The current dust storm (200-300 mempool level) started early in Feb, and still continues. Sure, there are pauses from time to time, but the patter is the same - mempool is at around 200-300 level, and when it stops, it goes back to a normal noise around 20 mp level (what was also pre-Feb).
The previous dust storms were basically brief (a couple of days or so) but very intensive (in thousands mp levels). The creator of the first one had a clear goal, and he talked about it, the other two/three were potentially highschoolers trying to use his code and their momâs computers.
In case even a small dust storm is present, it forces some folks to add some fees to their transactions. Isnât this the main reason for this dust storm? It is not really causing nodes to be knocked out, just slows down transaction processing / forces fees on those that need to get things done.
The only reason that I can think of for this storm is that it is conducted to really extort those fees. If that is the case, the only party that benefits are whales. Sure, all farmers get a share of those fees, but I get a coin here and there, and am nowhere close to be getting multiple per day or hour.
Not sure if anything can be done about it, but I would assume that it may be fairly easy for Chia to nail down the IP of that machine. Although, not sure what can be done (block it on the introducer side may not work as it looks like there are a third party introducers offered). Maybe if that entity is using the same fee to push his/her traffic, have transactions with that particular fee(s) being demoted to the end of the mempool queue (make it more expensive)?
Another thing is whether something can be done to have nodes process those transactions faster, so 300 level would be just too low to have any effects, and going higher would make the efforts more expensive.
There is no difference whether you pool or do it solo. Although, if one wants to change pool membership, somehow it is also affected (or maybe that was some bad code in one of the previous v1.3 releases).
The main distinction between this dust storm and those previous (at least for me) is that the previous ones were causing enough transaction traffic that nodes were stuck on passing data back and forth, and were basically choking (partial network contraction). This one potentially chokes the least optimized nodes (looks like not that many of those left), but extorts fees from those that need to send XCH right away. I mean, for all us farmers that is some extra mojos, but in case of smaller farms (say below few PB), those are just breadcrumbs.
I wasnât aware of that. Interesting. But I can think of at least one more reason other than extorting fees: trolling. Could be totally wrong on it, but it could be mischievous activity for no other reason than because its super easy to do. I think these dust storms have hugely increased the size of the blockchain DB which might have a lasting effect, and certainly is making a full sync hell right now.
If thatâs the case, itâll probably keep up until the value goes up. Until then, they can keep throwing pennies back and forth ad infinitum.
Iâd kinda been wondering for a while what benefit there might be and hadnât really considered the fee angle. Just was more thinking someone was saying âwhy notâ and are doing it because it is easy and isnât costing them anything to run this sustained activity.
trolling - Yup, but that would require two things, a consistent troll, and some free resources (a box or two) to pump those transactions. So, I would think that if that would be the angle, a âstandardâ troll would already move on to something else, as the original mission was already accomplished. Another reason that I would dismiss it is that if someone has it already fine-tuned this process, why not to ask his/her buddy to join and cause even more harm. As this "trollâ keeps it running for almost 3 months now, it has to be assumed that he/she doesnât feel that there could be any retaliation for doing that, so partnering with a buddy would be really the next step.
On the other hand, the extra db growth is a slow but real damage. I think that right now v2 db grows around 15-20 GB per month. Therefore, if this will continue, soon v2 will be in the same position as v1. So, this angle could be considered as a real attack against chia (both us farmers as well as the company). One could potentially engage in such activity if looking to buy chia or considers Chia Net as a competitor (thus devaluate the company value, discouraging potential business partners, âŚ).
So, seeing it going on for such a long time, my take would be that there is some financial goal behind that (as all the standard checkmarks were already hit long time ago).
Yeah, I see what you mean on the fees. I hadnât looked for a while but the block rewards are pretty high right now, so financial incentive makes good sense.
I think you could keep the dust storm running nicely on 1-2 VPS nodes costing less than $10 USD per month, so I wouldnât say resources were huge, but it would probably require a lot of babysitting.
As whatever the motives are it is either just a nuisance or potential financial exploits, the obvious question is what can be done, or if anything can be done to make it more expensive to run.
Of course, transaction processing is more or less processor / drive (with db) bound, not really network. So, this points to full node in need of a better network handling (backing off, if network is congested, better filtering what needs to be resent), and block crunching process (more transactions per second processed). Somehow, this overlaps 100% with all the slow syncing, and all is in chia dev court for a very long time.
The first dust stormer said âbecause I canâ. Ok, cool, you get some bragging rights. Perhaps they or the wannabe dust stormers repeated this again for âfunâ.
But to make it continuously, there is either an incentive of some sorts or it is virtually free to make transactions on the chain, so the stormer(s) run on a principle âbecause it is dirt cheapâ
I donât get why this has to be some malicious person doing this (guess is was the 1st time). What about devs working a project and want to test of network resiliency? They do play transactions equal to what might occur if the project was in actual use> aka a dust storm. Maybe even multiple actors doing this. How does this not fit whatâs occurring? If it was some nut-job, would they continue for so long, âjust becauseâ? Seems there is more intent behind what happens.
It is a possibility, but I think also less likely. They can do all kind of test ton testnets, as those are smaller, thus easier to influence, plus in case things need to be pushed a bit more, no one suffers (e.g., db growth, problems with switching pools, forcing people to pay fees). Also, you really donât need to stress test while you are coding, just when you want to test your code, and this thing is going on for almost 3 months already. Lastly, most likely they would have already stated that. So, still possible, although maybe not the actual reason.
The reason that I think it is just one person is that from the very beginning, the patter is really same. When it is on, it hovers around 200-300 level, then it drops to normal (20), and comes back at that 200-300. If the second person would join, I would assume that for some of those runs we would see higher numbers, also we would not see drops to normal (20), as they would need to coordinate stopping both sides, and starting them. So, we would be seeing three levels from time to time. However, it is steady 200-300 from the very beginning. As those last year storms reached 2k (maxed out mempool), I would think that there ie plenty of room to scale it higher (e.g., by getting multiple actors).
What brings another possibility that maybe this is some unintentional traffic where the person behind it is just not aware that it is going on. For instance, such person did some initial tests, and just didnât really turn the thing off for good, so it is somehow auto-starting, and unless such person will be notified to really stop it, it will be going on.
Bitcoin had a clever method of dealing with this problem: 25% of each block was reserved for transactions based on coin age, not fee.
That provided long term holders with fee-free transactions, and made it more difficult for any one person to completely fill a block.
Allowing anyone who can outspend the rest of the network to block all transactions makes certain attacks much easier. Having two different ways to âpayâ for inclusion (fees or long term holding) made the network much more resilient.
Unfortunately Bitcoin Core dropped support for Coin Age Priority in 2017.
Dust has been a problem for most blockchains in the early days and various countermeasures have been developed including minimum spends and minimum fees. For example Litecoinâs minimum fee per output.
As coins become more valuable the number of people willing to waste them on an attack decreases until it is not an everyday problem, but a network that only relies on fees for priority still remains potentially vulnerable to adversaries willing to spend large amounts to attack it.
The dust storm could simply be caused by a person who invested in the project and saw their investment drop off a cliff. Now they are pissed and want to get back at project by causing a ruckus. I read the article about the first dust stormer and if it only cost a few bucks a month to cause this problem then this could be a huge issue for chia. If an asshole can cause a problem for thousands of people and only spends $10 a month doing it and then I believe they will do it with a smile on their face. I donât have a solution for the issue, but just my 2 cents.
Plausible! We see daily the results of âone person attacksâ in our world. That is, one single person can wreck untold damage and itâs extremely hard to stop it from happening. The only difference between these individuals is what they choose to release their damage on.
I agree with you 100%. However, considering how little is needed to make it happen, I would say that whoever sees a profit in doing that would for sure outlast a pissed off guy.
On the other hand, I didnât know till recently, that the biggest fraud (combined) in US is actually medical fraud (white collar fraud). Another good example is Sacklers family that made over $20b also by killing people and destroying families.
So, yeah those guys in their momâs basement are doing some damage, and it grows at the same time not being prosecuted as it should be. However, it looks like all that is used to shift the focus from where the real crime.
I would not say that. I monitor a couple of web servers and my home router, and what I see is that excluding Russia, China, most of attacks are from just a handful of VPS providers. There is nothing being done to held those providers accountable. On the other hand, it is really easy to have them play a whack-a-mole game to show how active they are, where they collect another $10 when such accounts just rent another VPS.
At some point, I was trying to contact âabuseâ emails at those providers (that they are required to have it), but it ended up with them passing my info to the attackers.
On the bright side, it is showing the Chia blockchain to be resilient to the abuse, and handling the increased transaction volume fairly well?
I canât really explain the super high fees that have been occurring. Is that showing that people are using Chia, and are paying higher transaction fees because they want their transactions to complete quickly?
Itâd probably be unethical for the Chia Network, Inc. to be doing it as a sort of stress or litmus test. There are probably some things that just canât be simulated on the testnet, even with a really good network simulator like Shadow Simulator. There might be any number of reasons to run a proof like this, but there are ethics concerns and it doesnât seem to fit the profile anyway.
Regarding the recent price drop, that could be purely in relation to recent market events. It stands to reason that if Bitcoin drops some percentage, other coins, especially âless popularâ ones will drop along with it, and possibly at higher percentages because thereâs already way less demand for them.
OR, if Jacekâs original idea is on point, the whale(s) creating the storm could be collecting huge transaction fees and recently offloaded more XCH in the limited trading arenas that there are, thus dropping the price a bit more than usual.
But, all speculation of course Maybe one day weâll find out.
i think you are looking wrong perspective.
attacker do that because he can.
i blame developers about that.
think about that you just try join a pool but you cant because someone doing strom attack.
when developers doing their job better this problems will fix. question is, do you really belive developers will do their job better?
I also agree that the developers need to do a better job of protecting the chia network. If it truly is just one person or a few people spending $10 to $20 worth of chia every month to cause the storm then its just plain ridiculous to me. If the dust storm is caused by a whale making more money then what they are spending causing the issue then thatâs just another weak point of chia. At the end of the day chia holders need answers and fair transaction fees if the coin is to be feasibly used and become mainstream.
It was mentioned earlier about the size of the database increasing rapidly because of the dust storm. Could this be the intention of the dust storm, with the purpose that it might gradually cause farmers to abandon Chia as they run out of disk space, and on balance they decide itâs no longer worth buying a larger drive to store the db given the low price of XCH, and they leave the game? So the Netspace slowly falls, leaving more XCH to be farmed by those that remain.
Maybe one day weâll find out.