Pages:
Author

Topic: [1500 TH] p2pool: Decentralized, DoS-resistant, Hop-Proof pool - page 89. (Read 2591919 times)

hero member
Activity: 578
Merit: 501
Quite honestly I don't know why anyone takes notice of someone who has a vested interest in p2pool failing completely; remember CK takes a percentage of any money earned on his proprietery ckpool platform.

@nicklello  How are you making that connection? p2pool is not even a significant competitor to the pools that run ck's and kano's software.

Quite honestly I don't know why anyone takes notice of someone who has a vested interest in p2pool failing completely; remember CK takes a percentage of any money earned on his proprietery ckpool platform.

Incorrect.

Just like how forrestv gets a % of every p2pool block mined if the miners allow it,
-ck gets a % of every ckpool block mined, only if the pools allows it.

Both p2pool and ckpool are open source.

The proprietary ownership of ckpool by -ck is only at the same level as forrestv's proprietary ownership of p2pool.
Both are GPL

I think kano makes a good point, not that he or ck need any defending.
legendary
Activity: 4592
Merit: 1851
Linux since 1997 RedHat 4
The image with the dead horse.. mja I can't see that p2pool is dead...

The dead horse is the topic of how to grow p2pool not p2pool itself.

Quite honestly I don't know why anyone takes notice of someone who has a vested interest in p2pool failing completely; remember CK takes a percentage of any money earned on his proprietery ckpool platform.

Incorrect.

Just like how forrestv gets a % of every p2pool block mined if the miners allow it,
-ck gets a % of every ckpool block mined, only if the pools allows it.

Both p2pool and ckpool are open source.

The proprietary ownership of ckpool by -ck is only at the same level as forrestv's proprietary ownership of p2pool.
Both are GPL
member
Activity: 193
Merit: 10
The image with the dead horse.. mja I can't see that p2pool is dead...

The dead horse is the topic of how to grow p2pool not p2pool itself.

Quite honestly I don't know why anyone takes notice of someone who has a vested interest in p2pool failing completely; remember CK takes a percentage of any money earned on his proprietery ckpool platform.
hero member
Activity: 578
Merit: 501
The image with the dead horse.. mja I can't see that p2pool is dead...

The dead horse is the topic of how to grow p2pool not p2pool itself.
legendary
Activity: 3430
Merit: 3080
Why would this not work:

p2pool spontaneously splits itself into separated sharechains in sympathy with it's overall market share, with the objective of keeping the share difficulty as low as possible on one chain. That way, the pool can expand beyond that stage where the share variance becomes too high for entry level miners. Which pool a miner is assigned to at any given time is random, to the extent that the practicalities of competitive pooled mining need observing (healthy range of hashrates within a given sub-pool, sensible minimum time hashing on a given chain, etc).

I understand that larger miners could try to game that system, but there is a limit to how many controllers with their own NIC and/or the number of VM's running through each, the amount of work required, and the various extra overheads needed, I can't see large miners exploiting that to any significant extent. It's more productive to concentrate your resources on simply improving as a solo miner, not trying to fight with the lesser fauna.

Can two or more sharechains with different share difficulties be merged such that a block find on one carries rewards through on the other(s)? I ask because I do not know. Smiley

Ah, that's an important detail, and I ought to have thought about that previously....

If manual hopping between sub-chains is not possible, then from the low-block-variance perspective, it makes sense to split the block reward between all chains, but that would not necessarily scale up (all clients would need all copies of all sharechains they were not mining.... B/W requirements are low for one p2pool sharechain, but I'm not sure if multiples of that number is feasible).


So, a low variance model would split the reward between all sharechains, but would likely sacrifice ultimate scaling. Independent rewards for each sharechain could scale up well, but the variance would be similar to best-case scenario for p2pool currently. Perhaps the latter (also simpler) method would be the best compromise (and satisfies the basic objective of redesigning p2pool in any case).
newbie
Activity: 11
Merit: 0
Bitminter have a "Pay/Hour at 1 TH/s" column at their Shifts page. So i calculated my pay/h of the latest block (409346). = 0.234mBTC per 1TH/h. (@9Th)
Bitminters last record say 0.19471mBTC, but there are higher and super low values like 0.0000...

