Password manager says breach it disclosed in August was much worse than thought.
See full article...
See full article...
I would dearly love to have an easy-to-follow web manual on how to implement a Vaultwarden solution for people who are not super Terminal-friendly/on Linux.
I got Syncthing working on my Macbook and iPhone (using Möbius Sync), then got Docker up and running on the Macbook, but after installing the Vaultwarden server with Kitematic it started me off with the message "It looks like you did not configure a persistent volume!" along with a warning that any time I updated I'd lose all data. Yikes. A quick search for how to resolve that led me to "solutions" that were way past my comfort/computer-savvy level so I gave up for now.
KeePass seems easier to set up in this regard, but I've been using Bitwarden the past couple of years and I like it's format and function and would rather keep using it.
"Military grade" = AES256. Or possibly AES512.Agree. I was a long-time user of 1Password and that ended it for me. I don't care how much "military-grade" encryption you use, I'm not storing my password database on machine that I don't control. Period.
I have a Synology NAS. I installed VaultWarden in a Docker container and use the BitWarden client on the family phones and computers. Setting it up wasn't like falling off a log, but the experience has been very positive.
Well, Bitwarden is a bit of a different beast. That is totally 100% self-hosted, so the question then becomes: Am I better at managing this stuff than these online companies?"I think I'm done with LP now. Ever-increasing fees and yet another security failure.
How in f's sake could you fail to encrypt secure notes???
Switching to Bitwarden tomorrow most likely but how do we know that they are any better at this?
But they could have updated it the next time you logged in (which is quite frequent), a password change isn’t needed.Yeah, it's common for applications using PBKDF2 to update the number of iterations they do over time. For example, if you set a password with Lastpass back in like 2015, it'd probably only be like 5000 iterations. But if you had updated your password, it would use their latest default iteration count. It's not something they can update without you providing your master password, since it is basically a re-key of your entire vault.
That's how any reasonable symmetric encryption works. Your passphrase isn't the key, your passphrase unlocks the key. That's how you can change your password without re-encrypting everything or you can have multiole passphrases/keys/signatures to decrypt the same thing. It's also why you don't have to make a password exactly the length of your key.They announced that they generate a new key when someone logs in via web browser. That means the attackers know that all passwords are X length, and they know the algorithm used to generate the keys. That seems like the real threat here. I was unaware until this hack and some research that they create a new key that you can use to decrypt everything. The idea that it uses only your master password is false, because they allow master password resets using that new server generated key. You can see the key for yourself by clearing your cache, then logging into Lastpass, and seeing the new files generated. The attackers will know that all the passwords they need are the same length. LastPass never should have allowed a reset option for the master password. That isn't zero knowledge since they generated a new encryption key, and admit as much. Zero knowledge means only the user has the key, but that is empirically false since you can reset the master password using the new key generated upon login to the program. The user doesn't even know the contents of that key. It is downloaded from a server or generated in some fashion (the file is hard to read). The attackers know that all decryption keys are X length because LastPass creates one you can use to reset your own masterpassword. With true zero knowledge, I would have stayed because my master password is ridiculous. But that is moot since they generate one that can be used to reset said password. That means there is another key that can read everything, and the attackers have the code to generate it and know how long it is. Bad design choice which they call zero knowledge for marketing purposes. They admit on their own websites that a new key is generated which can read everything. The attackers will have the generation source code and the length of that key for the data they intend to crack. I was not aware of that until digging into this. I thought my master password was the only key and made it a doozy, but they mooted it by having all last pass instances create a new server derived key on first login that you can use to reset your master password. For that to happen, said key must decrypt the contents of the vault..
Bitwarden is open source and you can host it yourself, but they also offer it as a service themselves. They even have paid tiers and everything.Well, Bitwarden is a bit of a different beast. That is totally 100% self-hosted, so the question then becomes: Am I better at managing this stuff than these online companies?"
Probably yes, but mostly through the 'security through obscurity' model.
There is risk all around w/ storing this stuff online (self-hosted or not), which is why it is best (theoretically) to keep part of your locker in one area and the ability to access that locker in an entirely separate solution.
Example: BitWarden on your self-hosted solution, with a private key required to access it, which is stored on another online solution - say, Google Drive. Then of course using a strong password.
Oh interesting. But I thought the entire point of Bitwarden was to be a self-hosted solution. If there is an online, paid tier for them hosting data I would assume they're just as vulnerable as LastPass, 1Password, etc. The whole 'someone else's computer' thing.Bitwarden is open source and you can host it yourself, but they also offer it as a service themselves. They even have paid tiers and everything.
