AP1 ONLY Please -- life after 2018.50.6

[jaitch]

Tesla added the PIN to drive because the key fobs used an algorithm that used a relatively low bit encryption. Some cars in Europe were stolen by some sophisticated thieves. It appears they recorded a number of key fob presses, then figured out the algorithm from that. As soon as they got access to the car using a lap top that generated the code the car was expecting they turned off all communication from the car to the outside world so neither Tesla nor the owner could find it and drove it off.

I suspect the key fob update implements a more complex algorithm for communication that makes it harder for would be car thieves to hack the key.

The thefts we have seen here in UK are relay attack and not algorithm related and pin to drive was the solution. Not sure about Tesla but some newer fobs have motion on/off so relay attacks are not possible.

Relay car theft: What is it, how can you prevent it, and will your car insurance cover it?

 

[wdolson]

The thefts we have seen here in UK are relay attack and not algorithm related and pin to drive was the solution. Not sure about Tesla but some newer fobs have motion on/off so relay attacks are not possible.

Relay car theft: What is it, how can you prevent it, and will your car insurance cover it?

I believe some Teslas on the continent were stolen from places other than homes like office car parks and it's thought that the algorithm was hacked. Key fobs use a two part code every key press. There is a part that is always transmitted in a fixed pattern with a changing pattern tacked on that cycles through a set of changes that are calculated by an algorithm run on both the key fob and the car.

If someone is trying to hack the car, the key fob will not work until you hit buttons a couple of times and get the car and key fob back on the same part of the algorithm. I believe you could also get them out of sync by clicking the key fob out of range of the car.

At least on the Model S a few years ago Tesla used an unusually weak encryption to the algorithm so hackers monitoring the key and car could eventually crack it. Hackers could set up a key listening post in an office car park and monitor communication between keys and car for a few days or weeks to crack the code for cars that park in the lot every day. Once they crack a car, they can steal it with an ersatz key fob.

I believe relay theft can be made more difficult with handshakes between key and car that are timed. Relay equipment introduce a little more delay than the key directly as it passes through the electronics of the amplifier. If the car pings the key and gets an answer in the expected time frame, it assumes it is dealing with the key directly, but if the answer has too much delay, it assumes it is a relay hack and it won't respond.

For cars that require some kind of key press to unlock the doors a simple relay hack isn't going to work. The auto presenting door handles on the Model S can be disabled and I believe the auto unlock feature on the Model X can be disabled too. I don't know about the Model 3. The key card has even shorter range than the fob so a relay hack would be tough to pull off.

A relay hack for Teslas at home in the US would be difficult because a large percentage of US Tesla owners keep their cars in garages. At our house neither of the keys are anywhere near an outside wall a hacker could get their repeater close to either.

 

[dark cloud]

here is the best article I have found regarding the Tesla key fob security history:

Hackers Could Steal a Tesla Model S by Cloning Its Key Fob—Again

But my question remains: If this 2019.32.1 update contains a "new" security update for the fob, how could my V1 fob already be updated?

Or did it actually update and Tesla just screwed up the wording where it should not say "key fob (V1) already up to date" and should instead say " key fob (V1) updated" ?

 

[Droschke]

here is the best article I have found regarding the Tesla key fob security history:

Hackers Could Steal a Tesla Model S by Cloning Its Key Fob—Again

But my question remains: If this 2019.32.1 update contains a "new" security update for the fob, how could my V1 fob already be updated?

Or did it actually update and Tesla just screwed up the wording where it should not say "key fob (V1) already up to date" and should instead say " key fob (V1) updated" ?

Agree. I got the same prompts and it's confusing. I believe the wording of the message is not right. BTW, I did the update twice and I think the first time said it's up to date whereas the second time it did a search for the fob and the message was something like "it's already updated". I wish I took pictures.

On Edit: I think the key fob update tab should give us the last update version/date, like the OTA sw updates under the software tab, so one can confirm.

 

[Droschke]

I believe some Teslas on the continent were stolen from places other than homes like office car parks and it's thought that the algorithm was hacked. Key fobs use a two part code every key press. There is a part that is always transmitted in a fixed pattern with a changing pattern tacked on that cycles through a set of changes that are calculated by an algorithm run on both the key fob and the car.

If someone is trying to hack the car, the key fob will not work until you hit buttons a couple of times and get the car and key fob back on the same part of the algorithm. I believe you could also get them out of sync by clicking the key fob out of range of the car.

At least on the Model S a few years ago Tesla used an unusually weak encryption to the algorithm so hackers monitoring the key and car could eventually crack it. Hackers could set up a key listening post in an office car park and monitor communication between keys and car for a few days or weeks to crack the code for cars that park in the lot every day. Once they crack a car, they can steal it with an ersatz key fob.

I believe relay theft can be made more difficult with handshakes between key and car that are timed. Relay equipment introduce a little more delay than the key directly as it passes through the electronics of the amplifier. If the car pings the key and gets an answer in the expected time frame, it assumes it is dealing with the key directly, but if the answer has too much delay, it assumes it is a relay hack and it won't respond.

For cars that require some kind of key press to unlock the doors a simple relay hack isn't going to work. The auto presenting door handles on the Model S can be disabled and I believe the auto unlock feature on the Model X can be disabled too. I don't know about the Model 3. The key card has even shorter range than the fob so a relay hack would be tough to pull off.

A relay hack for Teslas at home in the US would be difficult because a large percentage of US Tesla owners keep their cars in garages. At our house neither of the keys are anywhere near an outside wall a hacker could get their repeater close to either.

The PIN is the way to go.

 

[jaitch]

I believe some Teslas on the continent were stolen from places other than homes like office car parks and it's thought that the algorithm was hacked. Key fobs use a two part code every key press. There is a part that is always transmitted in a fixed pattern with a changing pattern tacked on that cycles through a set of changes that are calculated by an algorithm run on both the key fob and the car.

If someone is trying to hack the car, the key fob will not work until you hit buttons a couple of times and get the car and key fob back on the same part of the algorithm. I believe you could also get them out of sync by clicking the key fob out of range of the car.

At least on the Model S a few years ago Tesla used an unusually weak encryption to the algorithm so hackers monitoring the key and car could eventually crack it. Hackers could set up a key listening post in an office car park and monitor communication between keys and car for a few days or weeks to crack the code for cars that park in the lot every day. Once they crack a car, they can steal it with an ersatz key fob.

I believe relay theft can be made more difficult with handshakes between key and car that are timed. Relay equipment introduce a little more delay than the key directly as it passes through the electronics of the amplifier. If the car pings the key and gets an answer in the expected time frame, it assumes it is dealing with the key directly, but if the answer has too much delay, it assumes it is a relay hack and it won't respond.

For cars that require some kind of key press to unlock the doors a simple relay hack isn't going to work. The auto presenting door handles on the Model S can be disabled and I believe the auto unlock feature on the Model X can be disabled too. I don't know about the Model 3. The key card has even shorter range than the fob so a relay hack would be tough to pull off.

A relay hack for Teslas at home in the US would be difficult because a large percentage of US Tesla owners keep their cars in garages. At our house neither of the keys are anywhere near an outside wall a hacker could get their repeater close to either.

Car theft is a movable feast with manufacturers always seemingly have to play catch up - which I guess is obvious due the nature of the act. A few years back you could jam the signal to the car and capture a code for use at a later date.

The pin to drive was a suggestion from the UK Tesla owners club. Also the moving code entry screen was a further request to Tesla - otherwise you just follows the grubby fingerprint marks to arrive at the code!

 

[aesculus]

Relay equipment introduce a little more delay than the key directly as it passes through the electronics of the amplifier. If the car pings the key and gets an answer in the expected time frame, it assumes it is dealing with the key directly, but if the answer has too much delay, it assumes it is a relay hack and it won't respond.

I thought the same thing but another poster felt that a variable voltage (like battery wearing down or a change in temperature) in the fob would also cause a delay and therefore it would be very hard to discern the two. Out of my league.

 

[Grumble]

I get an alert at least every 2nd or 3rd drive that the key can't be detected. It doesn't matter how new the battery is in the fob. I typically keep it laying in the tray next to the usb ports. Most of the time, when I pick it up, the alert does away. Sometimes, I have to wave it around or push one of the buttons. I used to keep my key in the armrest cup holders but found I would get the alert far more frequently with it in there.

My question is, will the update potentially make this worse?

***Edit***
I did take it in to the service center regarding this when I first got my car. They said it was a known issue and would be addressed in the next firmware update.

 

[dark cloud]

I get an alert at least every 2nd or 3rd drive that the key can't be detected. It doesn't matter how new the battery is in the fob. I typically keep it laying in the tray next to the usb ports. Most of the time, when I pick it up, the alert does away. Sometimes, I have to wave it around or push one of the buttons. I used to keep my key in the armrest cup holders but found I would get the alert far more frequently with it in there.

My question is, will the update potentially make this worse?

***Edit***
I did take it in to the service center regarding this when I first got my car. They said it was a known issue and would be addressed in the next firmware update.

