My 2 MNs now have 5 masternode in their list. It seems to me that 1 new MN appears each ~30m ...
v0.10.9.x Help test RC2 forking issues
- Thread starter eduffield
- Start date
- Status
Code:
"54.187.152.9:19999" : 1,
"54.209.226.236:19999" : 1,
"108.61.199.47:19999" : 1,
"87.230.94.57:19999" : 1my list after several hours
daaarkcoins
Member
Code:
xxx@lvps87-230-94-57:~$ ./darkcoind masternode list | grep 19999 | sort -V
"54.209.226.236:19999" : 1,
"87.230.94.57:19999" : 1
"108.61.199.47:19999" : 1Maybe most of the nodes on http://tdrk.poolhash.org/blocks/masterlist.txt are ghosts?
Last edited by a moderator:
flare
Well-known member
My current list with 62 active MNs - started with 1..2..3..4 (see one my previous posts) - and suddenly jumped to 55+ after 1h, very much like the graph on http://tdrk.poolhash.org/
Code:
testnet3@ip-10-65-166-48:~/.darkcoin$ ./darkcoind masternode list | grep 19999 | sort -V
"23.20.78.184:19999" : 1,
"23.22.13.31:19999" : 1,
"23.23.55.5:19999" : 1,
"50.16.10.185:19999" : 1,
"50.16.13.19:19999" : 1,
"50.16.50.164:19999" : 1,
"50.16.159.173:19999" : 1,
"50.17.9.225:19999" : 1,
"50.17.56.71:19999" : 1,
"50.17.76.216:19999" : 1,
"50.19.54.118:19999" : 1,
"54.80.118.25:19999" : 1,
"54.80.186.130:19999" : 1,
"54.80.217.163:19999" : 1,
"54.80.221.149:19999" : 1,
"54.80.248.174:19999" : 1,
"54.81.150.16:19999" : 1,
"54.81.244.239:19999" : 1,
"54.82.237.183:19999" : 1,
"54.83.143.1:19999" : 1,
"54.83.151.118:19999" : 1,
"54.86.103.191:19999" : 1,
"54.87.101.82:19999" : 1,
"54.87.121.203:19999" : 1,
"54.187.152.9:19999" : 1,
"54.197.189.67:19999" : 1,
"54.197.213.80:19999" : 1,
"54.198.109.108:19999" : 1,
"54.198.145.174:19999" : 1,
"54.198.191.99:19999" : 1,
"54.198.252.150:19999" : 1,
"54.209.226.236:19999" : 1,
"54.211.202.218:19999" : 1,
"54.224.28.113:19999" : 1,
"54.224.75.195:19999" : 1,
"54.224.75.199:19999" : 1,
"54.225.6.1:19999" : 1,
"54.226.159.60:19999" : 1,
"54.227.104.145:19999" : 1,
"54.227.119.61:19999" : 1,
"54.234.251.170:19999" : 1,
"54.237.191.208:19999" : 1,
"54.242.110.149:19999" : 1,
"54.242.139.236:19999" : 1,
"54.242.152.165:19999" : 1,
"87.230.94.57:19999" : 1
"107.20.4.185:19999" : 1,
"107.20.28.12:19999" : 1,
"107.20.38.255:19999" : 1,
"107.20.122.204:19999" : 1,
"107.21.138.152:19999" : 1,
"107.22.42.148:19999" : 1,
"108.61.199.47:19999" : 1,
"184.73.34.180:19999" : 1,
"184.73.48.127:19999" : 1,
"184.73.114.32:19999" : 1,
"184.73.179.148:19999" : 1,
"184.73.179.187:19999" : 1,
"184.73.179.196:19999" : 1,
"188.226.133.22:19999" : 1,
"188.226.248.36:19999" : 1,
"204.236.211.102:19999" : 1,
Last edited by a moderator:
eduffield
Core Developer
Thanks for the super detailed report. So what you posted actually looks fine. I'll try to explain what's happening.
Block 14217
Code:
Blockheight Pubkey Votes
14214 fgAKRZmRDp53yEPegeBGu3bKmH2pYjM4o 2
14215 nyua6yT2XjttboGdjLdNMWZqSmrtqHeUE 3
14216 QSp8gRBSez2Y6MQCgGSiMsoKxnsa9HVVk 2
14217 Z92SLSwb3z2D47g47Y1E9LczimJiVFtvv 1If you mined this block, your client added the pubkey for 14217. It's also allowed to increment each existing vote by one if it agrees with it. So the code goes something like this:
On block 14214 was the pubkey to be paid fgAKRZmRDp53yEPegeBGu3bKmH2pYjM4o?
- If yes increment votes from 1 to 2
- If no and votes=1, change pubkey to the correct one and don't vote (this only can happen get no voting has happening and it looks incorrect)
- If yes increment votes from 2 to 3
- If no, don't vote
Added block 14217, pubkey Z92SLSwb3z2D47g47Y1E9LczimJiVFtvv, votes 1
eduffield
Core Developer
hmm, try restarting your client and let me know if you get the full list. If so I think I know what's causing that.
I restarted one MN and now it only sees himself
Code:
{
"188.226.248.36:19999" : 1
}Should I look for something in the log?
