- cross-posted to:
- cybersecurity@infosec.pub
- cross-posted to:
- cybersecurity@infosec.pub
- Millions of people use password managers. They make accessing online services and bank accounts easy and simplify credit card payments.
- Many providers promise absolute security – the data is said to be so encrypted that even the providers themselves cannot access it.
- However, researchers from ETH Zurich have shown that it is possible for hackers to view and even change passwords.
You must log in or # to comment.
Many providers promise absolute security
This struck me as wrong, because that would be a technically impossible and liability-inviting thing to promise.
And after checking the homepages of the 3 services they tested, yep, none of them promise “absolute security.”
i really wasnt expecting a password manager related tech fearmongering on lemmy today
With pretty much every major company being hacked at some point, credit card companies being hacked, everyone selling our details and data, doge and palantir. Feels like post it notes under the keyboard isn’t that bad of an idea.
If someone breaks into my house to read them I have big problems already.
You have no idea how many times I’ve made that exact statement.
Let’s start a club
Don’t store your stuff in the cloud unless you don’t mind someone else accessing it.
If you store things in the cloud that you don’t want other people to access, you better be encrypting it yourself and only opening it locally.
This has been a cardinal rule since day 1.
I don’t want people to access my files but I wouldn’t really care if they did. I don’t understand people who keep things like compromising photos of themselves online, who’s benefit is that for, and why do you need quick access to your nudies?
If it’s something that you don’t really care about others seeing, that’s a prime candidate for cloud storage and more power to you.
This topic is about password lockers. I’m pretty sure you don’t want some schlub who happens to work at Cloud Password Lockers Inc. to be able to get at your PayPal account.
From the paper itself:
We had a video-conference and numerous email exchanges with Bitwarden. At the time of writing, they are well advanced in deploying mitigations for our attacks: BW01, BW03, BW11, BW12 were addressed, the minimum KDF iteration count for BW07 is now 5000, and their roadmap includes completely removing CBC-only encryption, enforcing per-item keys and changing the vault format for integrity. On 22.12.25 they shared with us a draft for a signed organisation membership scheme, which would resolve BW08 and BW09. At our request, to maintain anonymity, they have not yet credited us publicly for the disclosure, but plan to do so.
I didn’t look at the response to other Password managers, but the gist here is that the article is overblowing the paper by quite a bit and the majority of the “issues” discovered are either already fixed, or active design decisions.
I was also just looking for bitwarden information. Its just the best password manager and has never failed to do its job.
I dont know what they mean with less secure than promised. I didnt expect them to be perfect, and havent read that they promise no security flaws.
They advertise that passwords are only stored on the server in encrypted form, meaning they couldn’t read them even if they wanted to (or were forced to by a government agency) and you don’t have to trust them not to. This paper shows that several vulnerabilities exist in the protocol which could be exploited by malicious code running on the server (injected by hackers or a government agency), which would then allow an attacker to obtain cleartext-passwords. So you do, in fact, have to trust the servers integrity.
Thank you for taking the time to understand and comment, very valuable.
The beauty of open source
I pitty the fool that stores anything important on
the cloudsomebody elses computer.Would having a synced Keepass database with a composite key protect against this?
When I made my database I created a composite key file that never goes online. I locally copy it to any device that needs to access the database. The idea was even if the password got compromised you can’t access the database without the key file
What if you have a house fire and lose all devices with the key
What if there’s a nuclear war end the house gets vaporized?
To protect against this scenario I have this small portable computer that I keep in my pocket. They’re quite popular these days.
“We want our work to help bring about change in this industry,” says Paterson. “The providers of password managers should not make false promises to their customers about security but instead communicate more clearly and precisely what security guarantees their solutions actually offer.”
Great.
Now which password vault was the most cooperative and clear in their security communication and which one wasnt?
The author said that they have given the providers time to fix the issues. Now highlight the ones that did it the best… >_>They did gove some advice. They said to go with a vendor that is transparent about problems and reveals the results of their third party security audits. I’m sure if you read between the lines it means they likely reviewed several vendors and chose to spend their time attacking ones that are opaque about their security stance and used outdated encryption or bad implementations of E2E encryption. So all three are likely suspect. Like if 1Password were developed similarly to LastPass wouldn’t they have spent time attacking it?
Edit: https://support.1password.com/security-assessments/
1Password are posting the results of their external pen testing now.
Bitwarden did so too.
But IMO your assumption is a bit of interpreting bad/malicious faith into it.
