GingerAle
Legendary
Offline
Activity: 1260
Merit: 1008
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
June 30, 2015, 08:45:22 PM |
|
I did a transaction with a my usual transaction fee settings (slightly larger than standard fee) half an hour ago. The transaction was sent by my Electrum client via my Electrum server. Blockchain Info showed that the transaction confirmed in 6 minutes (next block).
However, at this network load my Electrum server was lagging and I didn't see the confirmation for some time. The server runs on a slow processor and the Electrum Server is using about 7 times as much CPU time as the bitcoin core node on the same machine. Another problem was that the Electrum client (on a different machine) still hadn't picked up the confirmation for several minutes after the Electrum server was done processing the block containing the confirmation. I restarted the client and this time it worked.
Performance (so far) of Bitcoin Core has been exemplary. Performance of Electrum, less so.
u should let the electrum team know about your experience, if they aren't hacking through it themselves. looks like they might be: https://github.com/spesmilo/electrum-serverupdate 7 hrs ago with some kind of mempool update
|
|
|
|
Quickseller
Copper Member
Legendary
Offline
Activity: 2968
Merit: 2370
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
June 30, 2015, 08:48:58 PM |
|
The attack is not meant to harm the network. This test is designed to measure the effect of what will happen when transaction volume starts to pick up naturally.
Yes it is. If the malicious spammer wanted to test something, we have a testnet for that. This is exactly what the testnet is ment for. It was explained in the thread regarding last week's test why the test net was not used - doing so would not properly measure economic incentives of both miners and other users in regards to TX fees. Eligius filtering out the test's transactions is doing nothing but harm to Bitcoin and is a good reason why no one should mine on eligius. Filtering out transactions that belong to a certain person harms bitcoin's fungibility and it is things like their actions that will lead to Bitcoin's failure.
Do you read and store all your email spam as well, forwarding copies to all your friends? You are comparing apples to oranges. Email I receive to my private email, is by definition private and will not (with very few exceptions) be shared with anyone. A better analogy would be would I forward emails sent to my mail server intended for my customers when the sender is paying for them to be forwarded? The answer would obviously be yes. Eligius is treating certain inputs differently then other inputs. Plain and simple. They are saying that Bitcoin owned by one entity is not allowed to have their transactions confirmed by their found blocks. This is not the first time that eligius (via Luke-jr) has done something similar to harm Bitcoin. Gambling sites were previously blacklisted by default by mining software when this was pretty clearly not a wanted "feature" of such software. In other words the software did things that its users did not want. This is how a lot of people describe malware.
|
|
|
|
sturle
Legendary
Offline
Activity: 1437
Merit: 1002
https://bitmynt.no
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
June 30, 2015, 08:58:11 PM |
|
Well I don't see how it could determine who the attacker is without some reference point, such as address, ip, node, or etc.
You can't spot the spam transactions yourself? Get … eyes. No, you don't need to know who it is to block the transactions.
|
Sjå https://bitmynt.no for veksling av bitcoin mot norske kroner. Trygt, billig, raskt og enkelt sidan 2010. I buy with EUR and other currencies at a fair market price when you want to sell. See http://bitmynt.no/eurprice.plWarning: "Bitcoin" XT, Classic, Unlimited and the likes are scams. Don't use them, and don't listen to their shills.
|
|
|
ChetnotAtkins
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
June 30, 2015, 09:00:45 PM |
|
So... was that it? Back to 2 tx/s.
Now what I really want to know is why on earth did they not send transactions at a pace of 32/s as planned.
|
|
|
|
AgentofCoin
Legendary
Offline
Activity: 1092
Merit: 1001
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
June 30, 2015, 09:06:16 PM |
|
Well I don't see how it could determine who the attacker is without some reference point, such as address, ip, node, or etc.