EDIT:
About 30m later another node shows up
Code:
{
"188.226.248.36:19999" : 1,
"98.165.130.67:19999" : 1
}Same behaviour as before
Last edited by a moderator:
flare
Well-known member
Hi Evan,
thanks for the explanation, i think i got it. The problem was not the algorithm, but my idea how the voting actually works in P2P networks - i guess i should have read the fucking manual :grin:
So this is my understanding now:
- A certain block is found by a miner
- The miner is looking up the previous block from the blockchain to get the current vote list
- Its appending the pubkey of a MN winner - the actual vote
- Additionally the miner is checking all previous vote pubkeys and if it agrees with it adding another vote - lets call it 'confirm vote'
- If the miner does not agree it replaces the pubkey with a new one. Its the job of the next miner to agree or disagree with that
So what basically confused me was the term 'vote'. My idea was that each miner is only giving ONE vote, but not confirming the others. That also explains why my math on the expected payout was wrong: I expected that only 1/6th of blocks get paid. But with the vote confirmations it makes sense again: That way each block will be paid out in the future.
And the scheme we are seeing on the explorer now is expression that the system is working as designed
Code:
Blockheight Pubkey Votes
* * 4
* * 3
* * 2
* * 1All normal, all pubkeys have been confirmed in sequence till now
Code:
Blockheight Pubkey Votes
* A 2
* * 4
* * 3
* * 2
* * 1So what you built with DRKs voting system is basically a blockchain (confirmed votes) inside the blockchain (confirmed blocks) - you are a genius man!
Thanks again,
Holger
Last edited by a moderator:
HI all, after a whole night's sleep, I have 67 masternodes on one MN's list, and 66 masternodes on the other. And the one that is missing happens to be my other one, LOL. Yah weird. Glad Evan says he thinks he knows what's causing this, so hopefully it'll get an easy fix.
daaarkcoins
Member
eduffield said:
Restarting the daemon purges the list and it only slowly starts to be filled again.
Code:
xxx@lvps87-230-94-57:~$ cat .darkcoin/testnet3/debug.log | grep masternode | grep dsee
2014-06-05 11:56:15 dsee - Got NEW masternode entry 54.209.226.236:19999
2014-06-05 11:56:15 dsee - Accepted masternode entry -1 -1
2014-06-05 12:05:24 dsee - Got NEW masternode entry 108.61.199.47:19999
2014-06-05 12:05:24 dsee - Accepted masternode entry -1 -1
2014-06-05 13:48:11 dsee - Got NEW masternode entry 93.188.161.89:19999
2014-06-05 13:48:11 dsee - Accepted masternode entry -1 -1
2014-06-05 14:24:23 dsee - Got NEW masternode entry 188.226.248.36:19999
2014-06-05 14:24:23 dsee - Accepted masternode entry -1 -1
2014-06-05 15:02:31 dsee - Got NEW masternode entry 98.165.130.67:19999
2014-06-05 15:02:31 dsee - Accepted masternode entry -1 -1
Last edited by a moderator:
flare
Well-known member
After what i learned from Evan about the voting system now, the list of MNs of a masternode does not matter - as long as your node is in the list of each pool/miner you should get voted and paid :smile:
P2Pool for DRK should include a statistics page for the MN list, so that the MN owner can check against the pools if their MN is alive and included in the votes.
eduffield
Core Developer
Opps. I found what's causing that. It'll be fixed when I push new binaries.
Ok, great to know, LOL
Oh cool, I just tried sending out more than my "extra coins" and I got an "insufficient funds" error, so that's all good. Looks like we can withdraw our extra funds (successfully sent some funds before that) and when we try to send even a small portion that cuts into the 1000 coins, we will get an error. And all this can be done while the MN is running, so no interruption.
You all probably already knew that, but I never tried, LOL
So I'm trying to mess with the masternode and cause trouble, so I setgenerate true to start mining, and it seems to have let me do this. Am I mining as a Masternode for real?
Last edited by a moderator:
Yes I sent funds from running masternodes with no issues so far... but I never took enough to leave the balance below 1k.
Not a coder, so since my node is working and getting rewards, I'll try to test the following
1) stop the node and time how long it takes to get off http://tdrk.poolhash.org/blocks/masterlist.txt
2) update and fire up my 2nd node and confirm success, and repeat step 1) again
I will then test local.online/remote. I have two VM Ubuntus in my Mac under the same router, but treated and physical machines. Any technical reason why two (or more) DHCP independent VM's under the same external IP could not control their own remote node?
1) stop the node and time how long it takes to get off http://tdrk.poolhash.org/blocks/masterlist.txt
2) update and fire up my 2nd node and confirm success, and repeat step 1) again
I will then test local.online/remote. I have two VM Ubuntus in my Mac under the same router, but treated and physical machines. Any technical reason why two (or more) DHCP independent VM's under the same external IP could not control their own remote node?
Last edited by a moderator:
To be clear, have you tried to send more than the remaining 1000 DRK? For instance, your balance is 1555 DRK. If you send out 556, will it just allow it and deactivate the node, or will is it "locked" with the masternode=1 flag?
I tried it now and got:
Code:
error: {"code":-4,"message":"Insufficient funds"}
chaeplin
Well-known member
I think you can spend it using copied wallet on other computer.
Not clear when network will invalidate spent Masternode.
1) as soon as tx received
2) tx get confirmed
vertoe
Three of Nine
I understand everything to the point where a node is chosen to be paid. Do the last 6 votes select the node to be chosen or is it only the miner which will randomly pick one?
- Status