Is it fine if a website says "email already in use" before you hit submit?
I remember hearing before that it's a sign they are storing your info unencrypted but I never checked.
Is this true? I was logging into a .gov website and noticed it does that.
The only issue with it is that it allows attackers to determine that a given person has an account on a site. Which if the site is pornhub or similar, could be embarrassing (try sign up to pornhub with your local politicians email).
The way around this is to do something like:
"We need to verify your email is correct, by sending you a code"
This doesnt tell the attacker anything, but if there already is an account, the email itself can just say "You already have an account, here are the links to reset and login".
Side note: encryption is reversible, hashing is not. Passwords should be stored hashed, but email only need to be encrypted (or plaintext, but less ideal). And because its reversible, they can get the original value back. They cannot reverse a hash to get the password back, so if a site ever tells you info about your password, that is a sign they might not be hashing it correctly.
Given many people use the same password.on many sites, it can allow the bad actor to "guess" their password based on data from other leaks.
Then whatever is inside that account may be accessible, such as financial info. Even protected data like cc info might show the last 4 digits, which can be used when talking to an agent to verify identity etc.
This is the way.
If you're going to encrypt the email, you need to be careful about how you use and store the key. Doing any operation with the email will be a lot more expensive, and you'll lose the benefits if an attacker that can access the db also has access to the key.
I personally don't think it's worth it and would prefer to spend more time hardening the app, especially if the email is displayed on the site (i.e. it gets decrypted frequently).
It probably makes sense when there's sensitive data (bank, medical care, etc), but for most things it's overkill.
I don't think many places encrypt/hash email addresses, but even if they did they could just apply the hash algorithm to what you entered to compare the hashes.
Anyone dealing with health information should definitely be encrypting email address; it's one of the HIPAA identifiers.
So ultimately hashing an email address could be a good thing, but its a matter of half measures. Sure, you can perform a basic hash before putting it in the database, but if we assume hashing is performed to prevent it being read by an attacker, why bother unless youre doing it properly?
Passwords, being more sensitive, should only be compared once finished being entered, so you can afford to run all the hashing, salting etc that is a requirement to keep the passwords safe.
If you were going to hash the email to the same standard, it becomes harder to retrieve and display, so when the user wants to look at their profile in the ui, you have to run an intense cryptographic algorithm just to display the email. Or if you want to contact the customer, or any other use for their email. Hence, people dont bother.
Hashing is completely irreversible. You cannot hash an email address and then unhash it. At most you can brute-force guess the email until the hash matches, but this is basically impossible.
Hashing the email address would break one of the main reasons to use an email address - the ability to send emails to users.
Encrypting email addresses is fine, but you wouldnt compare the encrypted data, you'd just decrypt and compare the original email address.
The email can be stored in 2 parts in the database. One hashed as a login and once as an email address to send to.
Sure, if your database is compromised, it's still public but it still makes the log in process more secure.
The only thing that needs to be encrypted or hashed is the password.
But telling that an email is already in use is leaking information. A bad actor can use this to figure out if you are using a particular service, or alternatively try random email addresses and check if they belong to a real user. This is why it's usually encouraged to just say "invalid combination of username/email and password", instead of specifying which is incorrect.
I keep telling my team this all the time... The push back is always from the support side that says "well users complain that they don't know what's incorrect to fix." and my answer is always "they got their own credentials wrong... it's ALL incorrect. Do it over".
But that's "user hostile".
"WIPE IT CLEAN. This fucking lady's getting a new account."
That's such brain dead reasoning. Only the password should be hidden - if the user can't tell whether their username is correct they need not to be using a computer...
Crackers. Why we can't have nice things.
User registration will still need to check if the email is the user id (which I loathe).
Not necessarily. If it's implemented well, the frontend will just show a "success" message, but the email sent will be different. This way, the owner of the account will know if they already have an account, or if it wasn't them, that someone else tried to use their email. Meanwhile the bad actor won't know anything new.