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

[in2tactics]

... Then again you advocate making Bitcoin's block size limit infinite...

Satoshi never imposed a block size limit in the beginning. There was something like a 32mb network message limit, but that is it. What is your point?

[1714151829]

1714151829

[1714151829]

1714151829

[1714151829]

1714151829

[1714151829]

1714151829

[veqtrus]

... Then again you advocate making Bitcoin's block size limit infinite...

Satoshi never imposed a block size limit in the beginning. There was something like a 32mb network message limit, but that is it. What is your point?

Satoshi also:

• Introduced numerous bugs like the output coin amount overflow
• Introduced soft forks in backdoor like manner like the sigop limit
• His whitepaper contradicts itself
• Isn't a god

[jtoomim]

That's because new shares aren't 1MB yet as your fork allows. Then again you advocate making Bitcoin's block size limit infinite...

New shares are already as big as they're going to get in the absence of explicitly malicious behavior or Bitcoin blocksize increases.

Do you mean in the sense that nobody has attacked the p2pool network with large, difficult-to-propagate shares, as an adjunct to doing a selfish mining attack, because somehow that's easier than just renting a bunch of hashrate on Nicehash and 51% attacking p2pool?

Or intentionally creating large shares with the intent of forcing low-bandwidth miners off of p2pool and onto other pools because somehow there's an incentive for someone to want to do that?

Keep in mind that with a 1 MB block size limit, the only way to have shares that require 1 MB of network traffic is to create block templates that use transactions that nobody has heard of before (i.e. transactions you created yourself). This also requires hacking p2pool so that it does not notify peers of transactions when they're first seen (i.e. disabling the have_tx p2p message). Normally, shares require only 2 to 32 bytes per transaction when transmitted regardless of the transaction size, but the 50 kB/1 MB limits apply to the total summed size of the transactions (e.g. 500 bytes).

Even to the extent these attacks are possible, they wouldn't be very effective. If you want to DoS someone, there are much easier ways than mining big shares.

[KorbinDallas]

I think I know exactly what would make everyone in this discussion feel better - Both P2pool chains hit a block!

[KorbinDallas]

Just some humor for perspective...  
Digital consensus is the easy part, the human part is far more difficult to obtain.  We are all P2Poolers, and discourse to improve the software is going well.  Perhaps we should use more formal improvement channels, such as GitHub?  Just a thought. 

[tubexc]

I would say that the part of the commissions is the most difficult to obtain 

[in2tactics]

... Then again you advocate making Bitcoin's block size limit infinite...

Satoshi never imposed a block size limit in the beginning. There was something like a 32mb network message limit, but that is it. What is your point?

Satoshi also:

• Introduced numerous bugs like the output coin amount overflow
• Introduced soft forks in backdoor like manner like the sigop limit
• His whitepaper contradicts itself
• Isn't a god

That is a red herring of an argument at best. For those who believe Satoshi is a god as you so finely put it, you missed the clear and obvious argument that it was Satoshi that introduced the 1mb block size limit. However, "who" introduced the change has nothing to do with my argument. The fact is that it is an arbitrary limitation that only constrains things and does not really solve the flooding problem. It only limits it.

[jtoomim]

Perhaps we should use more formal improvement channels, such as GitHub?

Been there, done that.

I think the proper response to our little flamewar is to bring popcorn.

veqtrus makes some good technical comments and observations at times, and I value his contributions. However, when he trolls me on how my fork is gonna doom p2pool to a centralized future, I think the proper response is to troll him back, laugh it off, and move on.

I'm not very good at trolling, though. Maybe I should ask Kano for a lesson.

[veqtrus]

The fact is that it is an arbitrary limitation that only constrains things and does not really solve the flooding problem. It only limits it.

Indeed, the solution is to reserve the blockchain mostly for higher value transaction which can afford higher fees and move low value transactions which don't require the security of the blockchain to off-chain systems, preferably decentralized ones like LN.

[KorbinDallas]

Talk about interesting timing: segwit2x now has over 80% of the global hashrate. Any thoughts on how an eventual 2MB block increase will effect this P2Pool fork?

Is there still an incentive @jtoomin to run your fork if the new block size is 2 MB?

[windpath]

Hey folks, just checking in with an update on my node. The absolute last thing I want to do is to inflame any debate further, but I also feel like I have an obligation to share what my node is doing...

So as of this morning the New York Agreement (NYA) A.K.A SegWit2x has reached almost 80% miner signaling of acceptance in just a couple days.

While it is certainly not my ideal outcome it appears to be time to accept compromise, and to that end http://p2pool.org is now signaling for the NYA while still running Core 0.14.1.

Code:

"subversion": "/Satoshi:0.14.1(NYA)/",

Once there is a full release of https://github.com/btc1/bitcoin, assuming continued support and p2pool compatibility, we will upgrade

[veqtrus]

segwit2x now has over 80% of the global hashrate

What is signaled is intent to signal intent to signal intent to enforce segwit. The hard fork part can't be enforced by miners.

