Take the 2-minute tour ×
Super User is a question and answer site for computer enthusiasts and power users. It's 100% free, no registration required.

It appears that Word's password protection is not really good, at least until Office 2003, if I read this SU entry correctly. I'm under the impression that Acrobat's PDF password protection should be better (it says 128-bit AES for Acrobat 7 and higher). Is that true?

Of course, it depends on the strength of the password used, but assuming I protect my PDF with a password like sd8Jf+*e8fh§$fd8sHä, am I on the safe side?

Like, say, for sending confidential patient information - not really valuable, but potentially highly sensitive.

share|improve this question
2  
Just as a side note: note that PDF can also have a password that only restricts certain usage, like to forbid printing. Those passwords are much easier cracked than passwords that protect the document itself. (But that is not what this question is about.) –  Arjan Oct 16 '09 at 13:45
    
Given a few of the answers: please make clear if you want protection to avoid opening the document, or to avoid things like printing or selecting/copying text. –  Arjan Oct 16 '09 at 16:04
    
I want to protect the document from being opened without the correct password. –  Tim Pietzcker Oct 16 '09 at 19:38

9 Answers 9

up vote 6 down vote accepted

From the Adobe site - Securing documents with passwords:

The Acrobat 3 And Later option uses a low encryption level (40‑bit RC4), while the other options use a high encryption level (128‑bit RC4 or AES). Acrobat 6.0 And Later lets you enable metadata for searching. Acrobat 9.0 And Later encrypts the document using the AES encryption algorithm with a 256-bit key size.

So apparently 7 will use 128-bit AES. I'd say you're very safe, especially with a password like that. The National Institute of Standards and Technology agrees:

Assuming that one could build a machine that could recover a DES key in a second (i.e., try 255 keys per second), then it would take that machine approximately 149 thousand-billion (149 trillion) years to crack a 128-bit AES key.

share|improve this answer

Of course, it depends on the strength of the password used, but assuming I protect my PDF with a password like sd8Jf+*e8fh§$fd8sHä, am I on the safe side?

With such a password your documents will be pretty much well protected. Especially under Acrobat 7 and 8.

Under Acrobat 9, Adobe made changes to the underlying algorithm. And while they upgraded the encryption to 256-bit AES, the algorithm allows for brute force and dictionary attacks to waste less processor cycles on each password interaction. You can read about it in Adobe's blog.

Necessarily, that type of password will be a strong one under Acrobat 9 and will render any brute-force or dictionary attack (pretty much the only means of breaking a pdf protected document) very inefficient methods. And while it needs to be said these tools will perform faster under Acrobat 9, it would still be years before a common user machine could eventually break your password.


One last comment, the size of your password will be the most determining factor in protection, as well as the unique count of characters. So, you can expect to provide a password such as mypaSwURD_frOM2009onMunTH#16, which is easier to memorize (includes purposed typos) and still obtain the same high security level.

share|improve this answer
    
On top of this, use words from a different language if you happen to speak one. A dictionary attack will always start with English as being the lowest common dominator (and very well so). –  Wolf Oct 16 '09 at 15:00
    
Klingon has worked well for me, very few collisions with English words. Elvish works well too. (Especially since one of our setups at work reject ANY password with ANY English Dictionary word. Do you know how hard it is to make up passwords that work?) –  lornix Jul 25 '13 at 3:12

Latest crackers can, on machines with the right video cards, use the GPU itself to crack passwords with a brute-force attack at a speed comparable to a super-computer.

If the password wasn't long enough, it will be cracked in a matter of minutes and up to several days.

Conclusion: Only if you use the latest Acrobat version and employ very longggg passwords and no dictionary words, will you be safe enough.

But then, all this will be a wasted effort if your password leaked to the web ...

share|improve this answer

I wouldn't trust either one, frankly. Password protections built into pdf, word processing, spreadsheets, archiving software...They're nearly all hobby-ist systems, put in place to stop people who are honest, not people who are determined. Doesn't matter how securely the password is stored if there are work-arounds (Acrobat is way better than Word, however).

I'd recommend looking into GPG or PGP for actual encryption (they're basically the same program, but PGP is polished, commercial, and expensive, and gpg is open source, little rough around the edges (as far as user friendliness goes), and free-as-in-beer.) You can integrate them with email, you can save whatever document format is convenient, and you can be sure that, as long as your key exchange procedures are solid, no one is going to be reading your mail.

From a more practical...shall we say legal...point of view, going to full encryption is going to do a lot to show that you're taken due dilligence with sensitive data.

share|improve this answer

I seem to remember that one could:

  • Obtain a free/open source PDF printer (i.e. you print to it from your application and it produces a PDF file)
  • Open the protected PDF in Acrobat Reader
  • Print the PDF to the PDF printer, thus ending up with a new PDF file with no protection.

Worth investigating.

share|improve this answer
1  
I assume the questioner is talking about a password to avoid opening the PDF file. –  Arjan Oct 16 '09 at 16:02

The simple test is to send a pdf file encrypted as V9.0 acrobat with a password similar to sd8Jf+*e8fh§$fd8sHa, and ask anyone to decrypt it. If after say 10 days no-one has replied with the contents on view then you know your data is safe. However, remember two problems with passwords. 1. Your recipient will have to know what it is - and may leak it as in the next item. 2. It's amazing how powerful key-loggers are. These read your passwords as you type them and potentially send them anywhere without you knowing. Your keyboard 'buffer' is your enemy in this respect. Even PGP suffers the same vulnerability. What's the answer? Place you data-files on a server - where you can only gain access via a two part process. E.g. see how PayPal now optionally allows access only via a new security code sent to your mobile. A PC key-logger would find this difficult to defeat unless your mobile is already infected by a key-logger!

share|improve this answer
3  
Simply asking someone to try decrypt something and getting no response is not any guarantee of security... –  David Fraser Mar 6 '13 at 14:57

This should be a comment to satanicpuppy, but the comments are limited to 600 characters. :-(

I support this (satanicpuppys) as being the most sensible answer.

You are looking at the strength of the password as a measure of how secure something is. In this case, you are - as an example - talking about patient data. So the security you are looking for is meant to secure the content not the algorithm or functionality (printing, saving, copy/paste).

While I agree that it might be superdifficult to print a document that is protected that way, PDF has been - and still is - dead easy to decrypt. That way the content can be descrambled and written into another file, with no restrictions whatsoever.

I am by no means a hacker, but the two phython scripts needed for that were so easy to use, even I managed to "free" my Adobe DRM-Protected ebook I just downloaded yesterday... No kidding.

Und natürlich würde man mal auf Elcomsoft gucken, denn dort findest Du Cracks für praktisch alles. PDF und Word ganz oben auf der Liste.

share|improve this answer
3  
You are looking at the strength of the password as a measure of how secure something is. -- no, I don't think so. I think it's just to say: "I know I have to be aware of brute force dictionary attacks. So, when limiting that chance by using a strong password, how secure is PDF protection?" And your "still is dead easy to decrypt" part needs some sources, for I think it's not true. (Your ebook DRM example seems irrelevant to me here as I think in this question, just like you wrote as well, the encryption "is meant to secure the content" whereas DRM is to stop duplication.) –  Arjan Oct 16 '09 at 15:10
    
You are looking at the strength of the password as a measure of how secure something is. -- no, I don't think so. -- That was what the author of the post said. He compares Adobes 128AES-Encryption to MS Word's... <br> And your "still is dead easy to decrypt" part needs some sources -- Although I don't like your agressive tone, Mr. Arjan van Bentem, I'd ask you to either google for "Adobe INEPT", look at Elcomsofts offerings or check a website with a name likely something like free my pdf. Just because my experience is different than your shouldn't make you arrogant. And don't lecture me –  Wolf Jan 19 '10 at 17:08

FYI, with a software like Passware Kit Forensic, you can crack a PDF with a massive password in seconds without any effort. Therefore, I think it's completely useless to use a password in a PDF document which is very unfortunate.

share|improve this answer
1  
The question says "but assuming I protect my PDF with a password like sd8Jf+*e8fh§$fd8sHä, am I on the safe side?" I doubt that the $995 software you mention can crack that. –  Arjan Jul 23 at 11:47

See: With 256-bit encryption, Acrobat 9 passwords still easy to crack

As it says:

"So, Adobe just makes [an] unbreakable thing stronger in Acrobat 9.

But actually security level is determined by the weakest link. In case if strong cryptography is used, the weakest link is a password"

So as long as your password is not of the type that would be iterated by a BFA, your doc is secure ... as of the time of that article's publication.

If one is using a BFA-iterable password there can't ever be any password based solution that would ever work - period. Then it becomes a PEBCAK problem - not a PDF encryption security problem.

[Update] Probably the best way to get the current status PDF security is to see what the current COTS PDF cracking software claims to be able to crack. Here is a tool that was added to CNET in 2011 -- this one, also available on CNET, was updated (as of this writing) on Aug,19 2013 -- I guess the point is that people are still downloading these cracker apps from CNET each week, hence, logic would dictate that you need to know what types of PDF files these apps say they can't crack. Also, some are trial-ware, so I suggest you try them out (from CNET) and see what they can and can't do on the files you are concerned about.

share|improve this answer
2  
This does not really answer the author's question. An article written in 20008 with regards to ANY security feature is worthless. I have to issue a downvote for only that reason. The article is 6 5 years old at this point. –  Ramhound Jul 24 '13 at 23:27
    
{Grin} I'm at a loss to think of a password that is NOT BFA vulnerable. How do you set the password then? Not saying it has to be something typeable... easy enough to use a program to insert password made up of randomized bytes. (UTF-8 is just a longer string of bytes.. it's all ones & zeros at the end) My philosophy... if it's on the computer, it can be broken... given enough time. Encryption is only really useful to slow the 'bad-guy' down enough so that the information gained is not useful by the time it's broken. –  lornix Jul 25 '13 at 3:16
    
@Ramhound: He clearly references "Acrobat 7 and higher" If he was't interested in v9, he wouldn't waste time asking about "Acrobat 7 and higher" My answer considers v9 coz he made a point to ask about "Acrobat 7 and higher" So it's not merely an assumption that he'd be interested in v9 since it falls within scope of "Acrobat 7 and higher" His answer is still pertinent to many, as it was to me when I 1st looked into it. Please forgive us but some of us still have to deal with lots of old PDFs which may not be altered (converted) in any way for legal reasons. See the big picture –  Michael.M Sep 29 '13 at 6:00
    
@Michael.M I never issued a a downvote but my feelings are the same a link to an old article is not helpful. Furthermore CNet is a horrible source to quote fu if non-technical people writing about technology –  Ramhound Sep 29 '13 at 11:58
    
@Ramhound CNET is not quoted in my post. I absolutely agree with you about the aspect of non-technical people writing articles however that has nothing to do with my post. I don't quote CNET and I certainly never quoted any of their posters either. The CNET reference is about trending: (1) the trend of software developers (technical people) publishing PDF crackers (2) and the trend of people downloading their software, what ever their intentions and success of using that software may be .. i.e. supply and demand –  Michael.M Oct 1 '13 at 19:13

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.