TERRACOIN ATTACK OVER 1.2TH ATTACK CONFIRMD

[crazyearner]

Someone got an ASIC working on TRC?

You can use Asic on TRC and others that are sha256D

[ohiwastedmylif]

Someone got an ASIC working on TRC?

You can use Asic on TRC and others that are sha256D

Application Specific, I would assume TRC is slightly different in the code and would require tweaks would it not? If not then why has an attack like this not happened yet?

[crazyearner]

Someone got an ASIC working on TRC?

You can use Asic on TRC and others that are sha256D

Application Specific, I would assume TRC is slightly different in the code and would require tweaks would it not? If not then why has an attack like this not happened yet?

It already has happened a few months back when the first asics where coming about it got attacked by 1 Avalon unit and it messed the difficulty up so they changed how it works and seems someone else has no found a way to attack and gain in the event of it.

[smscotten]

Application Specific, I would assume TRC is slightly different in the code and would require tweaks would it not? If not then why has an attack like this not happened yet?

Naw, I don't have ASICs but I have mined TRC with FPGAs. Same principle. The "application" isn't the miner, it's just the hashing algorithm. You can use the same setup to mine TRC that you would to mine BTC or PPC.

[markm]

Blockchains are insanely expensive to secure. Someone didn't care and wanted to fleece the suckers anyway so went ahead and made one anyway. The rest is pretty much just yet another repeat of why suckers are called suckers.

-MarkM-

[smoothie]

This is what happens when you keep the hashing algorithm the same when another chain that starts with a "B" and ends with an "N" has ASICs coming out for it.

This was not hard to see coming 8 months ago.

[Sunny King]

This has nothing to do with hash algorithm. FTC using scrypt didn't prevent someone showing up with 51% attack.

It's always been a problem in the implementation of terracoin's difficulty adjustment. Difficulty adjustment shouldn't be too fast.

[smoothie]

This has nothing to do with hash algorithm. FTC using scrypt didn't prevent someone showing up with 51% attack.

It's always been a problem in the implementation of terracoin's difficulty adjustment. Difficulty adjustment shouldn't be too fast.

How many scrypt ASICs are you aware of?

We are talking apples and oranges here. Gpus vs ASICs are a factor of 10 in speeds. Two totally different scenarios.

When I first mentioned this issue was before they changed the difficulty adjustment, back then the same issues still applies.

Difficulty adjustment is a moot point when it comes to 51% attack. If it was an issue then a new sha256 chain with adjustment every week would still have the same issues of a 51% attack because 1 sha256 ASIC would be enough to 51% attack it.

To be clear this is assuming fact someone is 51% attacking TRC, which I'm not sure is true.

[helipotte]

Are the Devs gonna fix this?  Kinda worried for Terracoin.  No activity on Git for over three months now.

[Trillium]

Dem 10 second blocks: 

Code:

[2013-07-25 11:58:00] LONGPOLL from pool 4 detected new block
[2013-07-25 11:58:04] LONGPOLL from pool 4 detected new block
[2013-07-25 11:58:07] LONGPOLL from pool 4 detected new block
[2013-07-25 11:58:13] LONGPOLL from pool 4 detected new block
[2013-07-25 11:58:26] Stratum from pool 1 detected new block
[2013-07-25 11:58:27] Stratum from pool 3 detected new block
[2013-07-25 11:58:30] LONGPOLL from pool 4 detected new block
[2013-07-25 11:58:38] Stratum from pool 1 detected new block
[2013-07-25 11:58:49] Stratum from pool 1 detected new block
[2013-07-25 11:58:58] Stratum from pool 2 detected new block
[2013-07-25 11:59:13] LONGPOLL from pool 4 detected new block
[2013-07-25 11:59:36] Stratum from pool 1 detected new block

[smscotten]

How many scrypt ASICs are you aware of?

We are talking apples and oranges here. Gpus vs ASICs are a factor of 10 in speeds. Two totally different scenarios.

Yeah, and Scrypt vs SHA-256  is a factor of 1000 in speeds. For a lot less than what it costs to put together a 1.2TH attack on Terracoin you could put together a 1.2GH attack on, say, Feathercoin with GPUs. It would cost a lot more in electricity, but an attacker wouldn't care about keeping the farm going on an ongoing basis.

Correct me if I'm making faulty assumptions or if my seat-of-the-pants math has led me astray.

However, the part about messing with the difficulty level is actually relevant. The attacker could have made it easier to attack by making the difficulty so unpredictable that miners gave up on it. I know I gave up on it for that reason. Difficulty would be enticingly low so I'd mine for about a half hour when I'd notice the difficulty had gone through the roof. After a while I learned not to pay any attention if TRC had a low difficulty. It wouldn't be worth my time to switch. Enough people give up and you keep the fluctuations going, eventually you can catch the network with its total hashrate down.

I'd be surprised if there weren't already a name for that tactic.

[roy7]

The last time TRC's difficulty adjustment was made was because the network hash rate was pretty low and someone would mine with an ASIC getting tons of blocks, difficulty skyrocket, and they move to a new coin until it was back down. Other coins have had that problem too (like FRC). That's not an attack per se, just someone optimizing their mining efforts but disrupting the whole coin because they have such a large % of the network hash rate.

I'm not sure that is happening now or not though, or if it some other sort of exploit. When the difficulty drops to the minimum 5k for quite a while, even if the 'attacker' wasn't mining, it should never go that low. There was a DDOS attack on TRC pools about a week ago but that seems over with at the moment.

Coinotron has 363.4 GH mining TRC right now. Even someone with 2-3 60GH/s miners should not be able to disrupt the difficulty that badly, or conduct a 51% attack. Maybe someone has way more than that? (A BFL minirig?)

The only solution, which is pretty much always the solution, is for more people to mine on the established pools so the "attackers" make up a smaller % of the network hashrate.

[roy7]

Maybe it's a repeat of this?
https://bitcointalk.org/index.php?topic=167493.0

Sure acts like it... I wonder if they found a way to do the same exploit even after the patch that was done in April.

[smoothie]

The last time TRC's difficulty adjustment was made was because the network hash rate was pretty low and someone would mine with an ASIC getting tons of blocks, difficulty skyrocket, and they move to a new coin until it was back down. Other coins have had that problem too (like FRC). That's not an attack per se, just someone optimizing their mining efforts but disrupting the whole coin because they have such a large % of the network hash rate.

I'm not sure that is happening now or not though, or if it some other sort of exploit. When the difficulty drops to the minimum 5k for quite a while, even if the 'attacker' wasn't mining, it should never go that low. There was a DDOS attack on TRC pools about a week ago but that seems over with at the moment.

Coinotron has 363.4 GH mining TRC right now. Even someone with 2-3 60GH/s miners should not be able to disrupt the difficulty that badly, or conduct a 51% attack. Maybe someone has way more than that? (A BFL minirig?)

The only solution, which is pretty much always the solution, is for more people to mine on the established pools so the "attackers" make up a smaller % of the network hashrate.

So in an attempt to make diff adjust more often due to ASIC attacks they still have another issue.

I still say a big part of the problem is they chose sha256 as the hash algorithm which there is now an army of ASICs to rape it at no cost to the attacker...well not much.

[flound1129]

Are the Devs gonna fix this?  Kinda worried for Terracoin.  No activity on Git for over three months now.

Devs?

[roy7]

As far as I can tell, the devs aren't active any more. Hoping I'm wrong. But someone posted a serious big report a month or two ago about the 32 bit version of the client being able to download/sync the block chain, and it was never replied to or fixed.

[STT]

The main factor seems to its relation to Bitcoin and its standing armies of asic.  So maybe it has to become more similar in its defense by raising difficulty in time

[NWO]

Someone got an ASIC working on TRC?

They are already available. TRC is SHA-256

[DeathAndTaxes]

Someone got an ASIC working on TRC?

You can use Asic on TRC and others that are sha256D

Off topic but what is up with the "D".  For the first couple years nobody got the hashing algorithm wrong and then recently there has been this poliferation of "SHA-256D" and "SHAW-256".  Once could be a typo, one person multiple times could been someone who is repeating a mistake but multiple people over multiple threads.

It is SHA-2 or SHA-256 (or without the hyphen SHA2, SHA256).

[coinerd]

t what is up with the "D".  For the first couple years nobody got the hashing algorithm wrong and then recently there has been this poliferation of "SHA-256D" and "SHAW-256".  Once could be a typo, one person multiple times could been someone who is repeating a mistake but multiple people over multiple threads.

It is SHA-2 or SHA-256 (or without the hyphen SHA2, SHA256).

I'm not sure whether it started with miner software using sha256d to describe the double sha-256 hash or if it came from outside of bitcoin circles. A quick google is inconclusive. I know the term has become pretty common and since it's the switch for algo selection in several miners it's probably here to stay.

Sha-2 and sha-256 are the related like squares and rectangles.  All sha-256 is sha-2 but not all sha-2 is sha-256.

I haven't seen shaw-256 but it reads like a phonetic typo. I'm sure I'll notice it now.