- We have this protocol to switch the streetlights remotely by modulating a signal on the main - but that's needing expensive hardware and it's cumbersome. Can't we just sent that over radio instead?
- There is all this decentralized renewable energy generation, we need a way to switch that off remotely if there is an overload in the grid - hey, we already have that hardware for swtiching streetlamps, let's just use that!
Of course encrption was never a concern and now anyone could remotely turn off / on power generation. But for that to cause real trouble, you'd need coordinated action that would require something like a state level actor.
H8crilA 6 days ago [-]
Authentication, not necessarily encryption. It's a common misconception to think that you need the latter while you actually need the former. And no, encryption does not mean authentication, not at all, usually you can meaningfully modify the ciphertext if a given protocol has no authentication.
Also, here's a fun thought experiment: consider two channels, one authentic but not encrypted, another non authentic but encrypted. Can you actually find a use for the second one? Can you find a use for securely talking to an unknown entity, other than running Omgele? :)
tialaramex 6 days ago [-]
We should distinguish whether we want everybody to be able to authenticate the messages or only our intended recipient. This is separate from the question of whether the message should be encrypted. It may be reasonable for infrastructure to work only with messages everybody may authenticate since there is nothing to hide. For this purpose a Signature Scheme is ideal - simply sign your messages.
Whereas for example in Signal two people could have made an Alice->Bob message. Both Alice and Bob have the keys to make such a message. Alice might have made it, and sent it to Bob, or, Bob might have just made it seem as though Alice sent him a message. Bob presumably knows if he's lying, but he can't prove it either way.
The unauthenticated link is basically useless. You aren't "securely talking to an unknown entity" because if you were that would be an authenticated link. TLS 1.3 can do "securely talking to an unknown entity" - but it's an authenticated link, the unknown entity is the authenticated remote party. You don't know who they are, but you do know they're your remote party whoever that is.
H8crilA 6 days ago [-]
Maybe I should have added what an encrypted, but not authenticated link looks like, because I meant it in both directions. An example would be doing unauthenticated Diffie-Hellman (without any signatures, or proving knowledge of a secret, or anything of this sort), then proceeding using the shared key with even the best of schemes. Another example would be a spy sending and receiving one-time-pad encrypted data via an untrusted, malleable channel - the only authenticity is in hoping that adversarial modifications will cause one of the endpoints to fail at "parsing" the message. It is indeed useless.
Also, this property of Signal is called repudiation (or non-non-repudiation :) ), meaning that you as a party in the communication can repudiate the origin of the message, i.e. say that you didn't write it. It is a nice extra feature, on top of authenticity and secrecy.
tialaramex 6 days ago [-]
> then proceeding using the shared key with even the best of schemes
Unlike with your "spy" scenario, this situation is in fact what TLS 1.3 builds initially and it's not useless at all, nor is the resulting link unauthenticated.
1. First Alice sends her fresh parameters to Bob
2a. Then Bob sends his fresh parameters to Alice
2b. Alice and Bob now have all the DH parameters and they now have a shared secret X
2c. Bob calculates a Digest of a transcript of the entire conversation so far and sends this Digest, encrypted with X to Alice, he can send this alongside the parameters if he wants
3. Alice also likewise calculates a Digest and sends the encrypted digest to Bob
Both Alice and Bob now have a shared secret and have an authenticated (because they've seen the other party's view of the conversation in the digest and confirmed it matches their expectation) communication channel which they can use. They don't learn each others' identity but, of course, it is easy to additionally offer this as part of the same protocol and HTTPS does so in one direction in the typical case.
Edited: Renumbered to clarify that only three messages are sent, parts 2a, 2b and 2c are actually a single message from Bob to Alice
5 days ago [-]
H8crilA 5 days ago [-]
I meant that man in the middle attacks make this effectively unauthenticated.
tialaramex 5 days ago [-]
Suppose that Eric is in the middle.
Eric could allow those initial three messages to pass unmolested. In this case Alice and Bob now have an authenticated connection and Eric is unable to read or modify their messages. So I suppose you don't mean that.
What if Eric just substitutes his own message for Alice's in step 1? He provides his own parameters. Since these were not Alice's parameters, Alice will not provide an acceptable Digest for the conversation, the parameters Eric sent to Bob are different and do not match the transcript, the connection is terminated.
