Sparkle Alpha Test Network Live: Mining to be Honored in Real Network

[sparkles]

The difficulty should have adjusted up 6.666 minutes per block so that means the difficulty will fall by 2% every 2.2 hours.

[turtletastic2]

Ok, that explains why the attackter had to wait every 20 blocks for a real block to show up...

Since the difficulty is sky high now, once we're at 20000 for the hard fork, it will take nearly forever to revert back to lower diff, wouldn't it? What is the max diff reduction now, 3%?

Could make sense to either start with a new blockchain then and honor all blocks up to whatever block it was were the truoble started, or allow for a faster diff correction downwards.

[sparkles]

https://github.com/sparkle5/Sparkle/releases/tag/alpha.6

[sparkles]

Found the problem and it has nothing to do with their hash power.   It has to do with some greedy miner who doesn't care to help fix the code but only to attack the code by exploiting it.

When I updated the code to only adjust the difficulty every 20 blocks, it started only CHECKING the difficulty every 20 blocks. 

I am preparing a new release that should fix that issue, it will simply hard fork off of this chain.

The fix has been checked in for everyone compiling from source.  The mac binaries are coming soon.

[lafona]

Is this person pulse mining as trogladactyl suggested earlier. Looking at where the system pauses(~3 min) from the difficulty rise, the first block after the pause comes from someone else, then this powerful miner fires back up again. Just a theory.

HEIGHT  TIMESTAMP           SIGNING DELEGATE                # TXS   SIZE    LATENCY PROCESSING TIME
===================================================================================================
18111   2014-12-07T14:39:29 SPKJm8qLn5hfgXxVg6QBx9B1XHmL... 0       129     0       0.001527       
18110   2014-12-07T14:39:28 SPKJm8qLn5hfgXxVg6QBx9B1XHmL... 0       129     0       0.001673       
18109   2014-12-07T14:39:27 SPKJm8qLn5hfgXxVg6QBx9B1XHmL... 0       129     0       0.0024         
18108   2014-12-07T14:39:26 SPKJm8qLn5hfgXxVg6QBx9B1XHmL... 0       129     0       0.001876       
18107   2014-12-07T14:39:25 SPKJm8qLn5hfgXxVg6QBx9B1XHmL... 0       129     0       0.001334       
18106   2014-12-07T14:39:24 SPKJm8qLn5hfgXxVg6QBx9B1XHmL... 0       129     0       0.001924       
18105   2014-12-07T14:39:23 SPKJm8qLn5hfgXxVg6QBx9B1XHmL... 0       129     0       0.003515       
18104   2014-12-07T14:39:22 SPKJm8qLn5hfgXxVg6QBx9B1XHmL... 0       129     0       0.002047       
18103   2014-12-07T14:39:21 SPKJm8qLn5hfgXxVg6QBx9B1XHmL... 0       129     0       0.001336       
18102   2014-12-07T14:39:20 SPKJm8qLn5hfgXxVg6QBx9B1XHmL... 0       129     0       0.001762       
18101   2014-12-07T14:39:19 SPKJm8qLn5hfgXxVg6QBx9B1XHmL... 0       129     0       0.001603       
18100   2014-12-07T14:39:18 SPK6ab1TgR2Tjx2YChhfSVeQTgMW... 0       129     0       0.002193       
18099   2014-12-07T14:36:51 SPKJm8qLn5hfgXxVg6QBx9B1XHmL... 0       129     0       0.001918       
18098   2014-12-07T14:36:50 SPKJm8qLn5hfgXxVg6QBx9B1XHmL... 0       129     0       0.00192       
18097   2014-12-07T14:36:49 SPKJm8qLn5hfgXxVg6QBx9B1XHmL... 0       129     0       0.001337       
18096   2014-12-07T14:36:48 SPKJm8qLn5hfgXxVg6QBx9B1XHmL... 0       129     0       0.001953       
18095   2014-12-07T14:36:47 SPKJm8qLn5hfgXxVg6QBx9B1XHmL... 0       129     0       0.002086       
18094   2014-12-07T14:36:46 SPKJm8qLn5hfgXxVg6QBx9B1XHmL... 0       129     0       0.001381       
18093   2014-12-07T14:36:45 SPKJm8qLn5hfgXxVg6QBx9B1XHmL... 0       129     0       0.001846       
18092   2014-12-07T14:36:44 SPKJm8qLn5hfgXxVg6QBx9B1XHmL... 0       129     0       0.001372 

[Frodo]

Indeed. That doesn't look right at all.

[arubi]

Something very weird is happening with the network.
Either someone has invested in a LOT of mining power, found a way to run the algorithm on a GPU, or has found some weakness.
Blocks are coming in FAST. Less than a second apart and always from the same public key.

Code: [Select]

# TXS   SIZE    LATENCY PROCESSING TIME
│===================================================================================================   ├
│16959   2014-12-07T13:23:39 SPKJm8qLn5hfgXxVg6QBx9B1XHmL... 0       129     0       0.001725          │
│16958   2014-12-07T13:23:38 SPKJm8qLn5hfgXxVg6QBx9B1XHmL... 0       129     0       0.00194           │
│16957   2014-12-07T13:23:37 SPKJm8qLn5hfgXxVg6QBx9B1XHmL... 0       129     0       0.001846          │
│16956   2014-12-07T13:23:36 SPKJm8qLn5hfgXxVg6QBx9B1XHmL... 0       129     0       0.001804          │
│16955   2014-12-07T13:23:35 SPKJm8qLn5hfgXxVg6QBx9B1XHmL... 0       129     0       0.001714          │
│16954   2014-12-07T13:23:34 SPKJm8qLn5hfgXxVg6QBx9B1XHmL... 0       129     0       0.00177           │
│16953   2014-12-07T13:23:33 SPKJm8qLn5hfgXxVg6QBx9B1XHmL... 0       129     0       0.00201           │
│16952   2014-12-07T13:23:32 SPKJm8qLn5hfgXxVg6QBx9B1XHmL... 0       129     0       0.001762          │
│16951   2014-12-07T13:23:31 SPKJm8qLn5hfgXxVg6QBx9B1XHmL... 0       129     0       0.001679          │
│16950   2014-12-07T13:23:30 SPKJm8qLn5hfgXxVg6QBx9B1XHmL... 0       129     0       0.001772          │
│16949   2014-12-07T13:23:29 SPKJm8qLn5hfgXxVg6QBx9B1XHmL... 0       129     0       0.001755          │
│16948   2014-12-07T13:23:28 SPKJm8qLn5hfgXxVg6QBx9B1XHmL... 0       129     0       0.001986          │
│16947   2014-12-07T13:23:27 SPKJm8qLn5hfgXxVg6QBx9B1XHmL... 0       129     0       0.001779          │
│16946   2014-12-07T13:23:26 SPKJm8qLn5hfgXxVg6QBx9B1XHmL... 0       129     0       0.00181           │
│16945   2014-12-07T13:23:25 SPKJm8qLn5hfgXxVg6QBx9B1XHmL... 0       129     0       0.001693          │
│16944   2014-12-07T13:23:24 SPKJm8qLn5hfgXxVg6QBx9B1XHmL... 0       129     0       0.001738          │
│16943   2014-12-07T13:23:23 SPKJm8qLn5hfgXxVg6QBx9B1XHmL... 0       129     0       0.001883          │
│16942   2014-12-07T13:23:22 SPKJm8qLn5hfgXxVg6QBx9B1XHmL... 0       129     0       0.001773          │
│16941   2014-12-07T13:23:21 SPKJm8qLn5hfgXxVg6QBx9B1XHmL... 0       129     0       0.001807          │
│16940   2014-12-07T13:23:17 SPKFEDZ7g7JPwmrTyiJcGBSzxsSn... 0       139     4       0.002266
It's still going on.

edit:

It's even weirder than it looks.
The fast blocks caused the difficulty to rise (all good so far), but their hashes are nothing like what the difficulty would suggest:

Code: [Select]

blockchain_get_block 17001
"previous": "0000008b60e210fd45bfa6678f701f5fcfcc15b4",
Block 17000 was mined by a "regular" miner : SPKFEDZ7g7JPwmrTyiJcGBSzxsSnFtaQpsiz

Code: [Select]

blockchain_get_block 17002
"previous": "dd9e35cd8ab6e4e23f82af69c2fdd52d1b2800af",
Block 17001 was mined by our "attacking" miner : SPKJm8qLn5hfgXxVg6QBx9B1XHmLCj4K5mMs

Does this even make sense?

[clayop]

Someone posted this in bitcointalk:

source: https://bitcointalk.org/index.php?topic=874406.msg9687816#msg9687816
Sparkle Alpha 4 QT win32 binary

The images are kinda borked, but it works and you can mine. I'll work on it more later if there's any interest.


Test it out:

Code: [Select]

username: upgradeadvice
pubkey: SPK7jfvFME4M82n1Sef3vXs28YtpzAnFz4rcXvaFZrAzG8RbLC6iB

It is based on alpha 4 and still does not work in my environment.

[bytemaster]

How is the network forking behavior?   I am curious if it is finding bugs in our fork management code for BTS.

[fran2k]

Someone posted this in bitcointalk:

source: https://bitcointalk.org/index.php?topic=874406.msg9687816#msg9687816
Sparkle Alpha 4 QT win32 binary

The images are kinda borked, but it works and you can mine. I'll work on it more later if there's any interest.


Test it out:

Code: [Select]

username: upgradeadvice
pubkey: SPK7jfvFME4M82n1Sef3vXs28YtpzAnFz4rcXvaFZrAzG8RbLC6iB

[iHashFury]

Yes I will honor it in the final chain.

Then testnet 65 will be ignored?

or (I hope) all testnets will be honoured in the final chain.

[clayop]

Can you update build_Ubuntu.MD?

[clayop]

Yes I will honor it in the final chain.

Then testnet 65 will be ignored?

[sparkles]

There are 40 connections to the test network. 

The difficulty adjustments appear to be much better now as we are averaging 3 per minute. 

There is about a 5% orphan rate. 

[sparkles]

Yes I will honor it in the final chain.