How safe or unsafe is this?
I’m asking because when I go to log in to vault.bitwarden.com it brings up community.bitwarden.com
And yes, I’ve tried changing the match detection to others but it still shows community.bitwarden.com on vault.bitwarden.com
When I tried a made up email rather than my own, it no longer showed community.bitwarden.com but that’s obviously not a solution.
Edit: I’m on iOS.
Well, if you trust it to securely store all your passwords, Bitwarden should be no exception.
And yes, I’ve tried changing the match detection to others
What exactly did you change?
It should be sufficient to change the “Website” field to https://community.bitwarden.com/
I can’t stop it from displaying the autofill for community.bitwarden.com on vault.bitwarden.com and I tried several different match detection options. It’s not a big deal.
Would you recommend creating a login for vault.bitwarden.com with my master password? Are there any potential downsides to doing that? I’m tempted to do it because my master password is very long. I don’t need to use the web vault much, I mainly just use the iPhone app nowadays.
It sounds like there may be a second password manager (e.g., Apple’s password manager) that is showing some of these autofill options.
What you’re reporting is not normal, and I would advise you to try to get to the bottom of it.
The main downside is that if your vault is compromised without your knowledge, then not only will your vault contents be stolen, but the impediments are now going to be much lower for the attacker to access or take over your Bitwarden account in the future. You can mitigate this risk by setting up a two-step login method with the second factor (OTP code or passkey) residing on a device where you do not use Bitwarden (e.g., a hardware key), and by paying close attention to security notification emails sent by Bitwarden.
You master password should be randomly generated passphrase consisting of 4 randomly selected words; this will generally contain a little over 30 characters (but they should be easy to type, being regular English words).
The general recommendation is to store the password in a safe place, e.g. in a Security Readiness Kit.
Personally, neither do I recommend nor do I advise against storing the master password in your vault. I am doing it, but the choice is yours.
In addition to what @grb already mentioned, there are additional risks involved, like when you perform an unencrypted export of your vault.
I don’t think so because Bitwarden has been the only password manager that I’ve used for a number of years now.
I find it a bit of an eyesore but unless I make a login for vault.bitwarden.com I may be stuck with it.
Why’s that?
I’m using TOTP 2FA from an app on my iPhone. I’m unable to view or edit the seed/secret though.
My master password has more words than that now, plus some numbers and symbols thrown in. I may forget it at some point but I’ve got it written down. I also recently switched to Argon2id which is apparently better than PBKDF2. I wonder if 4 random words alone (with no numbers or anything) would be secure enough for the next say 100 years?
I’m thinking of doing it now. It’ll fix the problem with the community.bitwarden.com autofill showing on vault.bitwarden.com too.
I don’t think I’d ever create an unencrypted backup/export of my vault unless it’s needed when switching to a different password manager. I’m not sure if that’s the case 
My point is that if you have passwords stored outside of your Bitwarden, then that may create a security risk for you. For that reason, if it were me, I would not be so cavalier about the issue. My recommendation is to investigate what is causing this behavior in your browser; we can help with this.
Please make sure that you have stored your two-step login recovery code in a safe location; without this code, you are at high risk of permanently losing access to all of your Bitwarden vault data (for example, if your iPhone stops working, is reset, or is lost or stolen).
Because the attacker will already have your email address and master password; thus, they can gain entrance into your Bitwarden account by just defeating the TOTP code.
As long as all of the words were randomly generated without any human input, that’s fine. Randomly selected numbers or special characters are also fine, but not necessary.
Yes, if you keep your KDF settings adjusted according to updated recommendations that may be issued in the future. The exception would be if your threat model includes an adversary who has already stolen an encrypted copy of your vault (or will do so in the near future), and who plans to warehouse the stolen data for 100 years before mounting a brute-force cracking attempt using computing technology available in the distant future. If that applies to you, your master password should probably contain 8 randomly generated words instead of four.
Doing what? Storing your vault master password in Bitwarden is not going to stop your community.bitwarden.com account from being offered as a suggested login when you’re on the vault.bitwarden.com site.
I don’t have passwords stored outside Bitwarden. I’m using Safari on iOS if it helps.
It’s written down. I wish I wrote down the seed/secret for my Bitwarden’s 2FA. Am I able to view it in the web vault or would I have to recreate the 2FA and write down the new seed/secret?
Do you know this for a fact or? If so, it kinda makes me wonder why cryptocurrency wallets use 12-24 word seed phrases. I know the word list is smaller (2048 words I think) but still, it’s massive overkill right? Or are they much faster to crack?
Oh? I thought it might if I made a new login for the master password and put vault.bitwarden.com or just bitwarden.com?
@grb You there?
In that case, you have not properly configured the URI match detection.
The latter former.
Yes. Using a random 4-word passphrase means that an attacker would, on average, need to test over a quadrillion guesses, on average, before hitting on the correct master password in a brute-force attack. Bitwarden’s default KDF settings are designed to throttle the guessing rate to 10,000 guesses/second per GPU. Thus, 6,000 GPU-years (and millions of dollars) would be required to crack a Bitwarden master password consisting of a 4-word passphrase.
I think you’re referring to a BIP39-enocding of a 128-bit or 256-bit cryptographic key. Different application.
No, you will just be offered both vault items as autofill options.
I’m not here 24/7 (although it may sometimes seem so), and I do have other responsibilities.
I have an update. I created a login for vault.bitwarden.com and set the match detection to “Exact” which is also what I have it set to for community.bitwarden.com and it shows the correct autofill on vault.bitwarden.com now.
Also, I can view Bitwarden’s 2FA seed/secret in the web vault which I’ll be writing down.
You’d think so but I tried like all of them available in the iOS app. Maybe the web vault differs?
Is that with PBKDF2 with 800,000 iterations or with Argon2id with default settings?
I’m referring to the autofill that appears on top of the keyboard when a username or password field is clicked. Sometimes for some sites, it simply has a blue button to press at the bottom which not only autofills the login but also logs in without me having to press log in on the webpage.
Understood.
I have no personal experience with the iOS mobile app; to my knowledge URI match detection is primarily for the browser extension — which is the context under which my comments have been framed.
The analysis I provided is for the default PBKDF2 settings. The default Argon2id settings are even more secure (more costly per guess).
Someone else will have to help you with iOS questions.
Hi @_Michael
I would doubt there are reputable sources that throw around that kind of number, given unknown technological advancements. The typical timeframe considered is usually 30-50 years (the typical remaining lifetime). @grb has rational reasons based on current computational power and future trajectories. However, if you have anxiety, irrational fears, or doubts about the more reasonable numbers experts are suggesting, that doubt may always nag at you. If you search for more information, you may find that some US agencies recommend 128-bit encryption protection (not password entropy) for long-term sensitive archives.
With that in mind, you can use a randomly generated passphrase of 10 or more words, making your password entropy 128+ bits. If you can touch type on a normal keyboard and remember it well, it probably takes less than 30 seconds to type, but it would be a pain on a mobile onscreen keyboard. On the other hand, you can configure it so you don’t have to type on mobile as often, or even on desktop, if you trust Bitwarden’s “Login with Device” implementation.
Your choice: a more manageable and reasonable password or something that may put your irrational mind at rest. You can even consider using an even more radical Bitwarden password that you keep in an offline password manager, stored on detached devices only.
To clarify my argument: As computing technology advances and brute-force guessing becomes more efficient, so will KDF algorithms evolve to make hash calculation slower and more costly. Thus, by periodically updating the recommended KDF settings (e.g., following OWASP recommendations, which are regularly updated) used for your Bitwarden vault, you can maintain a throttled hash calculation rate at around 10 kH/s, into the foreseeable future. Thus, there should be no need for future modifications to a master password consisting of a 4-word passphrase, unless there is reason to believe that the password has been compromised.
The only risk is if the attacker can access an old version of the encrypted vault data, which was encrypted using an old KDF configuration. One scenario under which this could occur would be “harvest now, decrypt later”, in which an adversary routinely acquires encrypted data by surveillance, and warehouses these data sets for many decades, until such time that brute-force decryption can be accomplished using future technology.
Because it has a smaller number of words, the entropy aspect for BIP39 (2,048 words) is about 12 words for approximately 128 bits of entropy, 18 words for 192 bits, and 24 words for 256 bits. You can achieve the same with the Bitwarden passphrase generator (EFF long-word list with 7,776 words) using 10 words (128 bits of entropy) and 20 words (256 bits).
To derive the numeric seed, the crypto wallets appear to use PBKDF2 with 2,048 rounds of iteration, which isn’t used to protect weak “passphrases” since it already has a minimum of 128 bits of entropy. There we go again with the magic number of 128 bits, regardless of the KDF used.
I’m not using a PC these days, I’m just using my iPhone. URI match detection is for mobile too. It must be because it’s letting me fill in logins in browsers and apps (provided I saved URI info beforehand of course).
I meant that maybe (but more than likely not I’m guessing) the URI match detection options differ in the web vault (vault.bitwarden.com) to the iOS app.
Let’s say an attacker/adversary gets a copy of your encrypted vault and knows that your master password is exactly 4 words with no separator. And let’s also say that the attacker has access to a massive amount of computing power from botnets and what not. Is it still going to take forever to crack due to the KDF iterations or whatever?
And what do you mean by an even more radical Bitwarden password that you keep in an offline password manager, stored on detached devices only?