Corrupted PDF won't open? What to try first
Match your error to the likeliest cause, then try the fixes in order, so you know what was recovered and what wasn't before you ask for a new copy.
A PDF that won't open is usually incomplete or has a broken internal index, not an empty file. You can try repairing the PDF with your own copy: the tool rebuilds what it can and reports how many pages it got back out of how many the file declared. Damaged PDFs are often the ones that matter, such as a thesis, a signed contract or a tax document, so the file is processed in your browser and isn't sent to any server.
Start by matching what you see to its likeliest cause.
What does your error mean?
| What you see | Likely cause | First thing to try |
|---|---|---|
| "The file is damaged and could not be repaired" | Incomplete download or broken page index | Download again, then try repair |
| "Not a valid PDF" or "does not start with %PDF" | The file isn't a PDF at all | Inspect it with File Inspector |
| Opens but some pages are blank | Damaged page content | Repair, then use "Rebuild as images" if pages stay blank |
| Won't open in one reader but opens in another | The first reader is strict | Keep the copy that opens and save a fresh one |
| Asks for a password | The file is encrypted | Unlock it first if you know the password |
Check the download before blaming the file
The most common cause is a transfer that got cut off, a flaky connection or a download interrupted halfway. Compare the file size with the one the sender shows. If it is much smaller, download it again, or ask for a resend. An email attachment opened from a preview and saved from there can also come out incomplete, so save it to a folder first.
A truncated PDF fails because the end of the file is missing, and that's where the PDF keeps the index of its pages.
Is it really a PDF?
Sometimes invoice.pdf is an HTML error page the server returned instead of the document, or an archive someone renamed. The repair tool looks for the %PDF- header anywhere in the file and, if it isn't there, tells you it isn't a PDF. To see what the file is, use File Inspector, which reads the first bytes instead of trusting the extension. The reasoning is in why the file extension can't be trusted.
What repair does, in three tiers
Repair PDF takes files up to 150 MB and works from least to most invasive:
- Structure rebuild. It reads the pages and writes a fresh cross-reference table. Text stays text.
- Complete-object salvage. If the file is cut off or scrambled, it reuses the intact objects and drops the incomplete ones. Text stays text.
- Rebuild as images. The last resort: each page is drawn and saved as an image in a new PDF. It usually recovers more pages, but the text can no longer be selected or searched, and the file may be larger.
The report names the method that worked, the pages recovered against the pages declared, and any warnings. Read them. Bookmarks, form fields and attachments are not kept, and a page with damaged content may come out blank or incomplete.
When repair can't help
Repair recovers what exists. If a page's content is truly gone, there's no way to invent it. Then your best option is another copy: the original email, your cloud storage, a backup, or the person or system that created the file. If the tool says no page could be recovered, that's the answer to take seriously.
Habits that prevent a repeat
- Check the file size after a long download or a transfer on a poor connection.
- Save attachments to a folder before opening them, and avoid editing straight from a mail preview.
- Keep a copy of important documents before running any tool on them, this one included.
- If you repair a file and then share it, review what data a PDF can hide, since the rebuilt file may still carry the original's metadata.
Frequently asked questions
Why does my PDF say it's damaged and can't be repaired?
Its internal structure is broken or a piece is missing, typically after an incomplete download or transfer. One reader may give up even though some content survives. Try another reader and a repair tool, which rebuilds whatever is recoverable.
Can a corrupted PDF be recovered 100%?
It depends on the damage. If only the page index broke, usually everything comes back. If parts of the content are gone, those pages can't be invented. The tool reports how many pages it recovered and which it couldn't.
What is lost when it rebuilds as images?
Selectable, searchable text, because each page becomes an image, and the file can grow. It's the last resort, and the tool warns you about it.
Are bookmarks and forms kept?
No. When the PDF is rebuilt, only the pages are recovered. Bookmarks, form fields and attachments are dropped, and the report says so.
Is it safe to repair a confidential PDF here?
The file is repaired in your browser and never sent to a server. You can check the Network tab in developer tools: analytics requests show up, but no POST or PUT the size of your file.
The tool says my file "is not a PDF". Now what?
It means the %PDF- header isn't in the file. It may have downloaded wrong or be another file type with a .pdf extension. Look at it with File Inspector and, if it's an error page, download the document again from the source.
Try repairing your PDF
It's rebuilt in your browser and you get a report of recovered pages and losses. Free, no signup.
Open Repair PDF →Related tools
- Repair PDF: rebuild a damaged PDF and see what was lost.
- File Inspector: check what the file really is, beyond its extension.
- PDF Reader: try the file in another reader before repairing.
- Unlock PDF: remove the password from an encrypted PDF you know.
You might also like: open a PDF without Adobe, can a PDF contain a virus? and PDF tools that don't upload files: a field guide.