Still haven't received an MN payment...

LucD88

Member
Aug 15, 2014
75
49
58
My Masternode is up and running (non-stop) for 2,5 weeks straight now, call it bad luck but I still haven't received a single payment. Not sure if this is a known issue, or me not having enough patience.

Anyone that could clarify what's wrong?
 
Last edited by a moderator:

fernando

Powered by Dash
Dash Core Team
Moderator
Foundation Member
May 9, 2014
1,527
2,058
283
That is really weird... mn.drk seems to be down, but all six masternodes in chaeplin's resocurces (http://drk.poolhash.org/masternode.html) see your masternode as working. I've filtered by your wallet address and you have never been expected to be paid (sometimes you are selected to be paid but aren't for different reasons, but that's not your case).

Can you give more info on your septup (remote/local, OS, version...)
 

Propulsion

The buck stops here.
Feb 26, 2014
1,008
468
183
Dash Address
XerHCGryyfZttUc6mnuRY3FNJzU1Jm9u5L

stonehedge

Well-known Member
Foundation Member
Jul 31, 2014
696
333
233
Do green lights on chaeplin and elebreth's sites prove that a master node is functioning correctly or simply that it is exists and is contactable on 9999?
 

fernando

Powered by Dash
Dash Core Team
Moderator
Foundation Member
May 9, 2014
1,527
2,058
283
Do green lights on chaeplin and elebreth's sites prove that a master node is functioning correctly or simply that it is exists and is contactable on 9999?
Those mean that chaeplin's masternodes see that masternode as working. As far as it is known up until now, if that happens, everything is working fine. Probably Propulsion is right and the explanation is that variance is a bitch. Statistically it can happen, so someone will be in this situation.
 

LucD88

Member
Aug 15, 2014
75
49
58
Is there anyone else that didn't receive a payment for nearly 3 weeks, or longer? I know variance can be a bitch, but how can I check if my MN is enlisted and available for receiving payments at all? Maybe the setup went wrong.. mehh.
 

vertoe

Three of Nine
Mar 28, 2014
2,573
1,652
1,283
Unimatrix Zero One
It seriously sucks but it's just variance. As long as everything is showing up as green in chaeplin's or elbereth's site, your MN is fully functioning.
No it's not, I'm about to do detailled stats with a statistician for my nodes and prove that the distribution is not random and that not recieving coins is not just bad variance.

Short preview:
  • 10 of 51 nodes did not recieve payments within 7 days
  • nodes with payments tend to accumulate even more payments

Is there anyone else that didn't receive a payment for nearly 3 weeks, or longer? I know variance can be a bitch, but how can I check if my MN is enlisted and available for receiving payments at all? Maybe the setup went wrong.. mehh.
Stay tuned, for now I suggest you stop the masternode, set up a new private key and start it again and watch it. (Maybe change IP too.)
 
  • Like
Reactions: stonehedge

stonehedge

Well-known Member
Foundation Member
Jul 31, 2014
696
333
233
I'm so glad to hear you say that vertoe. I've been going mad with my masternodes. I'm certain that payments are not random but don't possess the ability to prove it (or myself wrong)

Another thing that I have noticed is that EC2 t1.micro requirements for masternodes is rapidly increasing.

All of mine are operating at ~75% CPU with much higher network throughput. I'm going to keep an eye on it but I think I'm going to have to look at some higher specced options elsewhere. Running costs are about the same as I'm making in payments at the moment and thats if all three of my nodes are paying.
 
Last edited by a moderator:

stonehedge

Well-known Member
Foundation Member
Jul 31, 2014
696
333
233
Sorry, I added to my post while you were replying.

I'm a paranoid SAN specialist. I'd love to you prove me wrong so I can tell the paranoid part of my brain to shut up!
 

fernando

Powered by Dash
Dash Core Team
Moderator
Foundation Member
May 9, 2014
1,527
2,058
283
Can't promise anything, just collecting data right now!
That's gonna be interesting! You are probably the only one that can have a significant enough sample to do a decent analysis (it is easy to collect the addresses, but not so easy to get the uptime per MN).
 

thelonecrouton

Well-known Member
Foundation Member
Apr 15, 2014
1,135
813
283
It seriously sucks but it's just variance. As long as everything is showing up as green in chaeplin's or elbereth's site, your MN is fully functioning.
During RC4 testing I had 2 MNs on testnet - one was on Amazon Oregon and the other in the UK with OpenITC. With less than 30 MNs on testnet the UK one got paid about once every 3-4 days at best, orders of magnitude lower than the US one.

