Slashdot Mirror
Severe Vulnerabilities Uncovered In Popular Password Managers (zdnet.com)
chiefcrash shares a report from ZDNet: Independent Security Evaluators (ISE) published an assessment on Tuesday with the results of testing with several popular password managers, including LastPass and KeePass. The team said that each password management solution "failed to provide the security to safeguard a user's passwords as advertised" and "fundamental flaws" were found that "exposed the data they are designed to protect."
The vulnerabilities were found in software operating on Windows 10 systems. In one example, the master password which users need to use to access their cache of credentials was stored in PC RAM in a plaintext, readable format. ISE was able to extract these passwords and other login credentials from memory while the password manager in question was locked. It may be possible that malicious programs downloaded to the same machine by threat actors could do the same. The report has summarized the main findings based on each password management solution. Here's what ISE had to say about LastPass and KeePass -- two of the most popular password managers available:
"LastPass obfuscates the master password while users are typing in the entry, and when the password manager enters an unlocked state, database entries are only decrypted into memory when there is user interaction. However, ISE reported that these entries persist in memory after the software enters a locked state. It was also possible for the researchers to extract the master password and interacted-with password entries due to a memory leak."
"KeePass scrubs the master password from memory and is not recoverable. However, errors in workflows permitted the researchers from extracting credential entries which have been interacted with. In the case of Windows APIs, sometimes, various memory buffers which contain decrypted entries may not be scrubbed correctly."
The vulnerabilities were found in software operating on Windows 10 systems. In one example, the master password which users need to use to access their cache of credentials was stored in PC RAM in a plaintext, readable format. ISE was able to extract these passwords and other login credentials from memory while the password manager in question was locked. It may be possible that malicious programs downloaded to the same machine by threat actors could do the same. The report has summarized the main findings based on each password management solution. Here's what ISE had to say about LastPass and KeePass -- two of the most popular password managers available:
"LastPass obfuscates the master password while users are typing in the entry, and when the password manager enters an unlocked state, database entries are only decrypted into memory when there is user interaction. However, ISE reported that these entries persist in memory after the software enters a locked state. It was also possible for the researchers to extract the master password and interacted-with password entries due to a memory leak."
"KeePass scrubs the master password from memory and is not recoverable. However, errors in workflows permitted the researchers from extracting credential entries which have been interacted with. In the case of Windows APIs, sometimes, various memory buffers which contain decrypted entries may not be scrubbed correctly."
Fuck lazy horse batteries.
Are there any decent USB stick based password vaults? Something that stores credentials internally and manages decryption after entering the master password. You’d still need to take care that the master password or decrypted credentials don’t linger in memory, but I’d feel better having the master data offline instead of having everything floating around in the cloud.
Even better would be an unlock pin (or fingerprint) to be entered on the USB stick itself.
If construction was anything like programming, an incorrectly fitted lock would bring down the entire building...
So security researchers are scraping the bottom of the barrel to such an extent that having access to program data when you have total control over a computers memory is a severe vulnerability now?
If a malicious software can access the RAM, my computer is already compromised.
If my computer is compromised, it's much easier to steal the passwords when I'm saving it on the clipboard and pasting it on a website. Or having the keys logged when I'm typing the master password.
Twat did you say? I cunt hear you!
Your Password Safe should be the oldest and least feature ridden software on your computer, old and crusty is good. Forget all of the fancy cloud-auto-fill-automate-everything. Like they say: KISS
If I understand these two "vulnerabilities" properly, they require a piece of software installed/running locally which will steal/grab these passwords from RAM. However no normal/legitimate software will ever steal your passwords or access the RAM regions of other applications, which means this software is in essence malware which means you're already completely fucked and this software may just steal your master passwords, retreive all files, etc. etc. etc.
That's why I always use a yuge password: 1234abcd. It's a very good password. The best password, really.
Corruption is convincing someone that the selfless ideal is the same as their selfish ideal.
Like with FinTS.
You have a smard card that is its own computer, and a physically secured reader that lets you tell if it has been tampered with, with its own screen and keypad.
The card has a secure channel to the bank that not the computer and not even the reader can access (only forward), and a channel to the reader.