The image with the dead horse.. mja I can't see that p2pool is dead...
hero member
Activity: 578
Merit: 501
Why would this not work:

p2pool spontaneously splits itself into separated sharechains in sympathy with it's overall market share, with the objective of keeping the share difficulty as low as possible on one chain. That way, the pool can expand beyond that stage where the share variance becomes too high for entry level miners. Which pool a miner is assigned to at any given time is random, to the extent that the practicalities of competitive pooled mining need observing (healthy range of hashrates within a given sub-pool, sensible minimum time hashing on a given chain, etc).

I understand that larger miners could try to game that system, but there is a limit to how many controllers with their own NIC and/or the number of VM's running through each, the amount of work required, and the various extra overheads needed, I can't see large miners exploiting that to any significant extent. It's more productive to concentrate your resources on simply improving as a solo miner, not trying to fight with the lesser fauna.

Can two or more sharechains with different share difficulties be merged such that a block find on one carries rewards through on the other(s)? I ask because I do not know. Smiley
legendary
Activity: 3430
Merit: 3080

Why would this not work:

p2pool spontaneously splits itself into separated sharechains in sympathy with it's overall market share, with the objective of keeping the share difficulty as low as possible on one chain. That way, the pool can expand beyond that stage where the share variance becomes too high for entry level miners. Which pool a miner is assigned to at any given time is random, to the extent that the practicalities of competitive pooled mining need observing (healthy range of hashrates within a given sub-pool, sensible minimum time hashing on a given chain, etc).

I understand that larger miners could try to game that system, but there is a limit to how many controllers with their own NIC and/or the number of VM's running through each, the amount of work required, and the various extra overheads needed, I can't see large miners exploiting that to any significant extent. It's more productive to concentrate your resources on simply improving as a solo miner, not trying to fight with the lesser fauna.
hero member
Activity: 578
Merit: 501
-ck
legendary
Activity: 4088
Merit: 1631
Ruu \o/

Unfortunately the issues remain the same as they've always been and the suggestions for how to get more people to use p2pool haven't changed in years but none of that is working. I don't see a revolutionary way out of this predicament unless it's totally redesigned. Alas I don't think anyone has solutions for the existing design even if they were to start from scratch again.
newbie
Activity: 11
Merit: 0
Hi,

I think the key is to create (more) public nodes with a great user interface, maybe public nodes should run the same or similar interface? And one thing the user want is a graph of your own Hashrate.
At my node I run the "johndoe75 interface" (//213.66.205.194:9332), if it would be possible to klick on your "adress" to se a graph on your Hashrate. Add somewhere to klick for se the old list of all users payout?
Create nodes that run the webpage on port:80 instead of 9332.. ? is it possible? and then use port:9332 for miners.?

I have tested to run p2pool on one server and the bitcoinqt on another, but it did not work? the p2pool does not need much power to run but the qt needs lot of power to generate the blocktemplate quick. now i'm running the node on a core-i7 3,6ghz*8core + intel PCIE SSD and the time to generate the template is ~300ms. to slow.. ?
<--edit, by default in bitcoin core setting "rpcallowip=127.0.0.1" is set, by adding "rpcallowip=213.66.205.194" it is possible to use a remote bitcoin node for several p2pool nodes... but if the btc node is down.. all the p2 nodes is down...
--/>


One great thing with public p2pool nodes is the lazy users do not even need to create a user account. And u can mint on different node servers with the same "username" (Payout adress).

Best regards
Magnus
member
Activity: 193
Merit: 10

Thinking about how to attract more miners to P2Pool

The miners we need to attract are probably the type of person who doesn't want to mess around setting up software ; they probably also don't want the hassle of setting up full bitcoin node; it needs to be as simple as configuring an AntMiner (ie: pointing it at a pool)

SO...

  • how do we make setting up/using p2pool super simple ?
  • Is there value in maintaining a downloading VM image/archive containing the latest Bitcoin, P2Pool and up-to-date blockchain ?
  • Can p2pool operate using a remote bitcoin node, and what are the downsides of doing this ?
  • What RPC calls does p2pool need to make in order to operate ?
  • IS IT POSSIBLE to run p2pool on a miner controller ?
  • Is there any worth in converting (parts of/all of) the code to a compiled format (ie: so it can be distributed as a packaged executable to run as a service) ?

If we *dont* try and promote p2pool and encourage more miners to join p2p will simply fade away.....

Regards,
Nick
legendary
Activity: 1066
Merit: 1098
The "relayed by" on blockchain.info is driven by a JSON file, so that file would have to be updated with every p2pool operator's IP address.  Probably not feasible.

However, the coinbase signature could be modified.  That would identify the p2pool node if forrestv implemented it like you suggested.  Base sig would be "p2pool".  If a node operator started p2pool with your proposed -btcsig parameter, then the signature would be "xxxxxx p2pool".

where does this coinbase signature appear at? If its not on relay by msg ?
I did click the block number to see the information inside on blockchain.info website. There's nothing in there mention coinbase signature.

Thanks.

If you click on the coinbase transaction inside a block on blockchain.info, then click on "show scripts and coinbase", then you can see the actual coinbase signature, if there is one.  A lot of times blockchain.info will just show an IP address as 'relayed by', but you can see who mined it if they inserted a signature.  Not all signatures will show up in the "relayed by" field unless they are in the JSON file johnnybravo mentioned.

newbie
Activity: 48
Merit: 0
The "relayed by" on blockchain.info is driven by a JSON file, so that file would have to be updated with every p2pool operator's IP address.  Probably not feasible.

However, the coinbase signature could be modified.  That would identify the p2pool node if forrestv implemented it like you suggested.  Base sig would be "p2pool".  If a node operator started p2pool with your proposed -btcsig parameter, then the signature would be "xxxxxx p2pool".

where does this coinbase signature appear at? If its not on relay by msg ?
I did click the block number to see the information inside on blockchain.info website. There's nothing in there mention coinbase signature.

Thanks.
legendary
Activity: 1344
Merit: 1024
Mine at Jonny's Pool
The "relayed by" on blockchain.info is driven by a JSON file, so that file would have to be updated with every p2pool operator's IP address.  Probably not feasible.

However, the coinbase signature could be modified.  That would identify the p2pool node if forrestv implemented it like you suggested.  Base sig would be "p2pool".  If a node operator started p2pool with your proposed -btcsig parameter, then the signature would be "xxxxxx p2pool".
newbie
Activity: 48
Merit: 0
@forrestv:

Just some suggestion to let the p2pool node operator put prefix for the relay by message if it is achievable. Cheesy
maybe configurable with -btcsig parameter.

-btcsig ABC, and it will come out "ABC P2Pool" in the blockchain.info relay by messge if that node found the block.
if the node doesn't configure anything then it will come out default just "P2Pool"
newbie
Activity: 58
Merit: 0
Awesome, thanks!

However; My node has since overwritten that section of the log file.  Is there another place older log data would be stored? :\

I will post the section that I have on the issue.
hero member
Activity: 516
Merit: 643
Hey squidicuz, I started an issue to track this problem: https://github.com/p2pool/p2pool/issues/305

Can you upload a more complete section of your P2Pool log file? Ideally an hour behind and forward or so?
newbie
Activity: 58
Merit: 0
An easy check to see if transactions are being included in the default front end is to have a look at "Current Block Value", if it's 25BTC then no tx's are being included, if it's higher then they are....

http://minefast.coincadence.com:9332/static/

Yup... Just grabbed this from the node:
Code:
Current block value: 25.29108692 BTC  

I am not sure what caused this.  Everything I check that could be of issue, seems okay.
legendary
Activity: 1258
Merit: 1027
Then I'd have to say there was a lingering bug of some sort because somewhere in the 20 minutes it must have gotten the work info from bitcoind since bitcoind is saying it's generating GBT with the CreateNewBlock message. Check if you have the p2pool logs prior to that block find for some kind of unhappiness there.
Indeed.  There is something wrong here, but I don't have much feedback other than empty block and logs that show processing work.  Then the logs look normal except for
Code:
2016-04-14 23:21:13.403123 > Block submittal result: False (u'duplicate') Expected: True

It may be some sort of bug.

An easy check to see if transactions are being included in the default front end is to have a look at "Current Block Value", if it's 25BTC then no tx's are being included, if it's higher then they are....

http://minefast.coincadence.com:9332/static/
Pages:
Jump to: