[1500 TH] p2pool: Decentralized, DoS-resistant, Hop-Proof pool

[Ente]

From this quote, there is a space between 127.0.0.1: and 9332. It should be 127.0.0.1:9332 with no space. I don't know if that matters or could be your problem or not!

Edit: It doesn't look like that's the problem, from the poclbm output.

You have good eyesight! :-)
It looks like a linebreak to me, at least the command I actually entered in the shell doesnt have the space and poclbm does complain when the local p2pool node isnt running..

I have guiminer (with poclbm) running now, will check back every so often if anything changes there..

Ah, it would all be so easy if I could put linux on these boxes.. *sigh*

Ente

[Ente]

Woohoo!!

10:32:12.257000 Pool: 104GH/s in 18537 shares (13632/19553 verified) Recent: 0.1
4% >151MH/s Shares: 1 (0 orphan, 0 dead) Peers: 10

It works, yeehaw!
..and this is the day where a CLI program was beaten by a GUI program.. lol

Thank you all for your patience and help!
Now, finally, I can move some more gh to p2pool! :-)

edit:
One more question about this:
Its around six computers, with 12 miners (gpus) running. What would be the difference between running six local p2pool nodes and running one central p2pool "server"? As it takes quite long to do a proof-of-work-share, will my efficiency differ? Those miners run 24/7, so it should even out at the same efficiency eventually? I would love to have only one central p2pool node running, but my friend prefers more redundancy with independent computers..

Ente

[Prze_koles]

How do I know that we found a block? Someone said we found 2 this night, but I was mining for about 20 hours and I got only 1 block.
And what if I restart my computer? My "Payout if block" is again starting from zero...

[broken]

How do I know that we found a block? Someone said we found 2 this night, but I was mining for about 20 hours and I got only 1 block.
And what if I restart my computer? My "Payout if block" is again starting from zero...

Every found block contains donations to forrestv, so you can track them easily here:
http://blockexplorer.com/address/1Kz5QaUPDtKrj5SqW5tFkn7WZh8LmQaQi4

But we had a p2pool blockchain split, with miners running a old version on the one chain and miners running a new version on the other chain.

Block 164153 and 164207 were mined on the the old chain (old miners)
I think block 164142 was mined before the chain split, but im not sure about this block.
The new block chain (only new miners) didnt find a block since the chain split.

The donation address of forrestv contains donations from block of both chains.

edit:
Your "Payout if block" is starting from zero, because every time you restart p2pool it get a new address from your bitcoin client.
Shares previously submited with another address will still be payed out if a block is found. They are just not shown in the p2pool log, because p2pool doens't remember which addresses were yours.
If you start p2pool with -a <bitcoin address> you always use the same address and it won't start from zero.

[flower1024]

just want to say: thank you forrestv for your great work.
i really love your/our/this pool!

i hope you'll find a solution to the "share difficulty to high for one-gpu-miner"-problem as soon as it gets necessary.

EDIT: missed a word

[Costia]

"Payout if block"" doesn't reset to 0, just the shares counter in the client
also this:
http://yat.uukgoblin.net/p2pool-stats/current_payouts.html
dunno if it's accurate though

[Prze_koles]

Thank You, broken. I set persistent address now.
I've got this address: https://blockexplorer.com/address/1A3fqJJTzNEH9pUMpGDwnQS251rZktRLRq
As you can see there are two transactions connected to this address. But in my Bitcoinqt program I see only the first one. Should I rescan blockchain or something?
And from which p2pool version there was a blockchain split?

[Costia]

it takes 120 confirmations until you can use the BTC from a mined block

what is the difficulty of the fast shares?
edit:
nm found it

[Prze_koles]

it takes 120 confirmations until you can use the BTC from a mined block

But you can see it in block explorer as soon as the block is found.

Yes. First transaction showed immediately on my computer, but the second one only on blockexplorer.