I have had this happen a few times in the past 3 months. Definitely not every 3rd drive, but just a few times. Don’t know which update they are referring to, all I can provide evidence for at present time is it hasn’t happened in the past two days since getting 2019.32.1; meaning the 7 or 8 times I approached the car and drove with the fob in my pocket. Sometimes the fob is in a bag in the back seat.

 

[Grumble]

I have had this happen a few times in the past 3 months. Definitely not every 3rd drive, but just a few times. Don’t know which update they are referring to, all I can provide evidence for at present time is it hasn’t happened in the past two days since getting 2019.32.1; meaning the 7 or 8 times I approached the car and drove with the fob in my pocket.

Thanks for the info. I think it's the SC's canned response to glitches like this. It buys them time for you to just become complacent.

I can't drive with the fob in my pocket. It won't detect the key there more often than it will.

 

[wdolson]

I thought the same thing but another poster felt that a variable voltage (like battery wearing down or a change in temperature) in the fob would also cause a delay and therefore it would be very hard to discern the two. Out of my league.

I suppose it's possible a weak battery or change in temperature could affect the speed of the response, though I doubt it would be significant. With a battery going it would more likely put out a weaker signal rather than get slower.

I get an alert at least every 2nd or 3rd drive that the key can't be detected. It doesn't matter how new the battery is in the fob. I typically keep it laying in the tray next to the usb ports. Most of the time, when I pick it up, the alert does away. Sometimes, I have to wave it around or push one of the buttons. I used to keep my key in the armrest cup holders but found I would get the alert far more frequently with it in there.

My question is, will the update potentially make this worse?

***Edit***
I did take it in to the service center regarding this when I first got my car. They said it was a known issue and would be addressed in the next firmware update.

I had a similar problem with one fob. I put that into spare and use the other fob and it works fine. Last time a ranger was here he tested the bad fob, but couldn't find anything wrong. The problem was intermittent.

 

[jaitch]

Thanks for the info. I think it's the SC's canned response to glitches like this. It buys them time for you to just become complacent.

I can't drive with the fob in my pocket. It won't detect the key there more often than it will.

Does it make a difference not having your phone/cell/mobile on you potentially blocking fob signal? Apparently it is possible to lock your fob in your Tesla if your phone is left next to your fob. If it is same on both fobs then is it an antenna connection issue in the car?

Electronics will try your patience. My voice control only works if I shout really loud at the microphones - I guess there is a dodgy connection/failed component somewhere.

 

[Grumble]

Does it make a difference not having your phone/cell/mobile on you potentially blocking fob signal? Apparently it is possible to lock your fob in your Tesla if your phone is left next to your fob. If it is same on both fobs then is it an antenna connection issue in the car?

Electronics will try your patience. My voice control only works if I shout really loud at the microphones - I guess there is a dodgy connection/failed component somewhere.

I keep my phone on a wireless charging mount on the right side of the main screen. It's never in my pocket.

 

[jaitch]

I keep my phone on a wireless charging mount on the right side of the main screen. It's never in my pocket.

What happens if you turn off the wireless mount? Does that make any difference?

 

[Grumble]

What happens if you turn off the wireless mount? Does that make any difference?

I'll try but this issue started the day I took delivery so unless they've fixed this in the time I introduced the wireless charging mount, I'm going to assume it's still an issue. I'll unplug the mount and run for a few days to see if my experience changes.

 

[wdolson]

Does the key problem happen with both keys or just one? If its both keys the receiver in the car may have a problem.

 

[Grumble]

Does the key problem happen with both keys or just one? If its both keys the receiver in the car may have a problem.

Both keys. I just closed my eyes, grit my teeth and hit the update button. Tried to update my fob but it says v1 fob detected, no update necessary.

 

[RGloverii]

I read somewhere that the ‘Update’ feature was for the v2 remotes.

 

[Grumble]

I read somewhere that the ‘Update’ feature was for the v2 remotes.

Yeah, I thought I had a V2 remote based on the model number shown here

 

[Oldschool496]

I read somewhere that the ‘Update’ feature was for the v2 remotes.

Exactly its for Version 2. Version 1 is not updatable from my understanding of this whole situation.

I'm not purchasing version 2, I used pin from day one and it works great. I suggest everyone with version 1 UK or US or otherwise start using pin. Must remember though, for valet must be in valet to disable pin. I almost forgot last weekend and did not even think the valet would be coming back to ask me for pin.

Don't forget to take it out of valet, done that quite a bit and get on the freeway and the car is limited by app. It takes a few seconds to understand you forgot to take it out of valet. Could get you in sticky mess thinking you can accel hard and guess what, your snail meat.