged pages may lose formatting |
| "There was a problem reading this document (14)." | A font or embedded resource cannot be decoded | Repair; standard fonts are substituted for broken ones |
| "There was a problem reading this document (109)." | Damaged content stream — often a file that was edited while syncing | Repair recovers undamaged pages |
| "This document is protected. Enter the password." | Nothing is damaged — the file is encrypted | Not a repair problem; you need the password |
| "Access denied." on a network drive | A file permission or lock issue, not the PDF | Copy the file locally and open the copy |
The numbered errors — 14, 109, 110, 131 — are worth treating as a family. They all mean Acrobat parsed enough of the file to start reading a specific object and then found something it could not interpret. In practice they behave identically: the structure is rebuildable, and how much of the document survives depends on how many objects were hit.
Chrome, Edge and Firefox errors
Browsers use their own PDF engines and are dramatically more forgiving, so a browser error usually means something more serious than an Acrobat error does. Their messages are also vaguer, because the engine is not trying to diagnose the file — it is trying to render it and giving up.
| Message | What has actually failed | What to do |
|---|---|---|
| "Failed to load PDF document." | The engine could not parse the header or find any usable page objects | Genuine structural damage — repair it |
| "Couldn't preview file" / endlessly spinning viewer | The file is truncated; the engine is waiting for bytes that never arrive | Re-download first, then repair if it persists |
| Blank grey pane, no message at all | The document opened but nothing rendered — usually fonts or layers | A rendering problem, not an opening problem |
| Downloads instead of opening | The server sent the wrong content type, or the file is not a PDF | Check the file's first bytes |
| "This file cannot be opened because of an unexpected error." | Edge-specific wrapper around a parse failure | Treat as "Failed to load" |
The spinning-viewer case is the one people most often misread. A viewer that never finishes loading is not slow — it is usually waiting on a file that stops partway through. Compare the file size against the original source before concluding the file is corrupt; a download that reached 80% typically yields roughly 80% of the pages, intact, once you rebuild its index with Repair PDF.
Preview, mobile and email-app errors
Apple Preview, iOS Files, Android viewers and the in-app previewers inside Gmail, Outlook and WhatsApp are the least informative of all. They tend to produce a single generic message regardless of cause, which makes them poor diagnostics but useful in one specific way: they fail differently from desktop readers, so what they can and cannot do narrows the problem down.
- "The file appears to be damaged" (Preview). Preview is comparatively strict. It refuses files with a broken xref that Chrome would open, so this message alone does not mean the file is unrecoverable.
- "Cannot display PDF" inside Gmail or Outlook. The web previewer has size and complexity limits that the file exceeded. Download the attachment and open it locally before assuming anything is wrong with it.
- The file opens on desktop but not on a phone. Almost always memory, not corruption. Mobile readers give up on very large or image-heavy documents that a laptop handles without complaint.
- Opens, then closes immediately. The reader crashed rather than rejected the file — the same memory ceiling, hit a moment later.
- "Unsupported file format" on Android. Usually a file that is not a PDF despite its extension, or one that was renamed after download.
Four failures that are not corruption
A meaningful share of files brought to a repair tool are not damaged at all. Ruling these out takes a minute and saves you from applying the wrong fix:
- It is encrypted, not broken. A password prompt is not an error. If you know the password, the file is fine; if the document is yours and you want to change or remove that protection, use Unlock PDF rather than a repair tool.
- It is not a PDF. Open the file in any plain text editor and read the first line. A real PDF starts with
%PDF-1.4or similar.PKmeans it is a ZIP or an Office document;<!DOCTYPE html>means you downloaded an error page instead of the file. - It is 0 bytes. Check the size in your file manager. An empty file contains nothing to recover, and no tool can change that — go back to the source, the email attachment, or a backup.
- Your reader is too old. PDF 2.0 files using AES-256 encryption fail to open in readers released before roughly 2017, usually with a misleading "damaged file" message. Updating the reader fixes it outright.