However you slice that math, it's something more than statistical variance.

Redid it from scratch, same thing. At the time I was also running a MN on main-net with OpenITC, it went 3 weeks without payment, switched it to Amazon and got paid within 36hrs.

My gut feeling is that some kind of network harmonics emerge, and that the fix would be a MN-net tickrate, some kind of a clock that they all sync with before propagating their comms.
 
  • Like
Reactions: fernando and moli

moli

Grizzled Member
Aug 5, 2014
3,255
1,830
1,183
During RC4 testing I had 2 MNs on testnet - one was on Amazon Oregon and the other in the UK with OpenITC. With less than 30 MNs on testnet the UK one got paid about once every 3-4 days at best, orders of magnitude lower than the US one.

However you slice that math, it's something more than statistical variance.

Redid it from scratch, same thing. At the time I was also running a MN on main-net with OpenITC, it went 3 weeks without payment, switched it to Amazon and got paid within 36hrs.

My gut feeling is that some kind of network harmonics emerge, and that the fix would be a MN-net tickrate, some kind of a clock that they all sync with before propagating their comms.
Are you saying MNs on Amazon get paid better? I had a main net MN on Amazon that didn't get paid for a week. (And if I had left it there, no guarantee it could have got paid, I have seen some MNs not get paid for weeks). Someone on BCT said he moved his MN from Singapore to London and it got paid well. To me, this is all variance, combined with Evan's knowledge on economics/finance. If every MN was guaranteed with equal payments, every darkcoin would be bought and stored in the masternodes... Think of this scenario, it would be a disaster because there would be no incentive to buy and sell darkcoins, no incentive to create market adoptions for darkcoin, and how could masternodes be paid forever without the selling and buying and paying fees.. Ok maybe you guys then would have to fight harder to get fees from miners. But the masternode system is like a lottery, there will be winning and losing, more likely there will be winning... But if a masternode isn't profitable for someone, he/she should be out there promoting the coins so they can be bought and sold, and everyone wins.
 
Last edited by a moderator:

thelonecrouton

Well-known Member
Foundation Member
Apr 15, 2014
1,135
813
283
If every MN was guaranteed with equal payments, every darkcoin would be bought and stored in the masternodes... Think of this scenario, it would be a disaster because there would be no incentive to buy and sell darkcoins, no incentive to create market adoptions for darkcoin, and how could masternodes be paid forever without the selling and buying and paying fees..
I run MN's, but I don't just feed all my MN earnings back into new MN's, I spend some of it as it was intended to be spent - anonymously. I use DRK for what it's intended for in other words, and many others will do likewise.

Saying that MN's having a guaranteed income would mean no liquidity is like saying that because bank accounts exist that pay a guaranteed interest rate, nobody actually spends money - obviously not the case.

Plus the fees come out of newly minted blocks, nothing to do with the fees miners get for processing transactions.
 
  • Like
Reactions: moli

moli

Grizzled Member
Aug 5, 2014
3,255
1,830
1,183
I run MN's, but I don't just feed all my MN earnings back into new MN's, I spend some of it as it was intended to be spent - anonymously. I use DRK for what it's intended for in other words, and many others will do likewise.

Saying that MN's having a guaranteed income would mean no liquidity is like saying that because bank accounts exist that pay a guaranteed interest rate, nobody actually spends money - obviously not the case.

Plus the fees come out of newly minted blocks, nothing to do with the fees miners get for processing transactions.
I'm just going by what I've seen from what Evan explains on his thread: https://darkcointalk.org/threads/masternode-payment-variance.1636/
If this was like bank accounts, I'm just wondering why Evan wouldn't make the MN payments to be the same for every MN, why wouldn't he make it exactly like bank savings or checking accounts with interests? And I didn't mean to say there would be 0 liquidity at all for DRK if every or almost all of darkcoins were in the masternodes. What I mean is there would be no incentive to make the darkcoin market to grow beyond masternodes. Yes the fees come out of newly minted blocks, but wouldn't mining be done in the future after all 22 mil darkcoins have been mined? I think Evan sees the future of DRK beyond that.

Just my 2 duffs.
 

stonehedge

Well-known Member
Foundation Member
Jul 31, 2014
696
333
233
My gut feeling is that some kind of network harmonics emerge, and that the fix would be a MN-net tickrate, some kind of a clock that they all sync with before propagating their comms.
I have previously seen random variance issues on a distributed network where network latency was a factor.