I see it more like they are the more publicly known brands/services that do this and underwent the audit.
I have read the TLDR by the authors (linked a few times in the comments) and the answer by bitwarden.
Bitwarden said the, fixed the issue, are in the progress of doing it or are accepting it as “this is intended/a trade-off”.
What is a bit sad is that they had more vulnerabilities than other vendors. But I trust them more as they are mostly OSS.About 1password publishing their pentesting results. Why put it behind a ‘give me your email address’ wall?
That alone is enough for me to instantly disregard them as an option.
That’s why mine is a physical book.
Really depends on your threat model whether this is a good idea. If cops raiding your home is part of it, a physical book might not be your best bet.
That’s very true.
If you’re at the point where that’s a possibility that you need to defend against then you probably already have better security than using a password manager.
OMFG can people please fucking go away with this stupid “password managers are worthless” bullshit today. They are exactly as secure as promised, unless you went to the obviously shady ones that use web interfaces. People have been saying this for years, if you want security, keep your password manager offline.
So by that logic BitWarden is unsafe?
Yes, if you arent self hosting the web interface or using the desktop client.
But these issues were patched before even publishing the findings, right?
There is no way to patch the inherent flaw that comes with delivering client software through a web browser. If the entire client is delivered as a web page from a server you dont control, then that server can modify the software however it pleases. Same applies to e2ee encrypted chat clients that run as a web page like element-web (browser based matrix client).
This feels a bit extreme though. Can you even trust anything online at that point? Do you also never leave your home carrying your wallet in case someone might rob you?
Bro i have my bank details, all my private 2FA, work 2FA, health insurance access, my families master passwords, steam access, and more in there. Its literally the most important piece of software that can exist in this day and age. No im not taking chances with that. The only thing you can do with my physical wallet if you rob me is buy something up to 20€ beyond which you need the cards pin. Everything else i can just deactivate by calling the relevant parties.
But on another note, websites have never really been resistant to MITM attacks. So you dont just have to trust the hoster but also everything in between you and them.
Use a offline password manager. Problem solved.
Solves the security issue. Destroys the accessibility part
tl;dr:
- If the password manager server is hacked and compromised, then syncing your passwords with the compromised server will lead to compromised passwords (duh)
- None of the providers tested have (or have had in the past) compromised servers.
and an observation or two:
- Vaultwarden is free, self-hostable, and doesn’t rely on trust in a third party.
- Keepass (and its client variants, like KeepassXC which is pretty great) is even more secure because there is no server, just an encrypted file you can store anywhere.
Keepass (and its client variants, like KeepassXC which is pretty great) is even more secure because there is no server, just an encrypted file you can store anywhere.
And simultaneously less secure because it’s up to you to handle keeping your vault synced between various devices and most people are significantly worse at keeping systems secure than the professionals at the password managers.
Self hosting a server of some kind or using something like Keepass on a single device (with offline backups) is the most secure option, but as usual with security doing so trades significant convenience for security. For most people who are uninterested in making sure their servers are kept up to date week to week letting professionals handle it is the better option.
And simultaneously less secure because it’s up to you to handle keeping your vault synced between various devices and most people are significantly worse at keeping systems secure than the professionals at the password managers.
It is not less secure.
If the Bitwarden servers are compromised (either by hacking or by being forced to by the government of the country where they are hosted) then code could be run which would allow the attacker to receive your plaintext password and that is used to decrypt your data.
If a user is so horrible at syncing that they accidentally synced their database file to a public Twitter post, it is still protected by AES-256 which can’t be broken by a simple subpoena.
In either case, syncthing is pretty simple to use and is the common recommendation for the kind of small personal file sync that you need here. It also adds an additional security layer, on top of the unbreakable AES-256 encryption, to the whole setup.
I store my keypass database on several flash drives in different physical locations and update them several times per year to make sure that even if I do lose the copy I have, the versions on the flash drives, not at my physical location, are decently up to date, and so if I do lose any of the password data, it will be only for a couple of months worth if that.
If I add things that are extremely important, such as a new mortgage provider, or some sort of financial data into my keypass database, then I do an unscheduled immediate update to all of my flash drives in different physical locations to make sure that they all have that, but if it’s just a social media account, and I was to lose access to it, and not have the password for it, then… I wouldn’t be too upset about it.
In the absolute worst possible case, I stand to lose 3 months worth of data. It’s not often that I have to tweak stuff in my password manager, so that would be very few changes.
Great.
I am now your spouse and you want to give me access to the flash drive. What now?New requirement: I have several passwords I want to give you access to as well. What now?
As with everything: Your solution may work for yourself and a few others. The majority don’t want to collect 5 flash drives in different locations every 3 months to update a file (and making sure it’s the correct vault they have copied)
PThe master copy stays on my device. If I need to give somebody access to a specific password, I just give them that password locally and they put it in their password manager for that account.
Same thing occurs if they need to give me a password. They give me the password. I put it in my password manager and then I’m the one who updates the flash drives on the rotating basis like I mentioned above.
Great.
Now your data is (potentially) exactly where you are trying to keep it out of.So you made it more cumbersome to yourself by keeping your data as local as possible, yet still chosing to give up the tiny sliver of additional security for the comfort of others.
I don’t want to be annoying. But I hope you see what I am trying to convey.
I’m sorry, but I really don’t see what you’re trying to convey. The people I give my passwords to also don’t use cloud password managers.
If the password manager server is hacked and compromised, then syncing your passwords with the compromised server will lead to compromised passwords (duh)
What do you mean “duh”? The password managers claim that the exact opposite is true.
Most service providers therefore promote their products with the promise of “zero-knowledge encryption”. This means they assure users that their stored passwords are encrypted and even the providers themselves have “zero knowledge” of them and no access to what has been stored. “The promise is that even if someone is able to access the server, this does not pose a security risk to customers because the data is encrypted and therefore unreadable. We have now shown that this is not the case”, explains Matilda Backendal.
This would be true for a properly implemented end-to-end encryption scheme.
“Properly implemented” is doing the heavy lifting in that sentence.
Four paragraphs down from your quote is this:
Their attacks ranged from integrity violations affecting specific, targeted user vaults to the complete compromise of all vaults within an organisation using the service. In most cases, the researchers were able to gain access to the passwords – and even make changes to them.
If E2EE were properly implemented, the above would be impossible.
Everyone using online password manager services deserves everything he gets
Yes, let’s blame the victim and not the data hording mega corps that advertise their crap to collect more data, make big promises, hide the better options, and actively undermine open source in every way they can.
I’m pretty sure the average person hears “open source” and think “oh that’s insecure software made by hackers, I need to only use software from trusted sources”. Using only trusted software is still a good idea, but unfortunately the trusted sources of 2002 have betrayed us.
I use one of the password managers mentioned in the article, purely for the convenience of apps on all my devices, syncing and complex individual passwords. Should I be looking to move to self hosting something instead? Would my host (likely a synology Nas or raspberry pi) not then have the same risks?
I self host via vault warden. And I have it locked behind tailscale vpn. Aside from your server itself getting hacked, which is a risk, this is more secure than having passwords on the public internet.
I believe Proton Pass does not have the design flaws shown in the article. For instance, if you lose your password, you lose your data. Your data is encrypted and decrypted on your device.
This is what all the listed password manager claim.
What was done here was tricking the client through the server to do things. So the fixes went into the client application.
For people interested there were 3 cloud based password managers tested and this is what they found
The researchers demonstrated 12 attacks on Bitwarden, 7 on LastPass and 6 on Dashlane.
What I am wondering myself: Do the different amount of attacks mean the attack surface was greater or had more vulnerabilities or what made them only do 6 on Dashlane vs 12 on Bitwarden?
Edit:
In another article it was total identified vulnerabilities.Is there a reason why these attacks were on cloud based pw managers?
From what I scanned, there was no reason given on why they only attacked cloud based providers.
My guess is that these are paid ones and thus have a ‘market share’, easier to attack etc.
If you attack a ‘keepass’ password the attack vector is more crypto / memory based as far as my limited knowledge goes and not some funky inbetween attack.
Also, if you attack a cloud base provides, you will most likely have multiple victims per breach / exploit, whilst offline are targeted and thus not so interesting in most cases unless we’re talking about a person of interest
they ran the test on those pw managers because they were open source. that allowed the testers to implement a “dummy” provider on their own “compromised server.” so the results of failing the tests are based on the hypothetical situation of “what if bitwarden (or whoever) had an entire server taken over by hackers”. while the chances of that happening are greater than zero, it would take a lot for someone to completely hijack a server like that
edit to add-- these tests are one of the reasons these pw managers choose to be open source: to allow 3rd party tests like this to find vulnerabilities, so they can be fixed
nothing is 100% guaranteed safe, but if you don’t want to remember or write down dozens or hundreds of unique strong passwords, i still would recommend a pw manager