What if Eric substitutes Bob's only message in step 2? He provides his own parameters, and he can respond with a transcript digest for this alternate conversation. Now in fact the TLS 1.3 connection exists as normal, but it is between Alice and Eric. We're actually fine! We have a properly authenticated connection, with unidentified participants (we know they're Alice and Eric but Alice and Eric don't know that). Bob's connection fails, or he is unaware that Alice tried to connect.
Finally if Eric waits until Alice's second message in step 3, no message Eric knows how to construct is satisfactory. Only Alice's original message will work, other messages cause the connection to fail because Bob will not accept them.
H8crilA 5 days ago [-]
I'm sorry, are you trolling? You pretend to be Bob to Alice, and Alice to Bob, at the same time. Real Alice and Bob never compute the same transcript digests, but it doesn't matter. As post handshake data is flowing you decrypt it and reencrypt, modifying what you want.
Or, imagine that there isn't even one legitimate Bob in the world. But Alice is still talking to someone who follows the protocol, and is indistinguishable from a real Bob. Is that possible?
tialaramex 5 days ago [-]
I'm not trolling, I just don't think you've really thought about what "authentic" could mean if you don't have identity, which is a distinct problem.
For Alice there's no difference between a Bob whose identity you don't know and an Eric whose identity you don't know, they're each just another anonymous party.
Eric can't fool Alice and Bob into believing they're talking to each other, because they're using separate channels and identity systems will bind to the channel. Suppose Bob provides his identity, that's bound to the Eric-Bob channel, when Eric copies it to the Eric-Alice channel it's just obviously fraudulent, he could have made his own fraudulent claim of Bob's identity just as easily.
5 days ago [-]
npteljes 5 days ago [-]
This is very nitpicky, and not even valid at that.
First, I think authorization is even more valid than authentication. In this context, it's the authority is what is important, so that only the designated entities can assert control over the system, and not others.
Second, it's very hard to imagine authorization on an open channel like radio, without any sort of encryption. In fact, only the one-time pad comes to mind, although I'm far from being a proper security person. What I see is that authority is usually demonstrated through some encrypted means - even if the message itself is unencrypted, its digital signature is.
>another non authentic but encrypted
HTTPS is one such channel. The weakest guarantee of HTTPS is that the comms between the client and the HTTPS terminating server is encrypted, nothing more. HTTPS security can be upgraded to include authenticity information, but it's not mandatory, and it's still very useful even in this weaker form.
H8crilA 5 days ago [-]
Authentic channel is a channel typically uses digital signatures, or MACs, or something like the Dragonfly protocol used for example in WPA3. As you may know the padlock and alerts in browsers are exactly for that reason, to show that a channel is authentic. The client doesn't authenticate by default, we use passwords/API keys/oauth/etc for that. Though the client actually can, and some services use that.
> Second, it's very hard to imagine authorization on an open channel like radio, without any sort of encryption.
It's easy: you add digital signatures to sign plaintext packets/messages. And now you know they are authentic, and not generated by some random guy that bought a HackRF last week, and is now testing all firmwares that he can find.
nextaccountic 5 days ago [-]
I think the point made is that digital signatures are encrypted messages (eg. you may encrypt a hash of the message using your private key)
5 days ago [-]
5 days ago [-]
Scaevolus 5 days ago [-]
Authorization can be achieved by authentication with either a preshared key or a key derived using public key cryptography and some PKI.
It's trivial to implement on an open channel, HMAC being a common form. This is how old APIs using HTTP (Flickr, S3) handled authorization despite all communication being cleartext.
Confidentiality and integrity can be achieved with a single authenticated encryption primitive, or done separately with encryption and a hash-based authentication primitive, or exist as one without the other.
radicality 5 days ago [-]
Didn’t the parent talk about authentication vs encryption, and not authentication vs authorization ?
baobun 5 days ago [-]
You are confusing authentication (even less relevant in context than encryption actually) with authorization.
Doing proper authorization without cryptography is difficult. Mixing it up with authentication is unnecessary but popular.
The authn/authz distinction is more impactful than encryption/cryptography nitting. Signatures are just encryption with a public key, yknow?
5 days ago [-]
Muromec 6 days ago [-]
>But for that to cause real trouble, you'd need coordinated action that would require something like a state level actor.
luckily there isn't a state actor actively sabotaging all kinds of infrastructure in Europe right now with explicit interest to sabotage renewables
ForOldHack 6 days ago [-]
They could drag a boat anchor across it?
gorgoiler 6 days ago [-]
I really enjoyed how the payloads are encrypted, but the implementation leaves time synchronization in plaintext. With the street lamps that work to a fixed schedule, all you have to do is reset the time between 12pm and 12am to turn them on and off (the “lamplighter” attack, in the talk.)
happosai 6 days ago [-]
Listening to the talk, I don't think it was encrypted. They just said in early in the talk that it seemed encrypted due to high entropy. But later in the talk they decoded the payloads after they figured out the format.
But yeah, insecure time is a underrated attack vector.
pantalaimon 6 days ago [-]
As I understood it, that's likely weather data from a 3rd party (Meteocast) that they encrypt to protect their IP/subscription.
unbelauscht 6 days ago [-]
Like the one who's been messing with our deep sea cables?
mindcrime 6 days ago [-]
But for that to cause real trouble, you'd need coordinated action that would require something like a state level actor.
Or thousands of individuals using relatively inexpensive HackRF One SDR's, or home-brew radio transmitters which can be built even more cheaply. Of course all those people would need a way to communicate with each other over large distances... perhaps some kind of packet switching network running over a series of tubes (or avian carriers). Hmmm.
gunian 5 days ago [-]
[dead]
Eduard 6 days ago [-]
TL;DR: by law, German power stations are required to "turn off" (taken off the energy grid) when they receive specific radio messages. This is intended for energy grid load balancing.
Unfortunately, the message protocol is completely flawed security-wise, which allows malicious actors to control the power station.
It would require only a handful of strategically placed senders to control an estimated 20 gigawatt of load Germany-wide, causing havoc on the European energy grid (brown-out, cascading effects, etc.).
The security researchers followed a responsible disclosure towards the vendor, EFR, who reacted with sending letters from their lawyers.
Today's SPIEGEL online news magazine pre-talk report ( https://archive.is/p66as ) on this topic cites EFR that the proposed attack vector is not possible.
The security researchers therefore made the last minute decision to go full disclosure with today's talk to press on the urgency of the topic.
jdiez17 6 days ago [-]
Just read the SPIEGEL article and I think it’s a pretty balanced report on the positions of both sides. Basically, it comes down to the assertion that you can’t reach a large number of electricity generation plants with “simple radio equipment”. That is the position of EFR, and sadly, the Bundesnetzagentur (the radio communications regulator in Germany).
I haven’t watched the talk yet but I think it’s pretty clear to all of us on this website, that sending a specific short radio transmission to a large area is not an insurmountable challenge for our favorite terrorist state.
What I don’t understand is why there is such a reluctance to admit that these problems exist and work towards fixing them. Instead we pull the Ostrich maneuver every time. One day it’s going to really bite us in the ass.
EDIT: after watching the talk, the funny thing is that all of the “business secrets” that EFR is accusing our fellow hackers of leaking, are actually mostly DIN standards. In other words, they are just upset that someone is talking about the fact that no efforts have been made to proactively secure these receivers. Peinlich.
semi-extrinsic 6 days ago [-]
IANAL, and didn't watch the full recording yet. But if the EFR lawyers are threatening the hackers with "leaking business secrets", they have to be wildly incompetent. I won't give those guys any ideas, but I'm certain there are much more scary parts of DE/EU law that you could threaten with.
Etheryte 6 days ago [-]
Ass covering, so much ass covering everywhere. I've done a fair bit of consulting for the public sector and figuring out their office politics is often the only real way to get anything done, the actual technical discussion is often secondary.
trebligdivad 6 days ago [-]
I think they kind of have a point; they were talking about needing a 10kW transmitter - that's a heck of a lot of power for a transmitter, not easy to make at all. And at those frequencies, the antenna is a challenge.
Having said that, a bunch of few-hundred W transmitters in convenient places would be a lot easier, and there are probably easy but inefficient antenna hacks (drop a wire down a cliff/across a park/out of the top floor of a tower block?)
grumpy-de-sre 6 days ago [-]
I beg to disagree, 10kW at ~140khz is actually relatively straight forward with modern semiconductors and LiPo's. Eg. the inverters in a Tesla Plaid can do up-to 750kW, so I think two orders of magnitude more power is theoretically possible.