Fix a PDF that will not open, in three steps
Once you have ruled out the four cases above, the file is genuinely damaged and needs its structure rebuilt. That takes three steps in the browser, with no installation and no sign-up.
Upload the file exactly as it is
Go to pdfdocshift.com/repair-pdf and drop the file onto the upload area. Do not rename it, re-save it, or open and re-export it first — each of those can overwrite structure that is still recoverable.
Let the repair pass rebuild the index
The tool ignores the existing cross-reference table entirely, scans the whole file for valid PDF objects, reconstructs the page tree from what it finds, and writes a fresh index. Unreadable embedded fonts are replaced with standard equivalents so the text renders.
Verify more than page one
Download the result and check the page count against what you expect, then open a page from the middle and the last page. If pages are missing rather than malformed, the file was truncated and what you have is everything that survived.
Repair a PDF that will not open
Free, no sign-up, files deleted after 2 hours. Most files are rebuilt in under 10 seconds.
Open Repair PDF →If the repaired file opens but the pages look empty, the problem was never the index — see our guide to why a PDF opens blank, which covers fonts, layers and form documents. If it opens fine but is unusably slow, the file is too large rather than broken, and Compress PDF is the right tool.
Stopping it happening again
Nearly all real-world PDF damage traces back to a handful of habits, and most of them are avoidable once you know what they are:
- Let cloud sync finish before opening a file. Saving over a PDF that Dropbox, OneDrive or Google Drive is mid-upload reliably produces a file that is half old and half new — the single most common cause of "it worked yesterday".
- Verify size after large transfers. Interrupted downloads and network copies are the main source of truncated files, and a size comparison catches them immediately.
- Eject removable media properly. Pulling a USB drive before the write completes leaves the tail of the file unwritten, which is exactly the truncation case.
- Zip PDFs before sending through unfamiliar mail systems. A small number of legacy servers still apply text-mode encoding to binary attachments, corrupting them in transit.
- Keep the damaged original. Repair writes a new file rather than modifying the one you uploaded, but people routinely delete the broken copy once a repair looks successful. If pages turn out to be missing, that original is the only thing a second attempt can work from.
Frequently asked questions
Because they validate differently. Chrome scans the file for anything page-shaped and renders what it finds, working around a broken cross-reference table without telling you. Acrobat checks the structure against the PDF specification and refuses files that fail it. If a file opens in a browser, the page content is intact and only the index is broken, which is the most repairable form of damage there is.
It means Acrobat detected a structural problem, ran its own built-in recovery pass, and that pass did not succeed. It does not mean the document is unrecoverable. Acrobat's recovery is conservative and works from the existing index; a tool that ignores the index entirely and rebuilds the page tree by scanning the file for objects frequently succeeds where Acrobat gives up.
Almost certainly not. Corruption is consistent — a damaged file fails everywhere. A file that opens on one device and not another is hitting a resource limit, usually memory on a large or image-heavy document, or a previewer size cap inside an email app. Compressing the file or splitting it into smaller documents fixes it; repairing it will not.
Open it in any plain text editor and look at the first line. A genuine PDF begins with the characters %PDF- followed by a version number. If the file starts with PK it is a ZIP archive or an Office document with the wrong extension. If it starts with you downloaded an error page rather than the document, and the fix is to fetch it again from the source.
No, and confusing the two wastes a lot of time. A password prompt means the file is intact and encrypted. Some readers word this poorly enough that it reads like a failure, but there is nothing to repair. If the document is yours and you have the password, Unlock PDF removes the protection; if you do not have the password, no repair tool can recover it.
Usually not. A repair rebuilds the file's index and page tree while copying the page content across unchanged. The one visible exception is fonts: when an embedded font cannot be decoded, the repair substitutes a standard font, which can shift line breaks or spacing slightly. The words themselves, and the page order, are unaffected.
Ready to open your file again?
Free, no sign-up, files deleted after 2 hours. Works on every device, with no installation.
Open Repair PDF →