[1500 TH] p2pool: Decentralized, DoS-resistant, Hop-Proof pool - page 231.

corrow

member

Activity: 97

Merit: 10

Quote from: corrow on January 17, 2015, 02:16:15 PM

Quote from: IYFTech on January 17, 2015, 02:09:24 PM

I see you're using an Rc version: v99900 - Maybe that's what's causing your errors. I only use the stable release versions for mining.......

OMG, Why I didn't saw that myself? Embarrassed

I'm gonna deploy a stable release of Bitcoin Core to factor this out! Thanks!

I've been monitoring my P2Pool Node very closely since the update to the last Bitcoin Core version. I found out with Block #339.597 (which took about 50mins and had more than 2000tx did not grow beyond the the maxblocksize of 1.000.000! This is great news now the P2Pool did not gave any "playload too long" error at console. The update to a later Bitcoin Core did the trick. I would advise everyone to update their Bitcoin Core concerning this issue!

Thanks for the good help and your efforts everyone!!!

mahrens917

member

Activity: 78

Merit: 10

Quote from: corrow on January 18, 2015, 09:02:53 AM

Quote from: mahrens917 on January 17, 2015, 06:03:30 PM

Quote from: corrow on January 16, 2015, 05:46:32 PM

Well when recoding p2pool, It would be Nice to solve the "payload too long" issue as well.

Does somebody know how to deal with this? Huh

Please check this post: https://bitcointalksearch.org/topic/m.10164796

Roll Eyes

So doing some digging, the "payload too long" error comes about when a packet's length is greater than max_payload_length. This can be seen in p2pool/util/p2protocol.py.

This is called from p2pool/bitcoin/p2p.py and p2pool/p2p.py where it sets max_payload_length to 1,000,000.

I am going to change the code to print out the packet size, as well as, change the max_payload_length to twice the size. I will let the code run for awhile and see what happens and report back.

If I'm correct the 'maxblocksize' is hardcoded to 1MB into the Bitcoin protocol. So I guess changing the payload length does do the trick within P2Pool code (until the blocksize will be increased). Well it is still unclear to me what part of the code you have changed. Logically it had to do with Bitcoin Core which is responsible for "packaging" the block and verify the hash of it. It would be awesome if you still could update us with news. I've also updated my Bitcoin Core and I'm waiting for a long blocktime with lots of transactions to debug my new setup. So, I'll keep you guys posted! Thanks!

Yes, the block should not be coming into p2pool over 1,000,000 or perhaps it is sometimes coming in with a size of 1,024,000 (another meaning of MB). Or perhaps p2pool or bitcoind is somehow corrupting it at certain times which is causing this issue. We'll see and my logging should let us know a little better what is happening.

idonothave

full member

Activity: 238

Merit: 100

Quote from: Gumara on January 17, 2015, 06:52:55 PM

I have one question. Is it recommended to give the address with the -a flag and save this address is always the same? Does it make a difference?

Depends on what You preffer. If to mine to local wallet (where bitcoin daemon is running) or to exteral wallet. I preffer to mine to external wallet with -a option. Easier and more comfortable for me.

corrow

member

Activity: 97

Merit: 10

Quote from: mahrens917 on January 17, 2015, 06:03:30 PM

Quote from: corrow on January 16, 2015, 05:46:32 PM

Well when recoding p2pool, It would be Nice to solve the "payload too long" issue as well.

Does somebody know how to deal with this? Huh

Please check this post: https://bitcointalksearch.org/topic/m.10164796

Roll Eyes

So doing some digging, the "payload too long" error comes about when a packet's length is greater than max_payload_length. This can be seen in p2pool/util/p2protocol.py.

This is called from p2pool/bitcoin/p2p.py and p2pool/p2p.py where it sets max_payload_length to 1,000,000.

I am going to change the code to print out the packet size, as well as, change the max_payload_length to twice the size. I will let the code run for awhile and see what happens and report back.

If I'm correct the 'maxblocksize' is hardcoded to 1MB into the Bitcoin protocol. So I guess changing the payload length does do the trick within P2Pool code (until the blocksize will be increased). Well it is still unclear to me what part of the code you have changed. Logically it had to do with Bitcoin Core which is responsible for "packaging" the block and verify the hash of it. It would be awesome if you still could update us with news. I've also updated my Bitcoin Core and I'm waiting for a long blocktime with lots of transactions to debug my new setup. So, I'll keep you guys posted! Thanks!

idonothave

full member

Activity: 238

Merit: 100

Quote from: mahrens917 on January 17, 2015, 05:46:08 PM

Quote from: idonothave on January 17, 2015, 02:50:30 PM

Quote from: corrow on January 17, 2015, 02:03:21 PM

Quote from: idonothave on January 16, 2015, 09:08:26 PM

Quote from: corrow on January 16, 2015, 05:46:32 PM

Well when recoding p2pool, It would be Nice to solve the "payload too long" issue as well.

Does somebody know how to deal with this? Huh

Please check this post: https://bitcointalksearch.org/topic/m.10164796

Roll Eyes

What is Your Bitcoind GetBlockTemplate Latency? Has You p2pool node external IP to see stats?

Thanks for the reply!