[broken]

Every found block contains donations to forrestv, so you can track them easily here:
http://blockexplorer.com/address/1Kz5QaUPDtKrj5SqW5tFkn7WZh8LmQaQi4

I think those are payouts from forrestv's mining operation, not donations. (I could be wrong.)

Block 164153 and 164207 were mined on the the old chain (old miners)
I think block 164142 was mined before the chain split, but im not sure about this block.
The new block chain (only new miners) didnt find a block since the chain split.

I received payout for all three of those blocks. I'm using the latest P2Pool.

The --give-author param you can specify goes to that address (default 0.5).

I was running a old version and also received payout from all three blocks.
p2pool showed very little overall pool hash rate and difficulty of ~4, so I guess I was on a split chain.
I just updated my p2pool ( also re-downloaded p2pool chain ) and "payout if" was way below what i received.
I have no idea what happend there.

edit:
The split chain probably still had shares from before the split.

[eja]

I've created a p2pool rate icon/badge generator.

http://p2pool.us.to/p2pool_rate.php?address=<your p2pool payout address>

[Costia]

nice
and i think we got another block
http://blockchain.info/block-index/854717

[dani147624]

What caused the chain to split? Is there any reason why the new chain would be better than the old one?

[m3ta]

What caused the chain to split? Is there any reason why the new chain would be better than the old one?

Still under investigation, but what's certain, is that miners using old P2Pool versions, at this moment, are likely to have caused it. As for the causes and reasons... too technical for me, someone else might shed some light.

There is no "better" chain - all nodes can communicate with each other through the network, so i think payouts aren't affected. Just some more decentralization. (and a bit of confusion)

[Qoheleth]

What caused the chain to split? Is there any reason why the new chain would be better than the old one?

Still under investigation, but what's certain, is that miners using old P2Pool versions, at this moment, are likely to have caused it. As for the causes and reasons... too technical for me, someone else might shed some light.

There is no "better" chain - all nodes can communicate with each other through the network, so i think payouts aren't affected. Just some more decentralization. (and a bit of confusion)

My understanding from previous posts is that the old code was prone to splitting the share blockchain due to some bug. A fix was pushed to GIT, but some incompatibility between the fixed and unfixed versions meant that the share blockchain split one more time and then stuck that way.

I'm pretty sure it is affecting payouts, though; the whole point of the share blockchain is to decide who gets payed. So people who mined before the split got paid for those shares twice, and people who mined after the split are only getting payed by whatever branch they're on.

But that's just my impression from the talk in this thread. I'm not a developer either.

[forrestv]

Its around six computers, with 12 miners (gpus) running. What would be the difference between running six local p2pool nodes and running one central p2pool "server"? As it takes quite long to do a proof-of-work-share, will my efficiency differ? Those miners run 24/7, so it should even out at the same efficiency eventually? I would love to have only one central p2pool node running, but my friend prefers more redundancy with independent computers..

I would just run one... You get combined worker statistics and have less work to do setting it up.

EDIT: I can confirm that Qoheleth's summary is completely correct.

[Gabi]

Having a GUI version, maybe merged with CGminer would really help

[cheat_2_win]

I was just able to configure P2pool. Starting next week, I will move all of my 500 MH/s to P2pool from deepbit. It is nice to have decentralized mining pool. I feel P2pool will be the future of bitcoin mining.

Why next week?

Oh! I have the P2pool working in my laptop, but my miners are at another location. I will go there next week to reconfigure.

[niooron]

What decides share difficulty? I only have 300 mhash and 250 and 300 difficulty shares are kind of slow to mine.

[Qoheleth]

What decides share difficulty? I only have 300 mhash and 250 and 300 difficulty shares are kind of slow to mine.

From what I understand, it's automatically adjusted in the same way as any blockchain difficulty - just with the "network size" being the total hashpower on p2pool, and the target frequency being 10 seconds instead of 10 minutes.