[1500 TH] p2pool: Decentralized, DoS-resistant, Hop-Proof pool - page 442.

K1773R

legendary

Activity: 1792

Merit: 1008

/dev/null

Quote from: TierNolan on January 19, 2014, 02:20:28 PM

What does this mean?

Quote

Punishing share for 'Block-stale detected! height(1118c0efab8b0d8005a2995d3bf2c94d3823c91859cd30799) < height(2bee57eb6cf6965b98b6f34b8118b8ad863c2dc039618f41) or 19026666 != 19026666'! Jumping from 0b46a971 to 974819b9!

Is to cover "late" shares?

correct

TierNolan

legendary

Activity: 1232

Merit: 1094

What does this mean?

Quote

Punishing share for 'Block-stale detected! height(1118c0efab8b0d8005a2995d3bf2c94d3823c91859cd30799) < height(2bee57eb6cf6965b98b6f34b8118b8ad863c2dc039618f41) or 19026666 != 19026666'! Jumping from 0b46a971 to 974819b9!

Is to cover "late" shares?

smoothrunnings

hero member

Activity: 630

Merit: 501

Quote from: lenny_ on January 19, 2014, 11:38:42 AM

Quote from: xgtele on January 19, 2014, 08:22:20 AM

May be its time to do sudo apt-get update and sudo apt-get upgrade for your Ubuntu?

I have newest Xubuntu version, man. Fully updated.

Quote from: smoothrunnings on January 19, 2014, 08:18:43 AM

Have you restarted your Ubuntu to see if bitcoind and the P2Pool server startup again?

I just restarted p2pool. It's running fine so far on http://lenny.dnsd.me:9332/static/
Will see how it goes.

You know it's funny what you experienced just happened to me 2 minutes ago. I am using openSUSE, I restarted my pool and everything is normal again, it looks like there might be a bug in P2Pool. I wish there was a system that could stop the server and restart it when it errors out like this.

My status if anyone cares can be found at http://p2pool.smoothrunnings.ca:9332/

bkminer

full member

Activity: 216

Merit: 100

Don't let the nam-shub in your operating system.

Quote from: ikolubr on January 19, 2014, 11:57:40 AM

It is exactly what is happening with my node: bitcoin.breeba.com:3330
I deleted the data folder, restarted the server, after a while the error comes back and everything stales for few minutes. I don't know what to do.

Igor

After I lowered my bitcoind connections to 20 when this error happens my GBT latency doesn't spike anymore but my hash rate drops to zero on the pool until it's over. It's happening over and over now and I've checked and I'm on the latest code.

(also) My network rate never goes about about 60kBs on the graph and averages 12kBs)

-Rick

mdude77

legendary

Activity: 1540

Merit: 1001

Quote from: gyverlb on January 19, 2014, 12:33:04 PM

Quote from: mdude77 on January 19, 2014, 05:18:03 AM

[...]
Yes, I understand that. I still want to see some numbers!

You keep claiming you do better in p2pool than a conventional pool. Let's see numbers to prove it! We know theory and reality usually don't meet.

I don't compute these numbers anymore (at least not accurately) because:
- I read and understood the p2pool code,
- I validated them ages ago and have no reason to believe that p2pool behavior changed since (both because our income is still high enough and the code changes didn't introduce changes in the way the distribution works, stales are used, efficiency is computed, ... that could affect the basic principles),
- they are only one part of our above 100% PPS efficiency now (which makes isolating the p2pool's income and corresponding hashrate difficult, we simply use a 30-day sliding window and check the global income vs a theoretical 100+% PPS which is our target).

Reporting these numbers if they were available wouldn't even be of much interest: this would just say that our numbers match the expected p2pool behavior on our setup and you wouldn't even be able to verify that I didn't misrepresent them.
Either you trust my understanding of the system and my earlier experiment results and don't need recent numbers or you wouldn't have any reason to trust numbers I would provide (there's the possibility that I would fabricate good looking numbers hoping to lure you into a pool where your hashrate would work for me).
In my opinion you shouldn't trust anybody here (people luring others on "magical" pools looked like a sport on bitcointalk not long ago): if you can't use public data, your own understanding and/or your own experiments there's really no point in discussing this.

You're one of the few people who's number I'd assume is right.

But oh well, since you won't back it up, I see no reason to believe your claim that you come out ahead with p2pool would be accurate.

M

gyverlb

hero member

Activity: 896

Merit: 1000

Quote from: mdude77 on January 19, 2014, 05:18:03 AM

[...]
Yes, I understand that. I still want to see some numbers!

You keep claiming you do better in p2pool than a conventional pool. Let's see numbers to prove it! We know theory and reality usually don't meet.

I don't compute these numbers anymore (at least not accurately) because:
- I read and understood the p2pool code,
- I validated them ages ago and have no reason to believe that p2pool behavior changed since (both because our income is still high enough and the code changes didn't introduce changes in the way the distribution works, stales are used, efficiency is computed, ... that could affect the basic principles),
- they are only one part of our above 100% PPS efficiency now (which makes isolating the p2pool's income and corresponding hashrate difficult, we simply use a 30-day sliding window and check the global income vs a theoretical 100+% PPS which is our target).

Reporting these numbers if they were available wouldn't even be of much interest: this would just say that our numbers match the expected p2pool behavior on our setup and you wouldn't even be able to verify that I didn't misrepresent them.
Either you trust my understanding of the system and my earlier experiment results and don't need recent numbers or you wouldn't have any reason to trust numbers I would provide (there's the possibility that I would fabricate good looking numbers hoping to lure you into a pool where your hashrate would work for me).
In my opinion you shouldn't trust anybody here (people luring others on "magical" pools looked like a sport on bitcointalk not long ago): if you can't use public data, your own understanding and/or your own experiments there's really no point in discussing this.

ikolubr

newbie

Activity: 18

Merit: 0

Quote from: lenny_ on January 19, 2014, 11:38:42 AM

Quote from: xgtele on January 19, 2014, 08:22:20 AM

May be its time to do sudo apt-get update and sudo apt-get upgrade for your Ubuntu?

I have newest Xubuntu version, man. Fully updated.

Quote from: smoothrunnings on January 19, 2014, 08:18:43 AM

Have you restarted your Ubuntu to see if bitcoind and the P2Pool server startup again?

I just restarted p2pool. It's running fine so far on http://lenny.dnsd.me:9332/static/
Will see how it goes.

It is exactly what is happening with my node: bitcoin.breeba.com:3330
I deleted the data folder, restarted the server, after a while the error comes back and everything stales for few minutes. I don't know what to do.

Igor

lenny_

legendary

Activity: 1036

Merit: 1000

DARKNETMARKETS.COM

Quote from: xgtele on January 19, 2014, 08:22:20 AM

May be its time to do sudo apt-get update and sudo apt-get upgrade for your Ubuntu?

I have newest Xubuntu version, man. Fully updated.

Quote from: smoothrunnings on January 19, 2014, 08:18:43 AM

Have you restarted your Ubuntu to see if bitcoind and the P2Pool server startup again?

I just restarted p2pool. It's running fine so far on http://lenny.dnsd.me:9332/static/
Will see how it goes.

jedimstr

hero member

Activity: 798

Merit: 1000

Quote from: kjj on January 19, 2014, 09:57:23 AM

Quote from: clown on January 19, 2014, 09:16:46 AM

I was under the impression that pseudo shares are not included in the share chain, only shares with diff above the p2pool diff are included.

Another question. Let's pretend for a moment the p2pool share difficulty is at 10. A miner submits a share with difficulty 20 and the share goes into the share chain. During payout, is the value of the share capped to the share difficulty at the time of submission (10)?

I'm trying to build a proxy pool that'd would be able to payout to miners on a pseudo share basis (much less variance for miners at higher risk to pool operator). And perhaps have a vardiff system based on the pseudo shares. Has anything like this been attempted before?

Miners can intentionally pick a higher difficulty for themselves. These higher difficulty shares give a higher payout and free up space in the share window.

This was discussed to death a few pages back.

It would be best if the higher hashrate miners were to all set their min difficulties higher and thereby helping the rest of us, but having it as a voluntary action doesn't seem to be having much effect.

Anyway this could be enforced as part of VARDIFF for minimum diff shares for the big guns? Or if this is done already, can they be more aggressively applied?

bkminer

full member

Activity: 216

Merit: 100

Don't let the nam-shub in your operating system.