My worst masternode variance example was over a roughly three week period one of my masternodes was paid once and another was paid 43 times. I have since relocated the masternode that was barely getting paid to another EC2 zone and it is now being paid more regularly...three times in three days. These are short sample periods I know but still eye opening to me.

As it stands, now that the consecutive payment issues has (mostly, I still get some) been resolved, the number of payments I receive has dropped as expected.

I'm not far off it being better value to just buy DRK with what I pay Amazon every month although to be fair I should probably watch payments for a couple of months before I make such bold statements.
 

elbereth

Active Member
Dash Support Group
Mar 25, 2014
461
484
133
Costa Rica
dashninja.pl
Dash Address
XkfkHqMnhvQovo7kXQjvnNiFnQhRNZYCsz
Just so you know, now that I have a full list of blocks with payee and amounts in a database, I tried to do some analysis and while I am no mathematician it didn't look smooth repartition of payments (there was 4 visible groups, but I don't remember the details, will try to redo the analysis when possible). I can provide the full DB dump if needed, nothing secret.
 
  • Like
Reactions: fernando

karisu

Member
Jun 30, 2014
70
26
58
Well, if you want to see a small sample of "variance" check out https://drk.mn/ for the nodes
148.251.107.242 to 148.251.107.246
It is more or less 7days, exact same setup, all new private keys...
As you can see with 148.251.107.243 its kind of a winner-takes-all issue
 
  • Like
Reactions: BelStar

LucD88

Member
Aug 15, 2014
75
49
58
Anyone that can help me re-setup my MN? Previously Yidakee/GhostPlayer helped me out, but not seeing him online that much anymore..
 

DrkMiner

Member
Jun 7, 2014
204
63
88
Your IP looks okay (all green/port open).

If you wish to reinstall, I recommend a new IP and wallet address.

Even after the reinstall it might take few days for the first pay...
 

vertoe

Three of Nine
Mar 28, 2014
2,573
1,652
1,283
Unimatrix Zero One
Please stay on topic.
Cross posting:

There is one thing that I am afraid of at the moment is the masternode payment (I know some masternode owners are worried also).
The feeling if you got many masternode as I do (did..), is that the payment is weird and really not random:
  • Some of my nodes were having up to 3 payments every day for 4 weeks consistently
  • Some others nothing AT ALL during the same period
As discussed on another thread, now that I did my block page I tried to do some payment repartition to check if there was something fishy.
Here is the graph I obtained (was done for RC4 only on 21/08/2014 and I know it should be finetuned, therefore why I didn't publish it back then):


As you can see on the graph:
  • The vast majority of the masternodes received 5 or less payments since RC4 went live
  • Next group had received between 6 and 9 payments
  • And then some lucky masternodes are receiving 18+ payments.
How can this be caused by just variance? The payee selection seems not smooth at all or we would see a much smoother graph.

I know it is a difficult analysis to do and I am not mathematician, so bare with me if this is completly useless data.
As proposed in another thread, all data is available (I can provide with dumps of my database).

Long post already, I better stop.
 

LucD88

Member
Aug 15, 2014
75
49
58
So, anyone willing to help me re-setup my MN as the issue still remains...
 

GilAlexander

Member
Jun 26, 2014
84
23
48
Have my MN since 1st July, had about 42 payments till RC4, about 10 payments since RC4 till updating to .32 version, no payments at all since 27th September. Amazon EC2 Tokyo, reinstalled this one (new ip, privkey, wallet, resync blockchain, peers) a week ago. This is really strange.
Another thing, there were more payments in a day when I got one (for ex., I had 3 days with 5 payments when there were 800 mn at the network and a lot of days without any payments).
 

vertoe

Three of Nine
Mar 28, 2014
2,573
1,652
1,283
Unimatrix Zero One
*Ad* Investing in masternode farms is evening out the odds :) *Ad*

I'm still testing how reinstalling unpaid masternodes is having an impact on turnouts. Not having clear results yet.
 

oaxaca

Well-known Member
Foundation Member
Jul 8, 2014
573
832
263
How bout that variance!

3 masternodes with no payments since 8/31. Is there a concise writeup on the methodology for selection?
 
May 2, 2014
151
59
88
What is the absolute reason to even have variance at all? It sounds fishy to people.
I see other options. A fair split, 1 drk / # active nodes every mine payout.
That way, everyone is compensated equally.
Does this introduce too much complexity?
It'd be more trusted. We could verify its working properly. All celebrate we have mn online and receive some drk in short time.

If I could make Release Candidates for Darkcoin, I would make a version for a fair split for masternodes. Is there a blatant problem in that?