And then they left out that at such long wavelengths there are some unconventional antenna topologies available. Some of which are a lot more feasible than anything that was discussed in the talk.
The dismissal is quite concerning IMO.
Eduard 6 days ago [-]
Also, IMHO instead of a few strong senders, an attacker could use more low-powered senders placed in proximity to power stations.
randunel 6 days ago [-]
Insurmountable? How many russian citizens live in Germany? And how many russian fanboys with a non-russian citizenship? Now extend that to neighbouring countries, or the Schengen area.
jdiez17 6 days ago [-]
You may want to reread my post.
aunderscored 6 days ago [-]
Saw this in person, awesome look at street lamp control and then walking that all the way up to "oops we figured out a way to attack the European power grid"
BonoboIO 6 days ago [-]
What a great way for a state to cause havoc in all of Europe.
Russia definitely has the capabilities to send such signals in a coordinated attack and deny an wrong doing.
And this is just one example we know of, there must be hundreds.
ElectRabbit 5 days ago [-]
They have low-kHz transmitters for reaching submarines. So, for many decades already.
oger 5 days ago [-]
The researchers did a great job in pointing out the failures in what basically is an old DIN standard that should not be used in this century. I congratulated them after the talk as I did similar research and didn’t get it finished for 38C8. Their presentation is spot on. The attack vector is definitely feasible and publicly known for a while. I honestly don’t understand why nobody in the industry wanted to switch to a safer alternative.
The reaction by EFR will create an unnecessary Streisand effect and after all they will be able to upsell their customers to a (soon to be legacy) 450 MHz LTE system.
matchamatcha 6 days ago [-]
Talk starts around ~16:20 minutes in..
Torkel 6 days ago [-]
And the talk itself is in English.
__jonas 6 days ago [-]
That was an interesting talk!
I'm not very familiar with security stuff, but I didn't really get the responsible disclosure thing – is it really unreasonable for this company to ask them not to go public just three months after their initial disclosure?
I understand the 'it was known since 2013' thing, but they did also say the company was actively making improvements after the initial disclosure so they were not exactly just shoving it under the rug were they?
Hikikomori 6 days ago [-]
They got letter from their lawyers no?
__jonas 6 days ago [-]
Yeah? I’m saying I don’t get why the letter from the lawyer is unreasonable.
Sure, ideally it would have not been done via a lawyer but rather just asking them to delay going public directly since they were communicating before, but still it’s just three months after initial disclosure and they were actively making improvements and informing customers that they need to switch out hardware which I assume takes time, I think not wanting the researchers to go public just yet is pretty reasonable no?
Am I missing something?
As I said I’m not very familiar with security research stuff, maybe anything goes three months after disclosure, it just surprises me.
Also just to be clear: the work by the researchers here is super impressive, and it’s fantastic that they are doing it, I was just wondering about this disclosure process.
aunderscored 6 days ago [-]
If you always allow a company to say "wait no don't" with issues, it gives them a tool to quiet problems without solving them. Responsible disclosure is a tool , and part of that tool is the understanding that this will be public
Towaway69 5 days ago [-]
Are there any pointers to the software they built for the flipper?
It seems that they did create an app but it’s nowhere to be found on the flipper “app store”.
Why do we still build new remotely controlled things and then skip security? Like when was this ever a good idea?
avidiax 6 days ago [-]
I think it's a failure to solve 1 + x = 2. x is the percentage of the power grid controlled by this system, which has risen over time.
So at design time, the threat is just that people can turn off street lamps, which you can do with a BB gun. Then you expand to home solar. Also not so interesting.
But then you expand to be a significant fraction of the grid supply and load. Now there is a substantial target that actually needs security, but which requires a full redesign.
Rendered at 07:30:54 GMT+0000 (UTC) with Wasmer Edge.
- We have this protocol to switch the streetlights remotely by modulating a signal on the main - but that's needing expensive hardware and it's cumbersome. Can't we just sent that over radio instead?
- There is all this decentralized renewable energy generation, we need a way to switch that off remotely if there is an overload in the grid - hey, we already have that hardware for swtiching streetlamps, let's just use that!
