Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Why traditional password managers are flawed

github.com

Post reply on HN
Print view View original

41–50 of 64 posts

Re: Why traditional password managers are flawed

#41

Post by Xylakant »

tokenizerrr wrote: post #33
acdha wrote: post #32

Earlier quoted context omitted.

It's reduced but by how much? For example, 1Password uses PBKDF2 with something like 10k or 100K rounds to encrypt the local vault. If you were looking at ways that could be compromised, is it really likely that you need to worry about an attacker getting a copy and dedicating a lot of expensive hardware to brute-forcing it anywhere near as much as the attacker getting malware on your system which captures those keys…

Except 1Password is also pwnd if an attacker gets mallware on your system? They just snatch the database and record keystrokes when you unlock it.

Any algorithmic pw-manager can be attacked the same way: just record the keystrokes when the master pass phrase is entered. If the attacker has malware on your system, assume all is lost.

Re: Why traditional password managers are flawed

#42

Post by 45h34jh53k4j »

I strongly recommend that we stop inventing new implementations of a very flawed concept. Derived passwords from a seed is practically and operationally worse than a encrypted password database.

This keeps popping up lately, and every suggested implementation has objectively worse security, scalability and integrity than 1Password, KeePass, Dashlane or Lastpass.

Some simple problems that means this cannot work for me:

* Many password context's dont have a 'site' to derive from, now need to remember the IV for some random non-internet password (my luggage combo lock PIN, was the seed 'luggage' or 'travel bag' or 'samsonite'?)

* Might need multiple passwords per site; Multi users (need to remember all these shuffle codes)

* passwords are compromised all the time, now I have to remember the nonce/shuffle value for every one that gets leaked? (how do track this?)

* Storing non password data, like 'secret questions' or 'PINs' (not a generic secret store)

These same issues come up with every seed derived password scheme.

The only place this concept works is a very specific and metadata less context, such as HD derived cryptographic keys.

Re: Why traditional password managers are flawed

#43

Post by doppel »

Instead of remembering a master password, you now have to remember a sentence + a word for every site you want to use ("was it facebook, facebook.com or FB?").

Half the job of a password manager is to make sure you a) have unique passwords for each site and b) do not have to remember more than the master password. I sometimes use 1Password as a test to see if I even have an account if I visit a site I haven't been on in years.

Re: Why traditional password managers are flawed

#45

Post by KKKKkkkk1 »

I never understood password managers. What do I get from a password manager that I can't achieve by storing my passwords in a text file and encrypting it with $MY_OS's built-in file-encryption facility? And if I want timestamps and history, I can encrypt a git repo storing that file. Are password managers purely a Windows phenomenon, where you need a GUI for every little task you do?

Re: Why traditional password managers are flawed

#46

Post by mnarayan01 »

epistasis wrote: post #30

$ dpg "The sentence" word Doesn't this store all your secret info in your .bash_history file, an unencrypted plain text file, with a convenient marker for finding it in the history file? This is no better than just storing the password that DGP generates directly; it's deterministic and trivial to derive the password. For somebody concerned about a "master password" being insecure, this is a far far worse solution.

You could potentially add something like:

  HISTIGNORE='dpg *'
to e.g. your .bashrc -- not sure if the version of bash on OSX supports HISTIGNORE though (plus many other issues).

Re: Why traditional password managers are flawed

#47

Post by shinratdr »

Don't let perfect be the enemy of good. Have you seen how people manage their passwords? Even a deeply flawed password manager is better than the current method of using the same password for everything or resetting their password every time they need to log in.

Average people probably have 20-50 accounts they have to remember the passwords for regularly, and for people who do a lot of online shopping or internet browsing, it's probably closer to 50-100.

Even when we had to remember phone numbers, people retained what, like maybe 10, max? Expecting them to remember unique combinations of 50 different username/email/password combos isn't just bad security, its arguably near impossible for most people.

It's all well and good to warn about the flaws in trusting a browser (although this basically amounts to "too much is going on so don't trust it" which is a poor argument) or the inherent insecurity of a master password.

But if you're already in the state that most users are in, those "massive issues" are orders of magnitude better than the current state. The convenience compromises those solutions make is so that people will adopt them.

Is this more secure? Definitely. Will any significant number use it? No.

Re: Why traditional password managers are flawed

#48

Post by kuzimoto »

1ba9115454 wrote: post #11

I use KeePass and I'm quite happy. The database is local and enrypted with AES. My master password is around 72bits of entropy. I keep 2 files actually. Day to day stuff and critical services. So I only open up and therefore expose my master password infrequently for the critical services. (i.e. banking etc.) I've enabled 2FA on everything I can. So I think I'm safer than this deterministic solution as if my master p…

I love using KeePass. I keep the database file at home and only use it remotely with a plugin to allow accessing it via SSH so that the file is never (permanently) stored on disk.

I like to think that this is even more secure than other methods since even if someone were to log keystrokes, they would need to know the SSH server details as well to actually gain access to the file.

It can also be used to store 2FA secrets and KeePass2Android lets you generate the TOTP pretty easily so you're not relying on an app such as Google Authenticator which (last time I used it) doesn't backup secrets, and app itself isn't password protected.

Re: Why traditional password managers are flawed

#49

Post by kybernetikos »

mikeash wrote: post #39
kybernetikos wrote: post #10

Earlier quoted context omitted.

I use a system that I wrote https://github.com/kybernetikos/sinkless (incidentally it uses a similar system to this article - PBKDF with SHA512 - in browser plugins) that supports many of these situations. * The character sets can be entered manually. They form public information and are stored locally, can easily be exported as csv. * Each password has a 'version' which you can increment. The version is considered p…

What is the advantage of this over a traditional password manager? You need to sync a file so you can obtain the state needed to generate a password, just like a password manager. Your CSV file can be stored in a publicly accessible location without losing security, but the same is true of a password manager's database. Both are useless without the master password, and both unlock everything if the master password is…

I agree, there's not much in it. If you're not using proper salts there is a small advantage which is that it's usually possible to work out the settings for a site even if your sync file is lost since they will tell you the allowed characters and remembering a version number for the sites where you've changed it is not too hard especially since I show the corresponding emoji every time you log in.

I also prefer it to involving a third party service which is the other way password managers often work.

Re: Why traditional password managers are flawed

#50

Post by kuzimoto »

KKKKkkkk1 wrote: post #45

I never understood password managers. What do I get from a password manager that I can't achieve by storing my passwords in a text file and encrypting it with $MY_OS's built-in file-encryption facility? And if I want timestamps and history, I can encrypt a git repo storing that file. Are password managers purely a Windows phenomenon, where you need a GUI for every little task you do?

I'd assume that it's mostly for convenience really. Here's some things I can do with KeePass that a text file doesn't do:

1. Auto-Type

2. Access with mobile phone

3. Advanced search; by username, site, password, etc.

4. Audit password complexity

5. Password generator with many options (number of chars, what chars to use, etc.)

6. Ability to store other file types within the database which is associated with a specific entry

7. Password expiration rules

Those some of the things off the top of my head. While your method works for you, I really wouldn't trust 99% of people to know how to properly encrypt and store a password text file.

Post reply on HN