Quote from: nreal on January 19, 2014, 01:55:20 AM

I think that with your 5mbit upload the most you can handle is 10-15 bitcoind connections, more if you choose them carefully. Those random connections eats your upload bandwith - because theres allways someone downloading blockchain..

Thanks, nreal, I've adjusted the bitcoind connections down. Also I'm on Time Warner Cable and they have a habit of shaping large outbound transfers, or any P2P traffic they see. I'm trying to get my landlord to get FiOS but he doesn't seem interested.

kjj

legendary

Activity: 1302

Merit: 1026

Quote from: clown on January 19, 2014, 09:16:46 AM

I was under the impression that pseudo shares are not included in the share chain, only shares with diff above the p2pool diff are included.

Another question. Let's pretend for a moment the p2pool share difficulty is at 10. A miner submits a share with difficulty 20 and the share goes into the share chain. During payout, is the value of the share capped to the share difficulty at the time of submission (10)?

I'm trying to build a proxy pool that'd would be able to payout to miners on a pseudo share basis (much less variance for miners at higher risk to pool operator). And perhaps have a vardiff system based on the pseudo shares. Has anything like this been attempted before?

Miners can intentionally pick a higher difficulty for themselves. These higher difficulty shares give a higher payout and free up space in the share window.

This was discussed to death a few pages back.

mdude77

legendary

Activity: 1540

Merit: 1001

Quote from: clown on January 19, 2014, 09:42:00 AM

Quote from: mdude77 on January 19, 2014, 09:38:51 AM

It has been attempted before. I forget what it was called ... someone put up a server to do just that, complete with merged mining. It was fairly popular until it was hacked. I think you search the forums you can find it, I think the source might be available as well.

M

You'll have to enlighten me, I've searched long and hard and didn't see anything. Care to drop a name?

I just remembered the name. Was p2pmining.

https://bitcointalksearch.org/topic/offlinep2pminingcom-hybrid-p2pool-no-fee-btcnmcixci0cdevltc-66202

There wasn't any risk to the pool operator, to my knowledge. The basic principle was the subpool had it's own share chain (or equiv - could manage with a properly designed database) to track the p2pmining users. Each share was also submitted to the real chain under the p2pmining address. When a payout occurred, it was divided among the users in the subpool based on their piece of the subchain. This also allowed merged mining to occur at a sub pool level, instead of basically solo, which is what p2pool is now.

The downside, of course, was it was a mini centralized pool, subject to possible pool operator abuse and hacking. Hacking is what brought it down..

M

clown

member

Activity: 70

Merit: 10

Quote from: mdude77 on January 19, 2014, 09:38:51 AM

It has been attempted before. I forget what it was called ... someone put up a server to do just that, complete with merged mining. It was fairly popular until it was hacked. I think you search the forums you can find it, I think the source might be available as well.

M

You'll have to enlighten me, I've searched long and hard before I made this post and didn't see anything. Care to drop a name?

mdude77

legendary

Activity: 1540

Merit: 1001

Quote from: clown on January 19, 2014, 09:16:46 AM

Quote from: mdude77 on January 19, 2014, 08:57:26 AM

I believe it tells you on the p2pool output. You should also be able to divide your number of shares by the number of shares in the chain and multiply that times 25 to get an approximate value. (I think...)

M

I was under the impression that pseudo shares are not included in the share chain, only shares with diff above the p2pool diff are included.

Perhaps I misunderstood. I was referring to the shares above the p2pool diff share, which is what I thought you were asking.

Quote

Another question. Let's pretend for a moment the p2pool share difficulty is at 10. A miner submits a share with difficulty 20 and the share goes into the share chain. During payout, is the value of the share capped to the share difficulty at the time of submission (10)?

It's capped at 10, like most pools. AFAIK ozcoin is the only one that has an option to go by share value, and it uses a much different payout system.

Quote

I'm trying to build a proxy pool that'd would be able to payout to miners on a pseudo share basis (much less variance for miners at higher risk to pool operator). And perhaps have a vardiff system based on the pseudo shares. Has anything like this been attempted before?

It has been attempted before. I forget what it was called ... someone put up a server to do just that, complete with merged mining. It was fairly popular until it was hacked. I think you search the forums you can find it, I think the source might be available as well.