You can't spot the spam transactions yourself? Get … eyes. No, you don't need to know who it is to block the transactions. So you are automatically adding and then manually subtracting those "spam" txs before publishing the block, with your eyes?
|
I support a decentralized & unregulatable ledger first, with safe scaling over time. Request a signed message if you are associating with anyone claiming to be me.
|
|
|
|
coolbeans94
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
June 30, 2015, 09:12:03 PM |
|
How much longer is this supposed to go on?
|
(1.) Moral happiness depends upon moral order. (2.) Moral order depends upon the harmonious action of all our powers, as individuals and as members of society.
|
|
|
sturle
Legendary
Offline
Activity: 1437
Merit: 1002
https://bitmynt.no
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
June 30, 2015, 09:14:21 PM |
|
It was explained in the thread regarding last week's test why the test net was not used - doing so would not properly measure economic incentives of both miners and other users in regards to TX fees.
BS. If the attacker had a tenth of a clue, she could easily simulate all transactions over a period, and add his spam to that. Do you read and store all your email spam as well, forwarding copies to all your friends?
You are comparing apples to oranges. Email I receive to my private email, is by definition private and will not (with very few exceptions) be shared with anyone. But you do read and store it? No filtering of any kind? Treat all mail alike? I am sure the spammers won't mind if you share it, btw. A better analogy would be would I forward emails sent to my mail server intended for my customers when the sender is paying for them to be forwarded? The answer would obviously be yes.
So you take money for spammers to spam your customers. Great. I am sure you get a lot of them. Do your customers know? Eligius is treating certain inputs differently then other inputs. Plain and simple. They are saying that Bitcoin owned by one entity is not allowed to have their transactions confirmed by their found blocks.
Nope. They didn't say that. Even when you download and install a fresh Bitcoin Core form bitcoin.org or compile one from git, there are anti-DoS measures preventing most spammy transactions from getting forwarded and mined, and it prioritize non-spammy transactions over spam. This malicious attacker has tweaked his spam to get around the default anti DoS measures. Nobody can prevent miners to tweak the anti spam code in their server to ignore the new spam transactions or change the priorities. In fact, they don't need to mine any transactions at all, or they can choose to mine all transactions. Most miners are bitcoin friendly however. They keep keep the anti-DoS measures, include all sensible transactions waiting to get mined in their blocks and only mine spam if it doesn't get in the way. All miners prioritize sane transactions when blocks are full. This is not the first time that eligius (via Luke-jr) has done something similar to harm Bitcoin. Gambling sites were previously blacklisted by default by mining software when this was pretty clearly not a wanted "feature" of such software. In other words the software did things that its users did not want. This is how a lot of people describe malware.
I am a user. I want this. I want bitcoin to be successfull, not spammed to death.
|
Sjå https://bitmynt.no for veksling av bitcoin mot norske kroner. Trygt, billig, raskt og enkelt sidan 2010. I buy with EUR and other currencies at a fair market price when you want to sell. See http://bitmynt.no/eurprice.plWarning: "Bitcoin" XT, Classic, Unlimited and the likes are scams. Don't use them, and don't listen to their shills.
|
|
|
ChetnotAtkins
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
June 30, 2015, 09:18:33 PM |
|
Mempool at 16.5 Mb. Not bad but didn't they aim for 120 Mb?
|
|
|
|
sturle
Legendary
Offline
Activity: 1437
Merit: 1002
https://bitmynt.no
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
June 30, 2015, 09:18:54 PM |
|
Well I don't see how it could determine who the attacker is without some reference point, such as address, ip, node, or etc.
You can't spot the spam transactions yourself? Get … eyes. No, you don't need to know who it is to block the transactions. So you are automatically adding and then manually subtracting those "spam" txs before publishing the block, with your eyes? If you need more than a few minutes to code a filter after looking at three of the spam transactions with your eyes, you should find something else to do with your life.
|
Sjå https://bitmynt.no for veksling av bitcoin mot norske kroner. Trygt, billig, raskt og enkelt sidan 2010. I buy with EUR and other currencies at a fair market price when you want to sell. See http://bitmynt.no/eurprice.plWarning: "Bitcoin" XT, Classic, Unlimited and the likes are scams. Don't use them, and don't listen to their shills.
|
|
|
tspacepilot
Legendary
Offline
Activity: 1456
Merit: 1081
I may write code in exchange for bitcoins.
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
June 30, 2015, 09:25:38 PM |
|
I just registered to tell you that my bitstamp withdrawal has been stuck for 8 hours because of these "stress tests" I have now missed making my mortgage payment because of this. Thanks... I'm already completely broke and struggling to feed my family. Now I have more bank charges and lost value from today's peak.
Way to hit the small guys.
My last 4 BTC.
This is the biggest consequence to my mind of these stress-tests. Average guys who didn't hear about it beforehand having to wait days for transactions to confirm. I know that the dudes who are "stress-testing" say they're proving some important point, but the collateral damage, in my opinion, is too great just to confirm conclusions that we already knew. Thankfully Eligius pool is working against the DOS attack and that fact is encouraging, that at least some miners are willing to prioritize transactions based on something other then pure fee---in this case, the Eligius miners are standing up for the small guy and regular users who are being disrupted by this "experiment".
|
|
|
|
AgentofCoin
Legendary
Offline
Activity: 1092
Merit: 1001
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
June 30, 2015, 09:25:51 PM |
|
Well I don't see how it could determine who the attacker is without some reference point, such as address, ip, node, or etc.
You can't spot the spam transactions yourself? Get … eyes. No, you don't need to know who it is to block the transactions. So you are automatically adding and then manually subtracting those "spam" txs before publishing the block, with your eyes? If you need more than a few minutes to code a filter after looking at three of the spam transactions with your eyes, you should find something else to do with your life. Here is my original quote, that you then took part of, and responded to. Well I don't see how it could determine who the attacker is without some reference point, such as address, ip, node, or etc.
If you are just rejecting all transactions below a certain miner fee and not including in a block, then I have no problem, in theory. I'm only against outright blacklisting/whitelisting of addresses/coins/etc that will ultimately lead to Bitcoin/bitcoin's destruction.
Maybe you can answer my question and stop tap dancing around. Is your "filter" filtering based on transaction sizes and not miner fees? And if so, are all transactions at or around that chosen number being excluded, including non-attacker txs? Because my understanding was that the attacker's txs are being excluded, not all below a chosen limit.
|
I support a decentralized & unregulatable ledger first, with safe scaling over time. Request a signed message if you are associating with anyone claiming to be me.
|
|
|
sturle
Legendary
Offline
Activity: 1437
Merit: 1002
https://bitmynt.no
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
June 30, 2015, 09:32:21 PM |
|
Maybe you can answer my question and stop tap dancing around. Is your "filter" filtering based on transaction sizes and not miner fees? And if so, are all transactions at or around that chosen number being excluded, including non-attacker txs?
I don't have a filter, but I can easily see the malicious DoS transactions are all of two different forms, making them very easy to filter. No need to check transactions sizes or fees. Don't expect Eligius to reveal the exact filter, since this will enable the attacker to circumvent it.
|
Sjå https://bitmynt.no for veksling av bitcoin mot norske kroner. Trygt, billig, raskt og enkelt sidan 2010. I buy with EUR and other currencies at a fair market price when you want to sell. See http://bitmynt.no/eurprice.plWarning: "Bitcoin" XT, Classic, Unlimited and the likes are scams. Don't use them, and don't listen to their shills.
|
|
|
ChetnotAtkins
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
June 30, 2015, 09:37:18 PM |
|
The mempool is at 19.5Mb! Has it ever been that high?
|
|
|
|
AgentofCoin
Legendary
Offline
Activity: 1092
Merit: 1001
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
June 30, 2015, 09:40:09 PM |
|
Maybe you can answer my question and stop tap dancing around. Is your "filter" filtering based on transaction sizes and not miner fees? And if so, are all transactions at or around that chosen number being excluded, including non-attacker txs?
I don't have a filter, but I can easily see the malicious DoS transactions are all of two different forms, making them very easy to filter. No need to check transactions sizes or fees. Don't expect Eligius to reveal the exact filter, since this will enable the attacker to circumvent it. So what you are saying is you don't actually know how they are filtering, but have decided to comment that it is correct, fine, and easy to do. All that I was saying is that they should not filter coins, addresses, servers, nodes,ip and etc, IMO. BTW, the attacker, if wanting to take down the network, would keep increasing transaction amount and fees to circumvent the "filtering". The intention is not to harm, in theory. This isn't an attacker, but most likely a tester or spammer. IMO.
|
I support a decentralized & unregulatable ledger first, with safe scaling over time. Request a signed message if you are associating with anyone claiming to be me.
|
|
|
ChetnotAtkins
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
June 30, 2015, 09:45:00 PM |
|
Mempool at 20.5Mb. Is this the highest it has ever been?
This test is getting exciting again!
|
|
|
|
eleuthria
Legendary
Offline
Activity: 1750
Merit: 1007
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
June 30, 2015, 09:47:28 PM Last edit: June 30, 2015, 10:00:46 PM by eleuthria |
|
(snip) (TLDR) Long amount of bullshit claiming to be "testing" things which anybody with a functional brain could tell you about without having to do anything.
![](https://ip.bitcointalk.org/?u=http%3A%2F%2Fi.giphy.com%2FToMjGpJnf1UQN2tuwQ8.gif&t=665&c=K-IWp0O679aYYg)
|
RIP BTC Guild, April 2011 - June 2015
|
|
|
ChetnotAtkins
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
June 30, 2015, 09:50:00 PM |
|
Mempool at 21Mb. Has it ever been this high?
Meanwhile transactions per second have sharply decreased again
|
|
|
|
sturle
Legendary
Offline
Activity: 1437
Merit: 1002
https://bitmynt.no
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
June 30, 2015, 09:59:52 PM |
|
So what you are saying is you don't actually know how they are filtering, but have decided to comment that it is correct, fine, and easy to do.
If I knew, I wouldn't tell. For obvious reasons. All that I was saying is that they should not filter coins, addresses, servers, nodes,ip and etc, IMO.
Spam filters can have many configurations. Filtering IP addresses would be pointless in this case. Filtering addresses is generally pointless as well. You can have an unlimited number of addresses. BTW, the attacker, if wanting to take down the network, would keep increasing transaction amount and fees to circumvent the "filtering". The intention is not to harm, in theory. This isn't an attacker, but most likely a tester or spammer. IMO.
The attacker obviously doesn't have the funds to circumvent the filtering. Of course the intention is to harm. You don't wage a DoS attack if you don't intend to harm.
|
Sjå https://bitmynt.no for veksling av bitcoin mot norske kroner. Trygt, billig, raskt og enkelt sidan 2010. I buy with EUR and other currencies at a fair market price when you want to sell. See http://bitmynt.no/eurprice.plWarning: "Bitcoin" XT, Classic, Unlimited and the likes are scams. Don't use them, and don't listen to their shills.
|
|
|
Luke-Jr
Legendary
Offline
Activity: 2576
Merit: 1186
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
June 30, 2015, 10:01:09 PM |
|
Are you saying that all miners are blacklisting addresses? That's news to me and I'm sure other bitcoiners as well. Blacklisting is by definition about people, and no miner has ever attempted that to my knowledge. Addresses are not people, and do not even exist in the Bitcoin protocol. Instead, they are converted into scripts. Bitcoin Core has filtered what scripts it will relay or mine since Satoshi Nakamoto himself added filtering in 0.3.18 (2010 Dec 8th). To disable this, you must use (or rewrite) patches that I have maintained almost as long as I have been involved in Bitcoin. As far as I know, however, not a single miner is currently mining without any script filtering whatsoever. Eligius is treating certain inputs differently then other inputs. Plain and simple. They are saying that Bitcoin owned by one entity is not allowed to have their transactions confirmed by their found blocks.
This is not the first time that eligius (via Luke-jr) has done something similar to harm Bitcoin. Gambling sites were previously blacklisted by default by mining software when this was pretty clearly not a wanted "feature" of such software. In other words the software did things that its users did not want. This is how a lot of people describe malware. This is 100% false/FUD/lies.
|
|
|
|
|