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

Ask YC: Best Practices for User Authentication

news.ycombinator.com

Post reply on HN
Print view View original

1–10 of 50 posts

Ask YC: Best Practices for User Authentication

#1

Post by markbao »

Hey, I'm wondering about the practices about authentication with regard to passwords.

I'm looking to exclude the HTTPS / SSL cert method, which is obviously the most secure method, but it's not completely within reach right now.

What I'm currently looking at is simply a Javascript SHA2 implementation that hashes the password before it is sent to the server. After reading a recent post (http://news.ycombinator.com/item?id=205420), it seems that this implementation isn't the best.

Here's what I got from that:

     +----------+           +----------+
  +->|  SERVER  |---------->|  CLIENT  |
  |  +----------+   nonce   +----------+
  |       |                      |
  |  SESSION["nonce"] = "1234"   | SHA(nonce + password)
  |                              |
  +------------------------------+
(sorry for ASCII drawing failure)

What is good practice for user password authentication without SSL? (Feel free to yell at me for not using SSL, but I'm currently not able to implement it.)

Thank you.

Re: Ask YC: Best Practices for User Authentication

#2

Post by tlrobinson »

Hashing the password on the client with a nonce provided by the server and checking on the server is good for login, but not registration. The server still needs the password (or a hash of the password and a salt) in order to compute the login hash for comparison.

The only truly secure way (that I can think of) is some sort of public-key encryption, either SSL or some JavaScript library.

Re: Ask YC: Best Practices for User Authentication

#3

Post by jrockway »

You've basically reimplemented HTTP Digest auth, although not quite as well.

The problem with your implementation is that the server needs to know the cleartext password. With digest auth, the server never has the clear password, (only a digest thereof) although it can use the digest to authenticate to other services with the same digest.

It doesn't really matter though, since passwords are very insecure and users know that. Nobody is going to give your web app an important password. If they do, and it gets compromised, it's their problem, not yours. (For secure authentication, look at how ssh uses public/private keys. Much better. Compromising the server's password database will never yield the private keys and hence is a complete waste of time for an attacker.)

Anyway, I tend to transmit passwords over SSL and then store them in the database hashed with bcrypt (http://search.cpan.org/~zefram/Crypt-Eksblowfish-0.005/lib/C...)

Re: Ask YC: Best Practices for User Authentication

#4

Post by markbao »

tlrobinson wrote: post #2

Hashing the password on the client with a nonce provided by the server and checking on the server is good for login, but not registration. The server still needs the password (or a hash of the password and a salt) in order to compute the login hash for comparison. The only truly secure way (that I can think of) is some sort of public-key encryption, either SSL or some JavaScript library.

That's very true. Damnit. Not sure how I'm going to fix that.

I've looked into public-key encryption, and unless if I generate a new public and private key for every request to login, replay attacks are still possible.

Re: Ask YC: Best Practices for User Authentication

#6

Post by markbao »

dfranke wrote: post #5

What's the difficulty with implementing SSL? If you're not already familiar with it, read the manpage for s_server, which is part of OpenSSL.

No difficulty in implementing SSL, but I'm a student and don't have much money. Granted, this is a temporary solution until I'm able to get some money together for a dedicated IP and SSL cert.

Re: Ask YC: Best Practices for User Authentication

#7

Post by dfranke »

markbao wrote: post #6
dfranke wrote: post #5

What's the difficulty with implementing SSL? If you're not already familiar with it, read the manpage for s_server, which is part of OpenSSL.

No difficulty in implementing SSL, but I'm a student and don't have much money. Granted, this is a temporary solution until I'm able to get some money together for a dedicated IP and SSL cert.

You can do SSL very cheaply these days. GoDaddy has certificates for $15/year. Dreamhost gives you a dedicated IP for an extra $4/month.

Re: Ask YC: Best Practices for User Authentication

#8

Post by DaniFong »

markbao wrote: post #4
tlrobinson wrote: post #2

Hashing the password on the client with a nonce provided by the server and checking on the server is good for login, but not registration. The server still needs the password (or a hash of the password and a salt) in order to compute the login hash for comparison. The only truly secure way (that I can think of) is some sort of public-key encryption, either SSL or some JavaScript library.

That's very true. Damnit. Not sure how I'm going to fix that. I've looked into public-key encryption, and unless if I generate a new public and private key for every request to login, replay attacks are still possible.

If you hash the password on the clientside, say, and then use that as the password in your scheme throughout, then you don't need to transmit the password in cleartext, ever.

Re: Ask YC: Best Practices for User Authentication

#9

Post by tptacek »

No security scheme delivered over Javascript is going to make a difference in a security audit. Your auditor is going to tell your prospective customer that anybody who controls the DNS, routing, the network, ARP, the browser --- or 10 other places --- can rewrite the Javascript in transit.

If you can't do SSL, just do plaintext passwords. It means you're being open with your users about the risk.

Post reply on HN