Pages:
Author

Topic: p2pool - Decentralized, Absolutely DoS-Proof, Pool Hopping-Proof Pool [archival] - page 14. (Read 35513 times)

legendary
Activity: 1500
Merit: 1022
I advocate the Zeitgeist Movement & Venus Project.
Code:
08:14:55.132000 Error in DeferredResource handler:
08:14:55.132000 Traceback (most recent call last):
08:14:55.132000   File "twisted\web\http.pyc", line 773, in requestReceived
08:14:55.428000
08:14:55.584000   File "twisted\web\server.pyc", line 127, in process
08:14:55.880000
08:14:56.036000   File "twisted\web\server.pyc", line 147, in render
08:14:56.192000
08:14:56.489000   File "p2pool\util\deferred_resource.pyc", line 24, in render
08:14:56.645000
08:14:56.801000 --- ---
08:14:56.957000   File "twisted\internet\defer.pyc", line 133, in maybeDeferred
08:14:57.097000
08:14:57.253000   File "twisted\web\resource.pyc", line 215, in render
08:14:57.409000
08:14:57.565000 twisted.web.error.UnsupportedMethod: ['HEAD', 'POST']
08:14:58.033000 Pool: 13026MH/s in 10016 shares (1375/10016 verified) Recent: 0.
00% >0H/s Shares: 0 (0 orphan, 0 dead) Peers: 4
08:14:58.033000 Average time between blocks: 4.41 days
08:14:58.033000 Pool stales: 15%
08:14:58.845000 Processing 1001 shares...
08:15:03.166000 Requesting parent share ce4f10b8 from 94.23.34.145:9333
08:15:03.166000 ... done processing 1001 shares. New: 1001 Have: 11017/~17280
08:15:26.987000 Pool: 13026MH/s in 11017 shares (2376/11017 verified) Recent: 0.
00% >0H/s Shares: 0 (0 orphan, 0 dead) Peers: 4
08:15:35.661000 Average time between blocks: 4.41 days
08:15:35.676000 Pool stales: 15%
08:15:35.863000 Requesting parent share ce4f10b8 from 94.23.34.145:9333
08:15:36.066000 Error in HeightTracker._think2:
08:15:36.066000 Traceback (most recent call last):
08:15:36.066000 Failure: twisted.internet.error.TimeoutError: User timeout cause
d connection failure.

Only seems to appear when there is "new work for worker".
New version. Bitcoinqt 0.5.1 on Win7x64.
hero member
Activity: 516
Merit: 643
P2Pool 0.8 - tag: release-0.8

Windows py2exe binary: http://u.forre.st/u/gtrmtyze/p2pool_0.8_d893094.zip
Source tarball: https://github.com/forrestv/p2pool/tarball/release-0.8

Changes:
* Worker interface now caches merkle roots and changes the timestamp to improve latency
* Fixes merged mining block submit bug
* --debug now submits all merged mining solutions, whether they match the target or not, so it can be tested
* Message displayed whenever new merged mining work arrives:
11:29:09.781837 Got new merged mining work! Difficulty: 425748.907476
* Some other efficiency improvements for miners
* ± display bug fixed
legendary
Activity: 1526
Merit: 1134
Hmm, scriptSig is the sort of canonical place to put block data. Merged mining should be able to use any index into the scriptSig. If the tools can't do that, it'd be good to fix them.

CHECKMULTISIG is a basic form of threshold cryptography. In threshold ECDSA you can create N "key shares" and T shares can be used to sign where T < N for some arbitrary T and N.

The simpler forms require a trusted dealer to create the key and then split it. Some fancier algorithms eliminate the need for a trusted dealer.

CHECKMULTISIG can do thresholding without a trusted dealer, but it's very inefficient (expensive fees). Real threshold ECDSA doesn't have any cost to the block chain, but I don't know of any publically available implementations. AFAICT it's only described in academic papers.
hero member
Activity: 516
Merit: 643
What about auto-electing a 'CONGRESS' with like-i-say 100 members,
when a new block was found and a extra funds remain,
send  to MULTISIG by all 'CONGRESS' guys ' keys,
when withdraw, auto-vote?

--as you know i can't code, just some thoughts. Is this practical?

This is possible and would be really neat, but it'd be really complex. Also, there are some problems - What if half of the nodes go offline or were reset when it's time to withdraw?

I'll think about it more..
hero member
Activity: 516
Merit: 643
Hey Forrest, what's up with the first output of your coinbase transaction? It's not a valid scriptPubKey. Is this some kind of p2pool marker tag? Why not put it in the scriptSig if so?

It is indeed a p2pool tag - it's a hash of some data structures that is used to make shares unique. It used to be in the scriptSig, but with the rise of merged mining and miners needing to add their own data to the coinbase, it was simpler to give them complete control over the coinbase contents and make this separate.
legendary
Activity: 1526
Merit: 1134
Hey Forrest, what's up with the first output of your coinbase transaction? It's not a valid scriptPubKey. Is this some kind of p2pool marker tag? Why not put it in the scriptSig if so?
hero member
Activity: 714
Merit: 500
Can you make this pool running SMPPS mode , just like Eligius?

First, there is no way to store extra funds generated during lucky periods, because the pool can't keep any secrets from its users. (Using escrow payments might be possible, but has a lot of pitfalls - nodes could collude to steal pool's savings, nodes could go offline and money is lost forever.)

Some sort of hybrid scheme similar to...


What about auto-electing a 'CONGRESS' with like-i-say 100 members,
when a new block was found and a extra funds remain,
send  to MULTISIG by all 'CONGRESS' guys ' keys,
when withdraw, auto-vote?

--as you know i can't code, just some thoughts. Is this practical?
hero member
Activity: 516
Merit: 643
Can you make this pool running SMPPS mode , just like Eligius?

First, there is no way to store extra funds generated during lucky periods, because the pool can't keep any secrets from its users. (Using escrow payments might be possible, but has a lot of pitfalls - nodes could collude to steal pool's savings, nodes could go offline and money is lost forever.)

Some sort of hybrid scheme similar to...

A substantially different variant [of PPLNS] is to pay for every share at most once. If, when going backwards in the list of shares, we encounter some that were already paid, we skip them and move on to older shares.

...is definitely technically possible, but the size of the history of payments owed would have to be limited to prevent it from growing too large. Also, it would create a much greater reward from attacking the pool - instead of just nullifying the last 24 hours of other miners' work, an attacker with 51% of the mining power could create a separate sharechain stating that he's owed some enormous amount and override the history, forcing miners to pay any payouts during lucky periods to him.
member
Activity: 75
Merit: 10
Can you make this pool running SMPPS mode , just like Eligius?

I guess it is not possible to have a SMPPS-like "buffer" with distributed nodes.
hero member
Activity: 714
Merit: 500
Can you make this pool running SMPPS mode , just like Eligius?
newbie
Activity: 55
Merit: 0
I guess I was the one who found it.

Code:
 GOT BLOCK! Passing to bitcoind!

 Cheesy
sr. member
Activity: 445
Merit: 250
No problem, happy to help Smiley
newbie
Activity: 55
Merit: 0
omg ... thats probably it .. i feel so stupid Cheesy
that log made me fear the worst Cheesy

thank you
sr. member
Activity: 445
Merit: 250
The payment is a "generation" transaction, & takes 120 blocks to "mature". You should be able to see it if you "listtransactions", but it won't be in the balance until tomorrow.

EDIT: You'll have 11.19281017 when we get to block 157726
newbie
Activity: 55
Merit: 0
I mined for the for the past few days at this pool and while finding the block.
But bitcoind and namecoind show no balance Sad
What is going on?


EDIT:
Log shows this

14:36:17.209802 Getting payout address from bitcoind...
14:36:17.514771     Pubkey request failed. Falling back to payout to address.
14:36:17.516206     ...success!
14:36:17.517064     Payout script: Address. Address: 1GxyaXnqtU9FvmjrZHfLGTTXFxVCMDKVf3
newbie
Activity: 55
Merit: 0
Since your update 0.7.1, everything runs without a problem for days. Good work!
hero member
Activity: 516
Merit: 643
Ok, i switched to bitcoin 0.5, it seems OK now!

Yes, Bitcoin 0.5.0 is required now. I'll add that to the documentation.

but still get some "error like" messages:



Those are normal. There's an old peer mining a couple of shares.. I'll change P2Pool to make it obvious that these errors don't mean that something is wrong. EDIT: Fixed in https://github.com/forrestv/p2pool/commit/59a9680
hero member
Activity: 714
Merit: 500
I've never seen that error before.. Are you using wallet encryption? If so, it's probably because Bitcoin can't generate any more keys without the passphrase, and you should do as the error says - run bitcoind.exe keypoolrefill.
yes, i'll try that when i got home.

And, can you make this pool running SMPPS mode , just like Eligius?

I think that'll attract more miner, avoiding the "Chicken & Egg" thing.

EDIT:

I've got another exception:


EDIT again:
Ok, i switched to bitcoin 0.5, it seems OK now!
bu still get some "error like" messages:




Pages:
Jump to: