Bible Pay

Read 16002 times

  • Rob Andrews
  • Administrator

    • 2031


    • 27
    • June 05, 2017, 08:09:04 pm
    • Patmos, Island Of
    more
Re: Testnet - Test Proof of Giving
« Reply #60 on: December 10, 2018, 02:59:44 pm »
What happened to my donations
I sent 18*294 bbp and my pogpool says I have  294.64

Code: [Select]
{
  "Command": "pogpool",
  "ySwRWw1EEEMyGSz4iiWVAwryXbxBTumiPG": "Amount: 536.00, Weight: 1.0000, Payment_Tier: 13, Height: 85357, Address: ySwRWw1EEEMyGSz4iiWVAwryXbxBTumiPG, NickName: randrews",
  "yZVRb4Cbcoyqvw4VJmkSHPugyHxX7kbcF4": "Amount: 876.00, Weight: 0.7487, Payment_Tier: 0, Height: 85344, Address: yZVRb4Cbcoyqvw4VJmkSHPugyHxX7kbcF4, NickName: thesnat2",
  "yaSJ9FdsPcQvGjyrzxVFUPjg6wkJBQwaxc": "Amount: 1168.00, Weight: 1.0000, Payment_Tier: 10, Height: 85386, Address: yaSJ9FdsPcQvGjyrzxVFUPjg6wkJBQwaxc, NickName: thesnat1",
  "ydXPSPFWeoA9RUQoX6uBKmjnnJV6GxfwBa": "Amount: 880.00, Weight: 1.0000, Payment_Tier: 1, Height: 85297, Address: ydXPSPFWeoA9RUQoX6uBKmjnnJV6GxfwBa, NickName: jerky",
  "ygEXbdJhypUaVhYo5Q2jKATctQtVWL892K": "Amount: 1172.00, Weight: 1.0000, Payment_Tier: 11, Height: 85355, Address: ygEXbdJhypUaVhYo5Q2jKATctQtVWL892K, NickName: slybaby",
  "yhuFkP7UREXJrNrY62cBh5ZDMst7ATefCt": "Amount: 294.00, Weight: 0.2513, Payment_Tier: 0, Height: 85264, Address: yhuFkP7UREXJrNrY62cBh5ZDMst7ATefCt, NickName: testnet2",
  "0": "Count: 2, Total: 1170.0000",
  "1": "Count: 1, Total: 880.0000",
  "2": "Count: 0, Total: 0.0000",
  "3": "Count: 0, Total: 0.0000",
  "4": "Count: 0, Total: 0.0000",
  "5": "Count: 0, Total: 0.0000",
  "6": "Count: 0, Total: 0.0000",
  "7": "Count: 0, Total: 0.0000",
  "8": "Count: 0, Total: 0.0000",
  "9": "Count: 0, Total: 0.0000",
  "10": "Count: 1, Total: 1168.0000",
  "11": "Count: 1, Total: 1172.0000",
  "12": "Count: 0, Total: 0.0000",
  "13": "Count: 1, Total: 536.0000",
  "14": "Count: 0, Total: 0.0000",
  "15": "Count: 0, Total: 0.0000",
  "High Tithe": 1172.17,
  "Total Tithes": 4928.26999998,
  "My Tithes": 294.64,
  "My Payment Height": 85386
}

Here are some of the transaction ids

61836338359fa2641fb9a542d6e4bdcd242a1c505d2944ba19fc6d81074c3e1e-000
69410c348c2b6478d94f1bc0e8d0c61c6e00e613316f7628c53b78e2f1018df9-000
1bd2d45a10708ccf89a827fc12cd52f230b77a886888939cb1c1308b507062d4-000
121b947aea3282853ff90d7f79e583048d4988280ac26bf38918095207498b7a-000

So it looks like you tithed these on block 85280, and the first problem I see is txid #2, (if you look at showblock 85280 you see min diff of 1.12 age, 467 amount) and txid#2 spent a coin of amount of 98 (IE didnt meet the minimum diff req).

So that part is OK as we only induct legal tithes; but now to see if we have found a bug or not, can you please tell me - how did you tithe these on 85280, from the RPC or from the GUI?  Did the gui throw any errors and do you remember if it showed the minimum diff in red below the Send?

Also did you send from 1163?

Thanks,
Rob


  • mint
  • Newbie

    • 18


    • 0
    • December 10, 2018, 10:19:06 am
    more
Re: Testnet - Test Proof of Giving
« Reply #61 on: December 11, 2018, 01:12:07 am »
What is it that is random?  No, all tithes count if legal tithes (within diff parameter of that block).

I mean that hash is somewhat random and first legal tithe is ordered by hash. Code does not select for example youngest and smallest transaction to be used for tithe.


  • mint
  • Newbie

    • 18


    • 0
    • December 10, 2018, 10:19:06 am
    more
Re: Testnet - Test Proof of Giving
« Reply #62 on: December 11, 2018, 05:27:02 am »
So it looks like you tithed these on block 85280, and the first problem I see is txid #2, (if you look at showblock 85280 you see min diff of 1.12 age, 467 amount) and txid#2 spent a coin of amount of 98 (IE didnt meet the minimum diff req).

So that part is OK as we only induct legal tithes; but now to see if we have found a bug or not, can you please tell me - how did you tithe these on 85280, from the RPC or from the GUI?  Did the gui throw any errors and do you remember if it showed the minimum diff in red below the Send?

Also did you send from 1163?


They are send from GUI with version 1163.
Did the gui throw any errors and do you remember if it showed the minimum diff in red below the Send?

I did not see any error messages until I had no more aged coins. Until that I was able to send as many times as I wanted. I cannot recall seeing minimum diff in red.
I just reproduced this behavior and I can pump my tithes up in a very short time.


  • thesnat21
  • Administrator

    • 163


    • 15
    • March 28, 2018, 06:37:05 pm
    more
Re: Testnet - Test Proof of Giving
« Reply #63 on: December 11, 2018, 06:19:05 am »
I know we have a small sampling size, but this still seems odd, how are we ending up duplicating tiers so easily?

Also, i'm curious what happens when a tier is empty  the rewards just go to the miner?

"0": "Count: 0, Total: 0.0000",
  "1": "Count: 0, Total: 0.0000",
  "2": "Count: 0, Total: 0.0000",
  "3": "Count: 0, Total: 0.0000",
  "4": "Count: 0, Total: 0.0000",
  "5": "Count: 2, Total: 4054.0000",
  "6": "Count: 0, Total: 0.0000",
  "7": "Count: 0, Total: 0.0000",
  "8": "Count: 2, Total: 1450.0000",
  "9": "Count: 0, Total: 0.0000",
  "10": "Count: 0, Total: 0.0000",
  "11": "Count: 0, Total: 0.0000",
  "12": "Count: 2, Total: 3567.0000",
  "13": "Count: 1, Total: 869.0000",
  "14": "Count: 0, Total: 0.0000",
  "15": "Count: 0, Total: 0.0000",


  • mint
  • Newbie

    • 18


    • 0
    • December 10, 2018, 10:19:06 am
    more
Re: Testnet - Test Proof of Giving
« Reply #64 on: December 11, 2018, 07:11:10 am »
I know we have a small sampling size, but this still seems odd, how are we ending up duplicating tiers so easily?

Also, i'm curious what happens when a tier is empty  the rewards just go to the miner?

"0": "Count: 0, Total: 0.0000",
  "1": "Count: 0, Total: 0.0000",
  "2": "Count: 0, Total: 0.0000",
  "3": "Count: 0, Total: 0.0000",
  "4": "Count: 0, Total: 0.0000",
  "5": "Count: 2, Total: 4054.0000",
  "6": "Count: 0, Total: 0.0000",
  "7": "Count: 0, Total: 0.0000",
  "8": "Count: 2, Total: 1450.0000",
  "9": "Count: 0, Total: 0.0000",
  "10": "Count: 0, Total: 0.0000",
  "11": "Count: 0, Total: 0.0000",
  "12": "Count: 2, Total: 3567.0000",
  "13": "Count: 1, Total: 869.0000",
  "14": "Count: 0, Total: 0.0000",
  "15": "Count: 0, Total: 0.0000",

I wonder this duplicate too. There is small bias in each 205 block window, but it is travelling so it does not matter on long run.
There has been mention that empty blocks goes to miner totally,
 


  • sunk818
  • Full Member

    • 240


    • 7
    • April 24, 2018, 02:02:20 pm
Re: Testnet - Test Proof of Giving
« Reply #65 on: December 11, 2018, 10:50:02 am »
In exec pogpool I see BBP address twice. Is there a reason for this?

Is nickname required? Why is it there?


  • Rob Andrews
  • Administrator

    • 2031


    • 27
    • June 05, 2017, 08:09:04 pm
    • Patmos, Island Of
    more
Re: Testnet - Test Proof of Giving
« Reply #66 on: December 11, 2018, 11:05:15 am »
In exec pogpool I see BBP address twice. Is there a reason for this?

Is nickname required? Why is it there?

Yes, the address is the Key, so it listed once as the JSON key, then the rest of the row has all kinds of things in it.  This isnt really intended for prod, as I think prod would have 500 rows of data, so this is temporary.

The nickname is optional.  Its there so you can see your tithes in testnet.  I recommend using it.

For prod we can use it for the leaderboard.



  • Rob Andrews
  • Administrator

    • 2031


    • 27
    • June 05, 2017, 08:09:04 pm
    • Patmos, Island Of
    more
Re: Testnet - Test Proof of Giving
« Reply #67 on: December 11, 2018, 11:07:21 am »
I know we have a small sampling size, but this still seems odd, how are we ending up duplicating tiers so easily?

Also, i'm curious what happens when a tier is empty  the rewards just go to the miner?

"0": "Count: 0, Total: 0.0000",
  "1": "Count: 0, Total: 0.0000",
  "2": "Count: 0, Total: 0.0000",
  "3": "Count: 0, Total: 0.0000",
  "4": "Count: 0, Total: 0.0000",
  "5": "Count: 2, Total: 4054.0000",
  "6": "Count: 0, Total: 0.0000",
  "7": "Count: 0, Total: 0.0000",
  "8": "Count: 2, Total: 1450.0000",
  "9": "Count: 0, Total: 0.0000",
  "10": "Count: 0, Total: 0.0000",
  "11": "Count: 0, Total: 0.0000",
  "12": "Count: 2, Total: 3567.0000",
  "13": "Count: 1, Total: 869.0000",
  "14": "Count: 0, Total: 0.0000",
  "15": "Count: 0, Total: 0.0000",

When a tier is empty it pays 100% to the miner.  Not a problem in prod - we would easily fill up the tiers.

The clashing of tiers is not a problem; it is correct - the algorithm does not have a bug in it.



  • Rob Andrews
  • Administrator

    • 2031


    • 27
    • June 05, 2017, 08:09:04 pm
    • Patmos, Island Of
    more
Re: Testnet - Test Proof of Giving
« Reply #68 on: December 11, 2018, 11:11:41 am »
They are send from GUI with version 1163.
I did not see any error messages until I had no more aged coins. Until that I was able to send as many times as I wanted. I cannot recall seeing minimum diff in red.
I just reproduced this behavior and I can pump my tithes up in a very short time.


OK good to know.  So as of yesterday, I logged a couple things that could be bugs (the tithes not entering - sometimes - even though you didnt receive an error in the gui) and another one:  I believe our version of dash has a limited memory of utxo age; and I have a suspicion that some nodes are seeing a different memory utxo set - this is all really technical so what I need to do for the next version is create our own BBP map of these, so there is no chance of this "difference" and I believe this will also kill the possibility of a tithe not entering when it meets the parameters.  (On a side note I plan on adding a transaction fee to donations that increases by 1 bbp per donation - and giving that to the miner - I think this would stave off single block tithe attacks.  Although I do want to clarify to everyone, a legal tithe is a legal tithe, so dont get the idea that when one hogs a block they are doing something bad - if the coins are aged its legal - what Im trying to do is discourage it and make it financially unfeasible for someone to want to hog a single block and raise diff up for everyone at once by a massive amount).


Just to clarify on the fee:  It makes no difference what address or wallet # you tithe from, it would be like this:  Miner checks global memory pool, how many tithes on bbp network in this block?  If < 10, no extra fee to tithe.  If > 10, the tithe fee is 1 bbp for tithe #11, 2 bbp for tithe #12, ... etc.  Next block we start over at 0 again.  Also the tithing fees goes to the reaper.   The giver loses the tx fee.  This discourages tithe attacks by one "hog".



« Last Edit: December 11, 2018, 11:19:53 am by Rob Andrews »


  • Rob Andrews
  • Administrator

    • 2031


    • 27
    • June 05, 2017, 08:09:04 pm
    • Patmos, Island Of
    more
Re: Testnet - Test Proof of Giving
« Reply #69 on: December 11, 2018, 11:16:33 am »
I mean that hash is somewhat random and first legal tithe is ordered by hash. Code does not select for example youngest and smallest transaction to be used for tithe.
Well just to clarify this, the only element of randomness we have is the payment tier that a giver falls into.  If a user tithes, if the tithe is legal (within diff params) it most certainly goes in the pool (when we have no bugs in our code LOL), so there is no uncertainty there - no hash - no ordering - etc.

