Live data from Hacker News

Signal PINs

signal.org

101–110 of 199 posts

Re: Signal PINs

#101
post #71

Earlier quoted context omitted.

Interesting, mine keeps nagging me at the bottom of the screen every day, but hasn't forced it on me at all yet.

Samezies

At first mine behaved like parent, but then behaved like grandparent. Perhaps the was another update?

Re: Signal PINs

#102
post #35

Earlier quoted context omitted.

> I don't care about their Intel SGX whatever because I don't want to have to trust their servers in the first place. Yeah: particularly given how Intel SGX gets broken every year :/... prime+probe, foreshadow, load value injection, plundervolt... Moxie's fetish for Intel SGX is extremely concerning.

SGX has simply just been shown to be incompatible with secure software and is impossible to justify. Signal can happily operate without the need for this sort of thing to begin with, but they keep insisting on weak solutions for no reason.

These weak solutions are actually enough to keep the government requests at bay, for the moment. Why go into crazy crypto theory if that's good enough right? That's what Apple and Google are doing as well.

Re: Signal PINs

#103
post #6

This is all well and good. But the forced/unsilenceable pin reminders seems a bit obtuse to me. https://support.signal.org/hc/en-us/articles/360007059792-Si... "Can I turn off these reminders? It is important to memorize your PIN, and the reminders cannot be disabled. We cannot recover your PIN if you forget it. You will see the reminders less frequently if you consistently enter your PIN correctly. The reminders wil…

I actually like that it forces you to remember the PIN.

Re: Signal PINs

#104
post #68

I've been using Signal PINs for a long time to lock the app to my SIM card and unfortunately they are a real pain. This sounds exactly the same. First off the app is incessant about asking you to enter your PIN to prove you know it; this prompt is supposed to get less frequent and I suppose it does but is still way too frequent. Some of us are competent at storing secrets in a password manager and this is like a puni…

I've found that tapping next to the registration lock PIN prompt will dismiss it on Android. A "dismiss" button and a "don't ask me again, I won't lose this" checkmark would make much more sense to me.

how hard is it to enter 4 digits right

Re: Signal PINs

#105
post #90

Earlier quoted context omitted.

Basically, SGX: https://github.com/signalapp/SecureValueRecovery The enclave is hardened against Spectre and LVI with this (BOLT/LLVM based), and other techniques: https://github.com/signalapp/BOLT The last build step before signing is a verifier that checks that there are no missed mitigations, built using Intel Xed, to try to avoid potential missing mitigations due to an LLVM or BOLT bug.

I'm familiar with Signal's usage of enclaves. The case I'm wondering about is what happens if someone seized (or surreptitiously accessed) Signal's servers, ran the unmodified enclaves, but fed in different PIN requests to those enclaves in a brute-forcing attempt. What prevents that?

The enclaves will flush the data (much like the hardware you mention in your first post) after some number of wrong PIN guesses.

To do this securely made the whole thing quite difficult, because it means we can never persist anything outside the enclave, or an attacker could unpersist into 1000 unmodified enclaves (who would never know about each other) and get 1000*N guesses.

Re: Signal PINs

#106
post #90

Earlier quoted context omitted.

Basically, SGX: https://github.com/signalapp/SecureValueRecovery The enclave is hardened against Spectre and LVI with this (BOLT/LLVM based), and other techniques: https://github.com/signalapp/BOLT The last build step before signing is a verifier that checks that there are no missed mitigations, built using Intel Xed, to try to avoid potential missing mitigations due to an LLVM or BOLT bug.

I'm familiar with Signal's usage of enclaves. The case I'm wondering about is what happens if someone seized (or surreptitiously accessed) Signal's servers, ran the unmodified enclaves, but fed in different PIN requests to those enclaves in a brute-forcing attempt. What prevents that?

if it's anything like what other big players are doing: the attempts are limited and the code will destroy its knowledge of the key once you reach this limit.

Looking at the code quickly I see that nothing is documented and it's not clear what code gets executed there. So good luck understanding what they're doing. But it seems to be using Raft, probably to bring several HSMs to consensus on what's the state of the number of attempts. And it does seem to erase stuff if you reach some threshold of attempts: https://github.com/signalapp/SecureValueRecovery/blob/00d023...

My follow up question is: what prevents someone from destroying your account's backup key by entering wrong PINs for your account?

Re: Signal PINs

#107
post #6

This is all well and good. But the forced/unsilenceable pin reminders seems a bit obtuse to me. https://support.signal.org/hc/en-us/articles/360007059792-Si... "Can I turn off these reminders? It is important to memorize your PIN, and the reminders cannot be disabled. We cannot recover your PIN if you forget it. You will see the reminders less frequently if you consistently enter your PIN correctly. The reminders wil…

It is beyond comprehension that they think pin reminders being mandatory is ok. I have been a huge fan of signal from the beginning and this is single handily a deal breaker and has me looking for alternatives.

How did they not learn from this when they tried it with the activation lock.

Re: Signal PINs

#108

I'm extremely disappointed about how this Signal PIN rollout has been handled. Signal refused to let me view my received messages until I created a PIN. I filed a bug report about it [1]. I don't know if I was just caught in an A/B test or what because it hasn't happened to all of my Signal-using friends but it happened to me. I don't understand why it isn't just optional. They claim they want to protect my Signal da…

> I don't understand why it isn't just optional.

they probably got served with a secret order from the gov.

Re: Signal PINs

#109
post #89

Earlier quoted context omitted.

What are they for?

Read the blog post, friend! They're for restoring your data to a new device.

Still stupid, shouldn't they ask if I want that functionality? Not like it's going to be hard to brute force a user's PIN.

They should explain what the pins is BEFORE they ask. What it looks like is you are going to be locked out of signal if you make a mistake. Not like the average signal user is watching the whisper systems blog.

Re: Signal PINs

#110
post #109

Earlier quoted context omitted.

Read the blog post, friend! They're for restoring your data to a new device.

Still stupid, shouldn't they ask if I want that functionality? Not like it's going to be hard to brute force a user's PIN. They should explain what the pins is BEFORE they ask. What it looks like is you are going to be locked out of signal if you make a mistake. Not like the average signal user is watching the whisper systems blog.

> Not like it's going to be hard to brute force a user's PIN.

I guess it depends on if you buy into SGX. Lots of people don't.

> Still stupid

Yeah, this isn't making anyone happy it seems. Still, they are a small organization and having configuration increases maintenance burden. As long as Signal keeps operating as non-commercial product, I will keep cutting them slack.

Post reply on HN