The reader accepts an action (like a money transfer) request from the PC, displays the amount and recipient of the money, and lets you input a PIN. Then the card uses the internal key and the PIN instead of a TAN, to communicate with the bank over the secure channel, authenticate itself and order the action. All th reader and PC can do at this point, is block communication. In whch case no action is happening. They cannot communicate such matters with the bank themselves at al!
But all those things are of course pontless, if the manufacturer of the card, (maybe the reader), the bank servers, the CA, or the confines of your own home and PC aren't checked for trustworthiness by yourself.
Remember
1. there are dopant-level hardware trojans now, and
2. there is always the $3 wrench.
Yeah, you can destroy the card with a dead man switch by not triggering a refresh when being captured, or destroy it manually, ... but you're still gonna die from blunt wrench trauma. It's how my grandpa actually died. (Except it was an axe.)
And yeah you can make your own CPU. I did. But it won't be of much use, unless you can build your own chip making machines from scratch at a secured secret location.
Sooo... That's why pros make a threat assessment. See who, not what, has interests in such a thing, and what power they actually have over you.
You e.g. cannot solve an evil goverenment with technology. It's the goons actually following the evil leader that are the real physical threat. Without them, even the worst evil overlord is a joke.
"Breaking news - software reads data from memory"
This just in, you either earn the label of being secure or you regret the day you ever suggested it.
To absolutely no one's shocked surprise, secure password managers aren't secure because caching echos....
Alan?
Remember
1. there are dopant-level hardware trojans now, and
2. there is always the $3 wrench.
I'm not that worried about the wrench. It only comes into play in a targeted attack against me.
What I am more worried about are large scale fishing expeditions where I'm just one victim among many.
I do not trust my computer so any password management system has to be air gapped.
If your computer already has malicious software running on it, you're already screwed.
There is no way a password manager can protect against that. The malware could simply put up a screen looking like the master-unlock screen and wait for the user to type the password in. It could be pixel-for-pixel the same, so there would be no way to know.
This is why I just keep passwords in a text file on disk. With full disk encryption enabled they're even safe! (as long as the computer is off of course)
A simple org-mode (i.e. text) file encrypted with GNUPG... I'm confident enough in garbage collection, while there is no ram "scrub".
Apart of that ladies and gentleman consider a thing: your (and so mine) CPU have tons of crappy code we do not know, even worst
our's motherboards contains shitload of crap we do not know... I bet those are FAR bigger threat than any FOSS password managers.
Also having tons of different services (on-line, or on someone else computer normally) is a threat by itself. Ask yourself if you really
need such enormous mass of accounts of only few are enough and so you can manage your passwords in your mind.
Sorry for my English.
That's why I keep my passwords on a sticky note on my monitor! Never trust the cloud!
"However, errors in workflows permitted the researchers from extracting credential entries which have been interacted with."
The sentence above is from the last para of the post. "permitted the researchesr from extracting. Permitted from? Huh?
Every one of these "severe" vulnerabilities requires admin privileges. Why would I go digging though scraps of ram to find a password when I could just access the keyboard buffer?
"However, errors in workflows permitted the researchers from extracting credential entries which have been interacted with."
You don't need root/administrator, attacker can do DMA attacks via Expresscard, Firewire etc.
Are you kidding? That's easy, don't use system fonts to display the password on-screen. It takes a bit of effort to create letters from graphic elements like lines and semi-circles but it's much safer (/-\ = A, etc). You could even randomize the angles and lengths of the line segments slightly (like a captcha) to prevent automated pattern recognition.
Support Right To Repair Legislation.
Apple? Thats all good right? And Linux? All good?
Domestic spying is now "Benign Information Gathering"
Once someone has logged into a website you could also grab their cookies from memory or their NTLM hash or ... without the need of the actual password.
CVSS v3 Vector
AV:L/AC:H/PR:H/UI:R/S:U/C:L/I:N/A:N
CVSS Base Score:
1.8
Not a severe issue especially since low-hanging fruit is attempted by the password managers. Perhaps they will add more extensive methods, but each OS version will require custom API specific code for each scenario.
Media over-hyped.
Independent Security Evaluators (ISE) published an assessment on Tuesday with the results of testing with several popular password managers.
I have to snicker that anyone would fail so spectacularly. They realized just now that memory has to hold field data at some point?
Laws are rules for the court, but merely a bottom bar to hit for life. Think beyond laws in your actions always.
Bruce Schneier, thank you for the fish!
You're not worried because biometrics tied to your bank accounts are virtually unheard of. Make no mistake, people will be missing fingers if it becomes common.
Sides, identity is not a password.
I've been warning about this for a while. Everyone called me an idiot or paranoid. You can see it happening w/ keepass for example using windbg, the password is showing up multiple times in the pagefile for example. They decrypt the string a lot and don't properly scrub it. This happens frequently for example when gauging the "bit" strength of a password and when decrypting to copy to clipboard etc. There is a lot of moving around in memory that is unnecessary. And, no attempt is made to virtual lock the byte buffers. They rely on SecureString to protect them without really understanding the implications of its use. I never checked lastpass because they were compromised only a year after their business began and their db was stolen.
I finally decided to try it a couple of years ago. I got it all set up with about a dozen passwords.
The next time I opened it, they were all gone. No trace. I uninstalled it and went back to memorizing my passwords.
I find those tools as a single point of failure. I have a password scheme that I use, and keep a list of plain-text reminders for each site. The reminder is so vague that nobody could figure out the password, but I instantly know which scheme I use.
I still remember a password that a departing intern told me back in 1995. It was for one of our test systems, in case I needed it. It was the first letter of each word of a song lyric, and it still meets all modern password requirements. I have never used that password myself, but remember it to this day. I don't even need a hint.. but I could easily write down a simple fact about the artist if I ever needed to.
My beliefs do not require that you agree with them.
So what you're saying is, people who care about security in the first place are all unaffected. Nothing to see here.
There are two types of attacks against systems like this.
1 where the attacker modifies the system, hopes the victim doesn't notice and then steals information when the victim next uses the system
2 where the attacker steals the system and then tries to extract information
These attacks are against the latter, where I steal your laptop and then try and extract your passwords from the running machine. If your password manager is open and unlocked, then I can trivially get your passwords, but if the manager has been closed, then these attacks could reveal your passwords.
I once tried to bid on writing the standard for Canadian Interac point of sale devices. The spec at the time failed to make this distinction.
The problem with airgapped password storage is that if you don't absolutely despise typing your password in every time it's needed, it isn't long enough, isn't random enough, or you're a masochist.
Write some stuff in code. Don't put your passwords in any password manager perfectly, instead you have a few words you memorize to slap in the middle of a password... but instead of those words, you use other words or numbers that remind you of those memorized words.
Example being a real password of
123slashdot456
you put your codeword in there as
123snake456
says someone with no money. Biometrics do come into play when we're talking about fuck you amounts of money.
It doesn't look like the best one, Bitwarden, is affected.
Yes, of course. The good, old TLA infinite-budget porn.
Your position in the security food chain determines how much they are willing to spend. Even well-healed Q-class spooks answer to an ROI at scale.
Of course, part of the signal about your rung on the security food chain is determined by how effectively you armour yourself with effective prophylaxis.
This is why security culture can only work as a public good, wherein everyone on principle uses the highest caliber of security practical. When security is practiced exclusively on an as-needed basis, it only helps to paint a more accurate bull's eye on your backside.
All the TLAs must surely love the useful idiots who distract from the economic model that prevails here, by ranting at high pitch about naked capabilities, as there are no endemic constraints on their side of the fence.
What If your cpu is hacked or bad guy turns it on or can logon by some method. Are all your logons and passwords now available via selecting the LastPass icon? Hacker gets on and, opens yahoo mail, then pulls down the LastPass "recently used" or matching sights" ? Is hacker into much of your stuff??
The standard unix password manager is in many aspects more secure than the bloated ones: https://www.passwordstore.org/
- It is minimal. It is a short bash script, that you can read completely before using it.
- It uses standard tools like gpg for storage and pwgen for password generation
-It has a simple command line with some uncomplicated graphical frontends
- It does not leave anything in memory, as it terminates when it finished copying the password to your clipboard or writing it to the console (or in the pipe of some other program).
On *NIX systems running X11, you can use xdotool to 'type' a password as if it were typed from a keyboard. I use pass (password-store) with a script that decrypts through xdotool. The clipboard never gets used. I wonder if other OSes have a similar capability