However the payment tier your tithe ends up in is absolutely uncertain.  No once can predict what tier your tithe_weight will end up in.

Since we pay 12 blocks per 205 it should smooth it out very quickly (as far as even payments) - as long as tiers are pretty packed etc.

Ill take a look at those things in the prior post today and update asap.



  • thesnat21
  • Administrator

    • 163


    • 15
    • March 28, 2018, 06:37:05 pm
    more
Re: Testnet - Test Proof of Giving
« Reply #70 on: December 11, 2018, 11:24:14 am »
Testnet packages ready

sudo add-apt-repository ppa:biblepay-official/testnet
sudo apt-get update

I'm assuming the repository wont work for bionic? :)

Err:11 http://ppa.launchpad.net/dash.org/dash/ubuntu bionic Release
  404  Not Found [IP: 91.189.95.83 80]
Reading package lists... Done
E: The repository 'http://ppa.launchpad.net/dash.org/dash/ubuntu bionic Release' does not have a Release file.
N: Updating from such a repository can't be done securely, and is therefore disabled by default.
N: See apt-secure(8) manpage for repository creation and user configuration details.



  • MIP
  • Developer

    • 227


    • 26
    • February 13, 2018, 11:55:52 am
    more
Re: Testnet - Test Proof of Giving
« Reply #71 on: December 11, 2018, 12:36:28 pm »
I'm assuming the repository wont work for bionic? :)

Err:11 http://ppa.launchpad.net/dash.org/dash/ubuntu bionic Release
  404  Not Found [IP: 91.189.95.83 80]
Reading package lists... Done
E: The repository 'http://ppa.launchpad.net/dash.org/dash/ubuntu bionic Release' does not have a Release file.
N: Updating from such a repository can't be done securely, and is therefore disabled by default.
N: See apt-secure(8) manpage for repository creation and user configuration details.

We are not supposed to have any dependencies on dash.org. It must be some remnant left behind on the contrib/depends files. But this didn't show up on mainnet PPA, and nothing has changed about that on 1.1.6.3 from 1.1.6.1.

I'll check it.


  • thesnat21
  • Administrator

    • 163


    • 15
    • March 28, 2018, 06:37:05 pm
    more
Re: Testnet - Test Proof of Giving
« Reply #72 on: December 11, 2018, 12:42:25 pm »
We are not supposed to have any dependencies on dash.org. It must be some remnant left behind on the contrib/depends files. But this didn't show up on mainnet PPA, and nothing has changed about that on 1.1.6.3 from 1.1.6.1.

I'll check it.

Odd...  I removed the dashd repo and it lets me install.


  • sunk818
  • Full Member

    • 240


    • 7
    • April 24, 2018, 02:02:20 pm
Re: Testnet - Test Proof of Giving
« Reply #73 on: December 11, 2018, 01:47:03 pm »
in the gui, donate to foundation only allows me to select tithe amount.

in cli, exec tithe allows me to specify coin_age and coin_min_amount. I wonder if exec tithe should just allow me to specify a tithe amount only? so if I type exec tithe 300, i will either be told the tithe is legit and submit, or error out because tithe amount, coin age, or coin_min_amount is not met.

i also seem to be able to select a 300 input and tithe 290. when done automatically, qt wallet wants to select a bigger input amount. is an input of 300, a legal tithe when the automated way wants mincoinvalue of 500?


  • Rob Andrews
  • Administrator

    • 2031


    • 27
    • June 05, 2017, 08:09:04 pm
    • Patmos, Island Of
    more
Re: Testnet - Test Proof of Giving
« Reply #74 on: December 11, 2018, 06:08:45 pm »
in the gui, donate to foundation only allows me to select tithe amount.

in cli, exec tithe allows me to specify coin_age and coin_min_amount. I wonder if exec tithe should just allow me to specify a tithe amount only? so if I type exec tithe 300, i will either be told the tithe is legit and submit, or error out because tithe amount, coin age, or coin_min_amount is not met.

i also seem to be able to select a 300 input and tithe 290. when done automatically, qt wallet wants to select a bigger input amount. is an input of 300, a legal tithe when the automated way wants mincoinvalue of 500?
Good point.

I made a change where 'exec tithe amount' will honor the current diff level if you only specify the amount.

Regarding the min_coin_amount that is actually a bug in 1163, and is fixed in 1164 (building now).  I believe it is the source of a few issues for us.

I also removed the 2nd address you pointed out in exec pogpool.