ASIC botnet: The new threat?

[Korbman]

If not, then it's possible to own a machine with ASIC already attached.

Absolutely. I'm not saying that taking over the device isn't possible, I'm saying that keeping it undetected would be difficult. My guess is that ASIC devices won't be customized to go slower, such as having a Jalapeno go 3GH/s instead of 4.5GH/s just be tweaking it (though someone tell me if they have proof otherwise).
That said, if you've got one Single and suddenly it's not mining for your own account anymore, something tells me you might venture a look at what's going on.

[RodeoX]

I think these would be hard targets for a botnet. Anyone who owns an ASIC miner would likely know a lot about it and know about the risks.

[Jaw3bmasters]

I think these would be hard targets for a botnet. Anyone who owns an ASIC miner would likely know a lot about it and know about the risks.

Ya. Currently.

How bout a 0 day attack?

[RodeoX]

I think these would be hard targets for a botnet. Anyone who owns an ASIC miner would likely know a lot about it and know about the risks.

Ya. Currently.

How bout a 0 day attack?

Hmmm. interesting. When it comes to 1337 h@xoring I have learned to never say never. 

[bcpokey]

I think these would be hard targets for a botnet. Anyone who owns an ASIC miner would likely know a lot about it and know about the risks.

Ya. Currently.

How bout a 0 day attack?

0 day of what? Some flaw in the ASIC miners code? Or a flaw in the host machine? These are not typically the leads to botnets anyway.

Here's the thing, what are botnets used for? They are used for DDoS, and spam-mail traditionally. Why? Because every computer that is on the internet, has access to the internet. Traditionally low-hanging fruit is picked from this bunch, such that easy vulnerabilities that have remained unpatched for long periods of time allow for continuous usage of the targets.

So let's look at this: Every computer on the net has an internet connection, only a fraction are part of botnets.

Every computer has a GPU, only a fraction are "botnettable", only some portion of these GPUs could be used for effective mining, and it's a high-risk venture that would likely lose the machine, and thus is less profitable than other botnet ventures. This is likely why we've not seen any evidence of a GPU mining botnet on bitcoin.

An extremely tiny fraction of computers will have ASICs connected to them, of these some fraction will be botnettable at some point, and most use of this botnet will result in the botnet being broken up in very short time-scale. The difficulty will both be in even locating computers with ASICs, let alone exploiting them, let alone exploiting them effectively.

Is it impossible? I will go ahead and say it is not, just so I don't have egg on my face at some point. But will it happen? I will again say no, it is too difficult, with too little reward.

The only scenario I could really foresee would be some disruption hacktivist, who manages some malicious code that can collect the IPs of the vastly reduced mining world, and utilizing some "zero-day attack" disrupts the block chain, either taking control, forking it, or generally fouling it up. This would likely be as easily or more easily accomplished by finding the IPs and DDoS'ing the heaviest ASIC miners connections though.

Always ask yourself for the motivation behind actions, they are very important.

[Jaw3bmasters]

0 day of what? Some flaw in the ASIC miners code? Or a flaw in the host machine? These are not typically the leads to botnets anyway.

Here's the thing, what are botnets used for? They are used for DDoS, and spam-mail traditionally. Why? Because every computer that is on the internet, has access to the internet. Traditionally low-hanging fruit is picked from this bunch, such that easy vulnerabilities that have remained unpatched for long periods of time allow for continuous usage of the targets.

So let's look at this: Every computer on the net has an internet connection, only a fraction are part of botnets.

Every computer has a GPU, only a fraction are "botnettable", only some portion of these GPUs could be used for effective mining, and it's a high-risk venture that would likely lose the machine, and thus is less profitable than other botnet ventures. This is likely why we've not seen any evidence of a GPU mining botnet on bitcoin.

An extremely tiny fraction of computers will have ASICs connected to them, of these some fraction will be botnettable at some point, and most use of this botnet will result in the botnet being broken up in very short time-scale. The difficulty will both be in even locating computers with ASICs, let alone exploiting them, let alone exploiting them effectively.

Is it impossible? I will go ahead and say it is not, just so I don't have egg on my face at some point. But will it happen? I will again say no, it is too difficult, with too little reward.

The only scenario I could really foresee would be some disruption hacktivist, who manages some malicious code that can collect the IPs of the vastly reduced mining world, and utilizing some "zero-day attack" disrupts the block chain, either taking control, forking it, or generally fouling it up. This would likely be as easily or more easily accomplished by finding the IPs and DDoS'ing the heaviest ASIC miners connections though.

Always ask yourself for the motivation behind actions, they are very important.

So in summary, It's possible, but unlikely, unless.....

.....Ok then, my hypothesis is simple:
One day ASIC will be quiet affordable and user-friendly. (any objections?)

Normal people will continue doing unsafe computer actions. (any objections?)

More knowledgeable individuals will simply own those machines. (any objections?)

Thus we have ASIC botnet. (May not be worth it for YOU)

[2weiX]

if you're the botter and you're smart you're gonna tell the bot to

mine cpu
and if gpu exists, mine gpu
and if asic exists, mine asic

in effect turning it into a cpu/gpu/asic botnet, or short

"botnet"

[serp]

The target is likely too limited and diverse for a botnet attack.  There will be relatively few people out there using ASICS and of those that do they will likely be using many different types of OS/distributions.  Unless you have zero days for everything it's going to be unlikely to get anything on a mass scale.

Additionally, how many people will actually have their ASICS hooked up to a desktop machine?  Personally, I will have mine hooked up to minimal boxes with some minimal version of linux where every service will be turned off except the mining client and sshd on a non-standard port so I can access it locally on my network.

Luckily, ASICS will boost the difficulty up high enough so that having a botnet that mines off cpu/gpu as they do now will possibly pay less than the effort is worth.

[live627]

Was there ever FPGA botnet?

[bitcoindaddy]

Anybody that spent that much money on their ASICs is going to be watching the returns (the money flowing in). If they notice an anomaly, they will know something is wrong.

[Jaw3bmasters]

Anybody that spent that much money on their ASICs is going to be........

ASIC will, in the-not-so-distant future, cost about the same as GPU.

The cost is high now coz there's a HIGH demand for ASIC. Once that little bubble burst, I'd imagine ASIC price will fall.

I'll personally order a few and plug them in and forget em, I got other things to do. So does my rich friend who's even less tech savvy than I am.

With that, there's a chance they'll get own, becoming part of a net.

[MrTeal]

Anybody that spent that much money on their ASICs is going to be........

ASIC will, in the-not-so-distant future, cost about the same as GPU.

The cost is high now coz there's a HIGH demand for ASIC. Once that little bubble burst, I'd imagine ASIC price will fall.

I'll personally order a few and plug them in and forget em, I got other things to do. So does my rich friend who's even less tech savvy than I am.

With that, there's a chance they'll get own, becoming part of a net.

There's no incentive for the vast majority of people to buy ASIC mining hardware regardless of price. Everyone needs a CPU and some sort of graphics hardware in a computer. Even if Bitcoin really takes off the cast majority of people won't buy mining hardware; there's no reason to. They could just buy BTC on the market.

[bcpokey]

Anybody that spent that much money on their ASICs is going to be........

ASIC will, in the-not-so-distant future, cost about the same as GPU.

The cost is high now coz there's a HIGH demand for ASIC. Once that little bubble burst, I'd imagine ASIC price will fall.

I'll personally order a few and plug them in and forget em, I got other things to do. So does my rich friend who's even less tech savvy than I am.

With that, there's a chance they'll get own, becoming part of a net.

I don't mean to sound insensitive but it sounds like you're approaching a very technical problem from a very non-technical point of view. So let's try a different approach.

To date there have been 0 confirmed CPU botnets, 0 confirmed GPU botnets, and 0 confirmed FPGA botnets on bitcoin. Why would this change with ASICs? If as you say, ASICs become cheap like water, difficulty becomes high like a mountain. A 60GH/sec ASIC becomes the equivalent of a 300MH/sec GPU today, and there is as little incentive to seek out ASICs as there is for the other methods.

How many people are rich? By definition, very few (limited resource world). Thus very few people would "set it and forget it" after buying hardware that has no function except mining. Most people with money would as MrTeal said rather buy Bitcoins directly. It's faster, easier, can just throw it somewhere and forget about it. This idea of an ASIC botnet is really just a bugaboo.

[Jaw3bmasters]

There's no incentive for the vast majority of people to buy ASIC mining hardware regardless of price.

The incentive is Bitcoin. Or cryto-currency.

[Jaw3bmasters]

To date there have been 0 confirmed CPU botnets, 0 confirmed GPU botnets, and 0 confirmed FPGA botnets on bitcoin.

This made LOL. No, seriously....it did.

This is null.

The only thing left is don't get owned.

[live627]

ASIC will, in the-not-so-distant future, cost about the same as GPU.

The cost is high now coz there's a HIGH demand for ASIC. Once that little bubble burst, I'd imagine ASIC price will fall.

Did the FPGA get cheaper? BTCFPGA MM Quad is still $1k.

[bcpokey]

To date there have been 0 confirmed CPU botnets, 0 confirmed GPU botnets, and 0 confirmed FPGA botnets on bitcoin.

This made LOL. No, seriously....it did.

This is null.

The only thing left is don't get owned.

You have some contrary evidence? Or you just have no logical response?

[Rudd-O]

I can't confirm if it's hackable or not, just not that many people bother. With all the su/sudo things that need to be run...I just don't see it happening.

too complex for me, me to stupid can't understand -> UNBREAKABLE!!!

your logic is failing.

You are picking a fight with the wrong person, and I shouldn't be the one to talk about comprehension when you failed to comprehend first grade English grammar.

Which brings me to my question, are you kano? He is an idiot and has the same avatar. Spells the same way. I can only imagine it's you.

kokjo is an idiot who picks fights with everyone, constantly pretends to misunderstand what is being said, and initiates verbal abuse toward people.

[SolarSilver]

To date there have been 0 confirmed CPU botnets, 0 confirmed GPU botnets, and 0 confirmed FPGA botnets on bitcoin.

Actually there are plenty of GPU botnets and big pools like Deepbit have to take frequent action against them. Deepbit blocks single accounts that have more than 150 (?) different IPs connecting.

To see if they are CPU or GPU botnets, one should ask if Tycho would release the stats on such an account so one can judge how much MH/s per IP they generate.

A smart GPU botnet would of course create a different account for each worker but that is sooooo much work, right? ;-)

[bcpokey]

To date there have been 0 confirmed CPU botnets, 0 confirmed GPU botnets, and 0 confirmed FPGA botnets on bitcoin.

Actually there are plenty of GPU botnets and big pools like Deepbit have to take frequent action against them. Deepbit blocks single accounts that have more than 150 (?) different IPs connecting.

To see if they are CPU or GPU botnets, one should ask if Tycho would release the stats on such an account so one can judge how much MH/s per IP they generate.

A smart GPU botnet would of course create a different account for each worker but that is sooooo much work, right? ;-)

Yes? Can you link something to evidence the frequent action pool operators take? Article or transcript from pool operators talking about it? I'm curious to learn about this.