[veqtrus]

Hey folks, just checking in with an update on my node. The absolute last thing I want to do is to inflame any debate further, but I also feel like I have an obligation to share what my node is doing...

So as of this morning the New York Agreement (NYA) A.K.A SegWit2x has reached almost 80% miner signaling of acceptance in just a couple days.

While it is certainly not my ideal outcome it appears to be time to accept compromise, and to that end http://p2pool.org is now signaling for the NYA while still running Core 0.14.1.

Code:

"subversion": "/Satoshi:0.14.1(NYA)/",

Once there is a full release of https://github.com/btc1/bitcoin, assuming continued support and p2pool compatibility, we will upgrade

If the BIP148-like part of SegWit2x activates Core will probably merge it (otherwise there would be risk that blocks produced by Core could become stale). The hard fork part isn't mandatory at all.

Edit: Does p2pool.org also signal in its coinbase string?

[windpath]

...

Edit: Does p2pool.org also signal in its coinbase string?

Not currently.

[justdigging]

The last couple pages of BC and P2Pool discussions and science has been great.

Sorry to ask this, but what's up with "hitting a block"?

When i was looking into what pool to join, and I saw all the blocks that P2P was
hitting, and at a very high frequency, (luck factor off the charts), I got in just a
couple days after the last block was hit.... bad luck there I guess....

How does the ETA to the next block relate to the the last block, or how is
that estimated?

thank you

[daemondazz]

How does the ETA to the next block relate to the the last block, or how is
that estimated?

The last block found has absolutely zero impact on the next block found - it's purely luck.

The estimate comes from calculating how often you should statistically find a block based on the current pool hashrate compared to the total global network hashrate. But it is just that, an estimate, not a guarantee.

If you find blocks more often than statistically expected then you're having good luck. Likewise if you find blocks less often than statistically expected, you are having bad luck. Over the long term (>6 months) you should expect to get pretty close to what you statistically expect - that is the good luck and bad luck about even out. Short term it's just luck 

[chasaamm]

Let's try to make an analogy.
What's the difference betwen p2pool1,p2pool2,p2pool+ltc+segwit and bittorrent,utorrent or bitlord?
Is that you can change the torrent client and you don't loose your file.But you can't change p2pool chain without loosing
your work.
So don't play with miners work and time.
Pool operators should be interventive people,open to new ideias and forks and not open to chaos.

[in2tactics]

Let's try to make an analogy.
What's the difference betwen p2pool1,p2pool2,p2pool+ltc+segwit and bittorrent,utorrent or bitlord?
Is that you can change the torrent client and you don't loose your file.But you can't change p2pool chain without loosing
your work.
So don't play with miners work and time.
Pool operators should be interventive people,open to new ideias and forks and not open to caos.

Part of running a p2pool node is choosing your bitcoin node software, optimizing your node, and signaling your support for different BIPs when your node finds a block. When there is more than one p2pool sharechain, you also get to choose which one your node joins. If you are not running a node, you do not get a vote. If you do not like it, go to a regular pool. Quit trolling.

[chasaamm]

Let's try to make an analogy.
What's the difference betwen p2pool1,p2pool2,p2pool+ltc+segwit and bittorrent,utorrent or bitlord?
Is that you can change the torrent client and you don't loose your file.But you can't change p2pool chain without loosing
your work.
So don't play with miners work and time.
Pool operators should be interventive people,open to new ideias and forks and not open to caos.

Part of running a p2pool node is choosing your bitcoin node software, optimizing your node, and signaling your support for different BIPs when your node finds a block. When there is more than one p2pool sharechain, you also get to choose which one your node joins. If you are not running a node, you do not get a vote. If you do not like it, go to a regular pool. Quit trolling.

ATTENTION:no trolling intention,only helping.
About your answer although you are not the OP.
What's the point of runinng a node if you are afraid of mining on it?
Take a good look, there's a lot of empty nodes at this moment!

[jtoomim]

What's the difference betwen p2pool1,p2pool2,p2pool+ltc+segwit and bittorrent,utorrent or bitlord?
Is that you can change the torrent client and you don't loose your file.But you can't change p2pool chain without loosing
your work.

A better analogy would be using the bittorrent protocol to download different files. Each of the different p2pool forks represents a different share chain, a different set of users, and a different history.

If you switch share chains, you don't "lose" your work. If a block is found on the chain you left within ~3 days of you leaving, you get paid for the work you did on the old chain. The math works out such that if two p2pool chains are equally efficient, then no matter which chain you mine on, the expected value of your payout remains the same regardless of how much time you spend on each chain and how often you switch between chains.

(Of course, the two chains are not equally efficient. jtoomimnet has about 2x higher transaction fees, so your expected revenue will be higher on jtoomimnet than on mainnet. Recently, jtoomimnet's expected revenue has been about 10% higher than mainnet.)