M

clown

member

Activity: 70

Merit: 10

Quote from: mdude77 on January 19, 2014, 08:57:26 AM

I believe it tells you on the p2pool output. You should also be able to divide your number of shares by the number of shares in the chain and multiply that times 25 to get an approximate value. (I think...)

M

I was under the impression that pseudo shares are not included in the share chain, only shares with diff above the p2pool diff are included.

Another question. Let's pretend for a moment the p2pool share difficulty is at 10. A miner submits a share with difficulty 20 and the share goes into the share chain. During payout, is the value of the share capped to the share difficulty at the time of submission (10)?

I'm trying to build a proxy pool that'd would be able to payout to miners on a pseudo share basis (much less variance for miners at higher risk to pool operator). And perhaps have a vardiff system based on the pseudo shares. Has anything like this been attempted before?

mdude77

legendary

Activity: 1540

Merit: 1001

Quote from: clown on January 19, 2014, 08:51:53 AM

Is there a way to calculate (or even approximate) the current expected payout given the amount of pseudo shares accumulated?

I believe it tells you on the p2pool output. You should also be able to divide your number of shares by the number of shares in the chain and multiply that times 25 to get an approximate value. (I think...)

M

clown

member

Activity: 70

Merit: 10

Is there a way to calculate (or even approximate) the current expected payout given the amount of pseudo shares accumulated?

xgtele

sr. member

Activity: 288

Merit: 250

May be its time to do sudo apt-get update and sudo apt-get upgrade for your Ubuntu?

smoothrunnings

hero member

Activity: 630

Merit: 501

Have you restarted your Ubuntu to see if bitcoind and the P2Pool server startup again?

lenny_

legendary

Activity: 1036

Merit: 1000

DARKNETMARKETS.COM

Today I found my p2pool BTC crashed, it was running fine for couple of days, but today I looked there and stats were not loading, so I logged via ssh to my machine and found this:

Code:

2014-01-19 11:28:39.698426 >     share.check(self)
2014-01-19 11:28:39.698468 >   File "/home/pioruns1/p2pool/p2pool/data.py", line 320, in check
2014-01-19 11:28:39.698510 >     raise ValueError('share_info invalid')
2014-01-19 11:28:39.698550 > exceptions.ValueError: share_info invalid
2014-01-19 11:28:39.710423 > Share check failed:
2014-01-19 11:28:39.710527 > Traceback (most recent call last):
2014-01-19 11:28:39.710573 >   File "/home/pioruns1/p2pool/p2pool/p2p.py", line 283, in handle_shares
2014-01-19 11:28:39.710623 >     self.node.handle_shares(result, self)
2014-01-19 11:28:39.710683 >   File "/home/pioruns1/p2pool/p2pool/node.py", line 48, in handle_shares
2014-01-19 11:28:39.710725 >     self.node.set_best_share()
2014-01-19 11:28:39.710770 >   File "/home/pioruns1/p2pool/p2pool/node.py", line 295, in set_best_share
2014-01-19 11:28:39.710812 >     best, desired, decorated_heads = self.tracker.think(self.get_height_rel_highest, self.bitcoind_work.value['previous_block'], self.bitcoind_work.value['bits'], self.known_txs_var.value)
2014-01-19 11:28:39.710855 >   File "/home/pioruns1/p2pool/p2pool/data.py", line 452, in think
2014-01-19 11:28:39.710892 >     if self.attempt_verify(share):
2014-01-19 11:28:39.710935 > ---  ---
2014-01-19 11:28:39.710976 >   File "/home/pioruns1/p2pool/p2pool/data.py", line 431, in attempt_verify
2014-01-19 11:28:39.711017 >     share.check(self)
2014-01-19 11:28:39.711060 >   File "/home/pioruns1/p2pool/p2pool/data.py", line 320, in check
2014-01-19 11:28:39.711100 >     raise ValueError('share_info invalid')
2014-01-19 11:28:39.711139 > exceptions.ValueError: share_info invalid
2014-01-19 11:28:39.723094 > Share check failed:
2014-01-19 11:28:39.723195 > Traceback (most recent call last):
2014-01-19 11:28:39.723243 >   File "/home/pioruns1/p2pool/p2pool/p2p.py", line 283, in handle_shares
2014-01-19 11:28:39.723289 >     self.node.handle_shares(result, self)
2014-01-19 11:28:39.723334 >   File "/home/pioruns1/p2pool/p2pool/node.py", line 48, in handle_shares
2014-01-19 11:28:39.723386 >     self.node.set_best_share()
2014-01-19 11:28:39.723427 >   File "/home/pioruns1/p2pool/p2pool/node.py", line 295, in set_best_share
2014-01-19 11:28:39.723469 >     best, desired, decorated_heads = self.tracker.think(self.get_height_rel_highest, self.bitcoind_work.value['previous_block'], self.bitcoind_work.value['bits'], self.known_txs_var.value)
2014-01-19 11:28:39.723513 >   File "/home/pioruns1/p2pool/p2pool/data.py", line 452, in think
2014-01-19 11:28:39.723550 >     if self.attempt_verify(share):
2014-01-19 11:28:39.723589 > ---  ---
2014-01-19 11:28:39.723636 >   File "/home/pioruns1/p2pool/p2pool/data.py", line 431, in attempt_verify
2014-01-19 11:28:39.723677 >     share.check(self)
2014-01-19 11:28:39.723720 >   File "/home/pioruns1/p2pool/p2pool/data.py", line 320, in check
2014-01-19 11:28:39.723766 >     raise ValueError('share_info invalid')
2014-01-19 11:28:39.723808 > exceptions.ValueError: share_info invalid
2014-01-19 11:28:39.742410 > Share check failed:
2014-01-19 11:28:39.742704 > Traceback (most recent call last):
2014-01-19 11:28:39.742866 >   File "/home/pioruns1/p2pool/p2pool/p2p.py", line 283, in handle_shares
2014-01-19 11:28:39.743012 >     self.node.handle_shares(result, self)
2014-01-19 11:28:39.743167 >   File "/home/pioruns1/p2pool/p2pool/node.py", line 48, in handle_shares
2014-01-19 11:28:39.743315 >     self.node.set_best_share()
2014-01-19 11:28:39.743469 >   File "/home/pioruns1/p2pool/p2pool/node.py", line 295, in set_best_share
2014-01-19 11:28:39.743625 >     best, desired, decorated_heads = self.tracker.think(self.get_height_rel_highest, self.bitcoind_work.value['previous_block'], self.bitcoind_work.value['bits'], self.known_txs_var.value)
2014-01-19 11:28:39.743776 >   File "/home/pioruns1/p2pool/p2pool/data.py", line 452, in think
2014-01-19 11:28:39.743925 >     if self.attempt_verify(share):
2014-01-19 11:28:39.744068 > ---  ---
2014-01-19 11:28:39.744218 >   File "/home/pioruns1/p2pool/p2pool/data.py", line 431, in attempt_verify
2014-01-19 11:28:39.744400 >     share.check(self)
2014-01-19 11:28:39.744553 >   File "/home/pioruns1/p2pool/p2pool/data.py", line 320, in check
2014-01-19 11:28:39.744699 >     raise ValueError('share_info invalid')
2014-01-19 11:28:39.744848 > exceptions.ValueError: share_info invalid
2014-01-19 11:28:39.770142 > Share check failed:
2014-01-19 11:28:39.770299 > Traceback (most recent call last):
2014-01-19 11:28:39.770385 >   File "/home/pioruns1/p2pool/p2pool/p2p.py", line 283, in handle_shares
2014-01-19 11:28:39.770468 >     self.node.handle_shares(result, self)
2014-01-19 11:28:39.770549 >   File "/home/pioruns1/p2pool/p2pool/node.py", line 48, in handle_shares
2014-01-19 11:28:39.770661 >     self.node.set_best_share()
2014-01-19 11:28:39.770708 >   File "/home/pioruns1/p2pool/p2pool/node.py", line 295, in set_best_share
2014-01-19 11:28:39.770755 >     best, desired, decorated_heads = self.tracker.think(self.get_height_rel_highest, self.bitcoind_work.value['previous_block'], self.bitcoind_work.value['bits'], self.known_txs_var.value)
2014-01-19 11:28:39.770798 >   File "/home/pioruns1/p2pool/p2pool/data.py", line 452, in think
2014-01-19 11:28:39.770842 >     if self.attempt_verify(share):
2014-01-19 11:28:39.770885 > ---  ---
2014-01-19 11:28:39.770928 >   File "/home/pioruns1/p2pool/p2pool/data.py", line 431, in attempt_verify
2014-01-19 11:28:39.770972 >     share.check(self)
2014-01-19 11:28:39.771013 >   File "/home/pioruns1/p2pool/p2pool/data.py", line 320, in check
2014-01-19 11:28:39.771052 >     raise ValueError('share_info invalid')
2014-01-19 11:28:39.771095 > exceptions.ValueError: share_info invalid
2014-01-19 11:28:39.782044 > Share check failed:
2014-01-19 11:28:39.782151 > Traceback (most recent call last):
2014-01-19 11:28:39.782199 >   File "/home/pioruns1/p2pool/p2pool/p2p.py", line 283, in handle_shares
2014-01-19 11:28:39.782246 >     self.node.handle_shares(result, self)
2014-01-19 11:28:39.782293 >   File "/home/pioruns1/p2pool/p2pool/node.py", line 48, in handle_shares
2014-01-19 11:28:39.782333 >     self.node.set_best_share()
2014-01-19 11:28:39.782377 >   File "/home/pioruns1/p2pool/p2pool/node.py", line 295, in set_best_share
2014-01-19 11:28:39.782424 >     best, desired, decorated_heads = self.tracker.think(self.get_height_rel_highest, self.bitcoind_work.value['previous_block'], self.bitcoind_work.value['bits'], self.known_txs_var.value)
2014-01-19 11:28:39.782466 >   File "/home/pioruns1/p2pool/p2pool/data.py", line 452, in think
2014-01-19 11:28:39.782505 >     if self.attempt_verify(share):
2014-01-19 11:28:39.782547 > ---  ---
2014-01-19 11:28:39.782589 >   File "/home/pioruns1/p2pool/p2pool/data.py", line 431, in attempt_verify
2014-01-19 11:28:39.782652 >     share.check(self)
2014-01-19 11:28:39.782699 >   File "/home/pioruns1/p2pool/p2pool/data.py", line 320, in check
2014-01-19 11:28:39.782743 >     raise ValueError('share_info invalid')
2014-01-19 11:28:39.782787 > exceptions.ValueError: share_info invalid
2014-01-19 11:28:39.792553 > Share check failed:
2014-01-19 11:28:39.792661 > Traceback (most recent call last):
2014-01-19 11:28:39.792708 >   File "/home/pioruns1/p2pool/p2pool/p2p.py", line 283, in handle_shares
2014-01-19 11:28:39.792753 >     self.node.handle_shares(result, self)
2014-01-19 11:28:39.792799 >   File "/home/pioruns1/p2pool/p2pool/node.py", line 48, in handle_shares
2014-01-19 11:28:39.792844 >     self.node.set_best_share()
2014-01-19 11:28:39.792888 >   File "/home/pioruns1/p2pool/p2pool/node.py", line 295, in set_best_share
2014-01-19 11:28:39.792934 >     best, desired, decorated_heads = self.tracker.think(self.get_height_rel_highest, self.bitcoind_work.value['previous_block'], self.bitcoind_work.value['bits'], self.known_txs_var.value)
2014-01-19 11:28:39.792982 >   File "/home/pioruns1/p2pool/p2pool/data.py", line 452, in think
2014-01-19 11:28:39.793024 >     if self.attempt_verify(share):
2014-01-19 11:28:39.793063 > ---  ---
2014-01-19 11:28:39.793118 >   File "/home/pioruns1/p2pool/p2pool/data.py", line 431, in attempt_verify
2014-01-19 11:28:39.793163 >     share.check(self)
2014-01-19 11:28:39.793207 >   File "/home/pioruns1/p2pool/p2pool/data.py", line 320, in check
2014-01-19 11:28:39.793246 >     raise ValueError('share_info invalid')
2014-01-19 11:28:39.793287 > exceptions.ValueError: share_info invalid

Log was completely full of this crap.
Litecoin p2pool on same machine is running fine now, so I presume it's not my PC problem.
My PC details:

Code:

Linux 3.11.0-15-generic x86_64 GNU/Linux
Description:	Ubuntu 13.10
Python 2.7.5+
Current p2pool version: 13.4-4-ge66e6d5

Topic: [1500 TH] p2pool: Decentralized, DoS-resistant, Hop-Proof pool - page 442. (Read 2591964 times)