I checked my settings, and the value was set to 5000, I had never set it myself, but have had the account for at least 11 years. So I assume at some point the default was 5000 and was increased to something "woefully short" of recommended values. Which leads me to believe some have settings that are whatever 5000:100100 is if 100100:310000 is woeful.Those settings hash stored passwords using 100,100 iterations of the Password-Based Key Derivation Function (PBKDF2), a hashing scheme that can make it infeasible to crack master passwords that are long, unique, and randomly generated. The 100,100 iterations is woefully short of the 310,000-iteration threshold that OWASP recommends for PBKDF2 in combination with the SHA256 hashing algorithm used by LastPass.
They have free and paid service.Oh interesting. But I thought the entire point of Bitwarden was to be a self-hosted solution. If there is an online, paid tier for them hosting data I would assume they're just as vulnerable as LastPass, 1Password, etc. The whole 'someone else's computer' thing.
The entire reason I migrated away from my KeePass/Dropbox/Drive/password/key setup and to 1Password was because I liked their security model. They don't know anything about your stuff. So even if an attacker gained access to 1Password's servers, including password lockers, there wouldn't be a way to break into the lockers.
Agreed, no trust is best.They have free and paid service.
Yes, someone could break in and steal all the databases Bitwarden has. Same could happen to Dropbox, they'd just have to root around to find your database. Dropbox is someone else's computer too.
This is why you encrypt everything. I can't guarantee someone isn't going to make their way into service X, Y, or Z. Likewise I can't guarantee someone isn't going to break into my house and walk off with my desktop or my server. So, everything is encrypted. No need for trust.
If the central store contents were never going to change, absolutely, you could just auth in on first connect and never worry about it again.If it was purely for convenience, the strings could be decrypted, copied to a local repository, encrypted with a local device/app/user key combo upon the first authentication to the app (which could be required during install of the client).
I have yet to hear a valid technical explanation for why this field was left clear.
"Military grade" = AES256. Or possibly AES512.
I strongly dislike the 'military grade' description these password locker companies eschew in their promotional material. Tell me how you encrypt, don't fluff it.
AgreedI like it when someone says “military grade”. It tells me they're either incompetent or conmen and I don't need to spend more time learning about their product. It's rare that you can otherwise reliably evaluate a security product in 10 seconds.
For example: 1Password does not use the term. LastPass does.
Also... when we have places like the IRS getting attacked and leaking all of our social security numbers etc, It nice to know someone actually takes these threats seriously. It's not if but when there is a data leak, I'm glad they have our data secured so well. Do you thin the IRS took this much care with my social security number?“These encrypted fields remain secured with 256-bit AES encryption and can only be decrypted with a unique encryption key derived from each user’s master password using our Zero Knowledge architecture,” LastPass CEO Karim Toubba wrote, referring to the Advanced Encryption Scheme and a bit rate that’s considered strong. Zero Knowledge refers to storage systems that are impossible for the service provider to decrypt. The CEO continued:
I don't mind people knowing I bought something from Adam & Eve or used a dating app. I'm not even embarrassed that I use Bing as my daily driver for search. If someone wants to judge me for that it's a 'them' problem, not a 'me' problemI'm very surprised few people have mentioned the URL leaks. This is, IMHO, the most damaging leak I've seen.
ALL URLs that an account had stored were never encrypted. Whatever URLs you had logins for, those URLs are public + attached to your name, billing address, phone number, and email address.
It's now public who had .mil, .gov, OnlyFans, PornHub, Grindr, etc. accounts. That's just the USA.
A public key and a hashed strong password are essentially the same thing. The one difference is how a password vs key is sent to the server. You send a password to the server over an encrypted channel, then it is hashed server-side to compare to the stored hash, but it is the actual password being sent. With keys, you essentially sign a random number with your private key, then send that.Um, no? It's using the same tech as is used for accessing secure websites, or SSH or Wireguard or anything else decent: public/private keys. The site you're authenticating to never has your private key. You don't need to send it. If the site is hacked and their entire key database is dumped it doesn't matter. You could take a public key and post it in the forum for all to see and it doesn't matter, anymore then you being able to click the lock and inspect Ars Technica's certificate and then the certificates of the chain to the CA (Amazon in this case) means you can now run a website and impersonate Ars.
You're not overly cynical, you're ignorant. Asymmetric auth isn't the same as symmetric. And none of these attacks, nor any scalable attack, are actual physical attacks. "Real-world meatspace" is precisely the advantage! Having a PIN or even just a touch to prove operator presence and then a blackbox to handle keys and identities is both easier and completely obsoletes entire classes of attacks.
anyone who's ever been in the military knows that phrase is code for "will break when used" and not "not as cool as you think it is" lolAgreed
I used to use Last Pass but decided to switch when VC's got a hold of them. I use Windows PC's for home and work and when shopping for a new Password Manager the one I ultimately chose was Bitwarden. It's easy to switch over - you can export your LP passwords to a CSV file and then import them into Bitwarden (this is a good time to review your accounts and possibly deactivate those you no longer use). Bitwarden can be installed as a browser extension and also has a mobile version which are both similar to LP. All this means once you get things setup on your devices you'll more or less just have to train them to look for the new app. Everything will work the same way. Of course, you'll want to delete the CSV file when done, too (which is to say - purge it from your recycle bin)Dan (or anyone with cybersecurity knowledge): In your honest opinion, would you advise changing password manager services if you were on Lastpass? I have considered switching to 1Password but didn't want to go through the hassle of teaching my entire family to use another service. Lastpass sort of has a hold on me over that major inconvenience. However, I have zero desire to have accounts breached out of my laziness to retrain people. If so, what other services would you recommend for family accounts / family sharing and why?
Yikes, I too moved to bitwarden then too (and love it). Thankfully I did delete* my LP account. I got my Dad on LP years ago, now I need to get him migrated as well. Unfortunately, I doubt he's using a strong password, so I'll also have to impress on him to change all sensitive passwords (banking, etc). Ugh, there goes a few hours...I switched to bitwarden when they launched their pay-for service. But I never came back to delete my account. So I went back and fixed the problem. Sadly, my encrypted data is out there.
There are numerous comments here about the 5000 iterations that many people have, because they set up their lastpass master password when that was the standard. But aren't all other password managers affected by the same issue? If they don't have your master password, they can't re-encrypt with newer, stronger algorithms.Sadly, I'm in the same boat. 5000 iterations (now set to 200k until I migrate to bitwarden/1password), so starting the arduous journey of changing every password.![]()
Absolutely. Total loss of faith.Currently using LastPass with 2 factor authentication enabled wherever possible. My master password is around 40 characters long, with numbers, special characters and mixed case. The password has nothing in it that could be tied to anything about me that could be found online or in any documents, but it's easy for me to remember. The password also isn't a famous quote or from any literature, movie or song or any trivia, fact or historical event. It's not used elsewhere. I'm not worried about it getting cracked, but I'm moving off LastPass. It's inexcusable as far as I'm concerned to have any customer data unencrypted for a service like this. I'm pissed about having to move. I'm pissed about having to create new passwords. And I'm really pissed about needing to come up with a new master password.
Isn't that usually jargon for FIPS140-2 compliance? Which, until relatively recently, meant including the backdoored Dual_EC_DRBG generator.I like it when someone says “military grade”. It tells me they're either incompetent or conmen and I don't need to spend more time learning about their product. It's rare that you can otherwise reliably evaluate a security product in 10 seconds.
If you are using a shell script to generate password candidates like I do, you can make up your own word lists. For example, if I wanted a word from this comment page, I could do something like:For what it’s worth, I can say 1Password and Bitwarden offer somewhat more entropy than just four words in their passphrase generators.
The divider can be nearly any character, you can add capitalization, you can add numbers, and from what I can tell, it’s significantly more than the most common 4096 words.
I recently got rather esoteric words like moravia, kibitz, pibroch, neology, and rowel.
See here: https://1password.com/password-generator/ (this has some fewer options than 1Password itself; see below)
View attachment 49295
EDIT: according to this site, Bitwarden uses 7,776 words. 1Password uses 18,300 words.
https://passwordbits.com/passphrase-cracking-calculator/
$ cewl -d 0 https://meincmagazine.com/information-technology/2022/12/lastpass-says-hackers-have-obtained-vault-data-and-a-wealth-of-customer-info/?comments=1\&comments-page=9
CeWL 5.5.2 (Grouping) Robin Wood (robin@digi.ninja) (https://digi.ninja/)
and
the
LastPass
data
that
The
Share
Ars
customers
...
encrypted
stored
storage
hackers
hit
single
Google
...