Of course encrption was never a concern and now anyone could remotely turn off / on power generation. But for that to cause real trouble, you'd need coordinated action that would require something like a state level actor.
Also, here's a fun thought experiment: consider two channels, one authentic but not encrypted, another non authentic but encrypted. Can you actually find a use for the second one? Can you find a use for securely talking to an unknown entity, other than running Omgele? :)
Whereas for example in Signal two people could have made an Alice->Bob message. Both Alice and Bob have the keys to make such a message. Alice might have made it, and sent it to Bob, or, Bob might have just made it seem as though Alice sent him a message. Bob presumably knows if he's lying, but he can't prove it either way.
The unauthenticated link is basically useless. You aren't "securely talking to an unknown entity" because if you were that would be an authenticated link. TLS 1.3 can do "securely talking to an unknown entity" - but it's an authenticated link, the unknown entity is the authenticated remote party. You don't know who they are, but you do know they're your remote party whoever that is.
Also, this property of Signal is called repudiation (or non-non-repudiation :) ), meaning that you as a party in the communication can repudiate the origin of the message, i.e. say that you didn't write it. It is a nice extra feature, on top of authenticity and secrecy.
Unlike with your "spy" scenario, this situation is in fact what TLS 1.3 builds initially and it's not useless at all, nor is the resulting link unauthenticated.
1. First Alice sends her fresh parameters to Bob
2a. Then Bob sends his fresh parameters to Alice
2b. Alice and Bob now have all the DH parameters and they now have a shared secret X
2c. Bob calculates a Digest of a transcript of the entire conversation so far and sends this Digest, encrypted with X to Alice, he can send this alongside the parameters if he wants
3. Alice also likewise calculates a Digest and sends the encrypted digest to Bob
Both Alice and Bob now have a shared secret and have an authenticated (because they've seen the other party's view of the conversation in the digest and confirmed it matches their expectation) communication channel which they can use. They don't learn each others' identity but, of course, it is easy to additionally offer this as part of the same protocol and HTTPS does so in one direction in the typical case.
Edited: Renumbered to clarify that only three messages are sent, parts 2a, 2b and 2c are actually a single message from Bob to Alice
Eric could allow those initial three messages to pass unmolested. In this case Alice and Bob now have an authenticated connection and Eric is unable to read or modify their messages. So I suppose you don't mean that.
What if Eric just substitutes his own message for Alice's in step 1? He provides his own parameters. Since these were not Alice's parameters, Alice will not provide an acceptable Digest for the conversation, the parameters Eric sent to Bob are different and do not match the transcript, the connection is terminated.
What if Eric substitutes Bob's only message in step 2? He provides his own parameters, and he can respond with a transcript digest for this alternate conversation. Now in fact the TLS 1.3 connection exists as normal, but it is between Alice and Eric. We're actually fine! We have a properly authenticated connection, with unidentified participants (we know they're Alice and Eric but Alice and Eric don't know that). Bob's connection fails, or he is unaware that Alice tried to connect.
Finally if Eric waits until Alice's second message in step 3, no message Eric knows how to construct is satisfactory. Only Alice's original message will work, other messages cause the connection to fail because Bob will not accept them.
Or, imagine that there isn't even one legitimate Bob in the world. But Alice is still talking to someone who follows the protocol, and is indistinguishable from a real Bob. Is that possible?
For Alice there's no difference between a Bob whose identity you don't know and an Eric whose identity you don't know, they're each just another anonymous party.
Eric can't fool Alice and Bob into believing they're talking to each other, because they're using separate channels and identity systems will bind to the channel. Suppose Bob provides his identity, that's bound to the Eric-Bob channel, when Eric copies it to the Eric-Alice channel it's just obviously fraudulent, he could have made his own fraudulent claim of Bob's identity just as easily.
First, I think authorization is even more valid than authentication. In this context, it's the authority is what is important, so that only the designated entities can assert control over the system, and not others.
Second, it's very hard to imagine authorization on an open channel like radio, without any sort of encryption. In fact, only the one-time pad comes to mind, although I'm far from being a proper security person. What I see is that authority is usually demonstrated through some encrypted means - even if the message itself is unencrypted, its digital signature is.
>another non authentic but encrypted HTTPS is one such channel. The weakest guarantee of HTTPS is that the comms between the client and the HTTPS terminating server is encrypted, nothing more. HTTPS security can be upgraded to include authenticity information, but it's not mandatory, and it's still very useful even in this weaker form.
> Second, it's very hard to imagine authorization on an open channel like radio, without any sort of encryption.
It's easy: you add digital signatures to sign plaintext packets/messages. And now you know they are authentic, and not generated by some random guy that bought a HackRF last week, and is now testing all firmwares that he can find.
It's trivial to implement on an open channel, HMAC being a common form. This is how old APIs using HTTP (Flickr, S3) handled authorization despite all communication being cleartext.
Confidentiality and integrity can be achieved with a single authenticated encryption primitive, or done separately with encryption and a hash-based authentication primitive, or exist as one without the other.
Doing proper authorization without cryptography is difficult. Mixing it up with authentication is unnecessary but popular.
The authn/authz distinction is more impactful than encryption/cryptography nitting. Signatures are just encryption with a public key, yknow?
luckily there isn't a state actor actively sabotaging all kinds of infrastructure in Europe right now with explicit interest to sabotage renewables
But yeah, insecure time is a underrated attack vector.
Or thousands of individuals using relatively inexpensive HackRF One SDR's, or home-brew radio transmitters which can be built even more cheaply. Of course all those people would need a way to communicate with each other over large distances... perhaps some kind of packet switching network running over a series of tubes (or avian carriers). Hmmm.
Unfortunately, the message protocol is completely flawed security-wise, which allows malicious actors to control the power station.
It would require only a handful of strategically placed senders to control an estimated 20 gigawatt of load Germany-wide, causing havoc on the European energy grid (brown-out, cascading effects, etc.).
The security researchers followed a responsible disclosure towards the vendor, EFR, who reacted with sending letters from their lawyers.
Today's SPIEGEL online news magazine pre-talk report ( https://archive.is/p66as ) on this topic cites EFR that the proposed attack vector is not possible.
The security researchers therefore made the last minute decision to go full disclosure with today's talk to press on the urgency of the topic.
I haven’t watched the talk yet but I think it’s pretty clear to all of us on this website, that sending a specific short radio transmission to a large area is not an insurmountable challenge for our favorite terrorist state.
What I don’t understand is why there is such a reluctance to admit that these problems exist and work towards fixing them. Instead we pull the Ostrich maneuver every time. One day it’s going to really bite us in the ass.
EDIT: after watching the talk, the funny thing is that all of the “business secrets” that EFR is accusing our fellow hackers of leaking, are actually mostly DIN standards. In other words, they are just upset that someone is talking about the fact that no efforts have been made to proactively secure these receivers. Peinlich.
And then they left out that at such long wavelengths there are some unconventional antenna topologies available. Some of which are a lot more feasible than anything that was discussed in the talk.
The dismissal is quite concerning IMO.
Russia definitely has the capabilities to send such signals in a coordinated attack and deny an wrong doing.
And this is just one example we know of, there must be hundreds.
I'm not very familiar with security stuff, but I didn't really get the responsible disclosure thing – is it really unreasonable for this company to ask them not to go public just three months after their initial disclosure?
I understand the 'it was known since 2013' thing, but they did also say the company was actively making improvements after the initial disclosure so they were not exactly just shoving it under the rug were they?
Sure, ideally it would have not been done via a lawyer but rather just asking them to delay going public directly since they were communicating before, but still it’s just three months after initial disclosure and they were actively making improvements and informing customers that they need to switch out hardware which I assume takes time, I think not wanting the researchers to go public just yet is pretty reasonable no? Am I missing something?
As I said I’m not very familiar with security research stuff, maybe anything goes three months after disclosure, it just surprises me.
Also just to be clear: the work by the researchers here is super impressive, and it’s fantastic that they are doing it, I was just wondering about this disclosure process.
It seems that they did create an app but it’s nowhere to be found on the flipper “app store”.
Nothing there for Blinkencity yet.
So at design time, the threat is just that people can turn off street lamps, which you can do with a BB gun. Then you expand to home solar. Also not so interesting.
But then you expand to be a significant fraction of the grid supply and load. Now there is a substantial target that actually needs security, but which requires a full redesign.