Well my stats looks like this:

Code:

P2Pool Status
Local rate 	155.77 GH/s (DOA 3.67 GH/s / 2.35%) 	Shares 	Total: 2 (Orphan: 0, Dead: 1, Unknown: 1)
Global pool rate 	1.93 PH/s (DOA 321.61 TH/s / 16.67%) 	Share difficulty 	6,707,631.61
Current block value 		Expected time to share 	2d 3h 22m 33 s
Total node payout if block found NOW 		Expected time to block 	1d 3h 11m 10 s
Node peers 	[in] 2 / 30 [out] 	Node uptime 	2d 6h 35m 22 s
Node p2pool version 	13.5-73-g12ca9d8-dirty 	Protocol version 	1300
Node efficiency 	60.00% 	Node getwork latency 	305 ms
Node fee 	0% 	Node donation 	0%

Bitcoin Status
Network rate 	303.41 PH/s 	Blocks 	339,372
Current block 	998,775 bytes / 1,600 txs 	Block difficulty 	43,971,662,056.08958
Hashes to win 	9.22 EH 	Block probability per GH 	5.295015761671106e-12%
Total Bitcoins 		Expected time to block 	0h 0m 30 s
Exchange rate 		Block interval 	0h 10m 30 s
Connections 	[in] 12 / 523 [out] 	Node uptime 	2d 6h 37m 53 s
Node bitcoind version 	99900 	Protocol version 	70002
Relay fee 		Peer latency 	675 ms

I assume my Bitcoind GetBlockTemplate Latency is 675ms, right? From what I've read, the latencies are OK and they are stable when the "payload too long" error is not occurring. Only when the error comes up in the console of p2pool then those latencies are indeed skyrocketing. As I understand from the logs and statistics this payload error does not look like a cause from latencies, but latencies are the symptoms of the payload error. So where all the sudden arises this payload error from? Is it the "maxblocksize" of Bitcoind Core which is 1MB? Because the payload error is always in combination when the blocktime between blocks were +10mins (and +2000 transactions). Get the idea the problem comes from that direction.

Any thoughts?

Yes.
1) Switch to latest p2pool version.
git clone https://github.com/forrestv/p2pool
2) to limit outgoing/incomming connections of p2pool use this settings:
--max-conns 6 \
--outgoing-conns 6 \
3) use 0.9.4.0 bitcoin core version with bitcoin.conf like this (nothing more is necessary):
server=1
rpcuser=your_rpc_user
rpcpassword=your_rpc_password

Thanks for the tips. I have the latest p2pool and bitcoind code and have limited my incoming and outgoing p2pool connections. However, my bitcoin.conf file does set blockmaxsize, minrelaytxfee, maxconnections, blockminsize, blockprioritysize, and mintxfee. Do you believe that one of those settings could cause this issue within p2pool?

Do You know the reason for each one parameter in Your bitcoin.conf file? If no, or they are not really necessary from very good reason, remove them. Give it a chance to work. Wink

Gumara

full member

Activity: 141

Merit: 100

I have one question. Is it recommended to give the address with the -a flag and save this address is always the same? Does it make a difference?

mahrens917

member

Activity: 78

Merit: 10

Quote from: corrow on January 16, 2015, 05:46:32 PM

Well when recoding p2pool, It would be Nice to solve the "payload too long" issue as well.

Does somebody know how to deal with this? Huh

Please check this post: https://bitcointalksearch.org/topic/m.10164796

Roll Eyes

So doing some digging, the "payload too long" error comes about when a packet's length is greater than max_payload_length. This can be seen in p2pool/util/p2protocol.py.

This is called from p2pool/bitcoin/p2p.py and p2pool/p2p.py where it sets max_payload_length to 1,000,000.

I am going to change the code to print out the packet size, as well as, change the max_payload_length to twice the size. I will let the code run for awhile and see what happens and report back.

IYFTech

hero member

Activity: 686

Merit: 500

WANTED: Active dev to fix & re-write p2pool in C

I leave my conf file with the bear minimum needed, and I don't adjust my p2pool settings either - everything works perfectly using the standard settings for me apart from maxconnections to reduce the bandwidth usage - my ADSL line is shite Cheesy

mahrens917

member

Activity: 78

Merit: 10

Quote from: idonothave on January 17, 2015, 02:50:30 PM

Quote from: corrow on January 17, 2015, 02:03:21 PM

Quote from: idonothave on January 16, 2015, 09:08:26 PM

Quote from: corrow on January 16, 2015, 05:46:32 PM

Well when recoding p2pool, It would be Nice to solve the "payload too long" issue as well.

Does somebody know how to deal with this? Huh

Please check this post: https://bitcointalksearch.org/topic/m.10164796

Roll Eyes

What is Your Bitcoind GetBlockTemplate Latency? Has You p2pool node external IP to see stats?

Thanks for the reply!

Well my stats looks like this:

Code:

P2Pool Status
Local rate 	155.77 GH/s (DOA 3.67 GH/s / 2.35%) 	Shares 	Total: 2 (Orphan: 0, Dead: 1, Unknown: 1)
Global pool rate 	1.93 PH/s (DOA 321.61 TH/s / 16.67%) 	Share difficulty 	6,707,631.61
Current block value 		Expected time to share 	2d 3h 22m 33 s
Total node payout if block found NOW 		Expected time to block 	1d 3h 11m 10 s
Node peers 	[in] 2 / 30 [out] 	Node uptime 	2d 6h 35m 22 s
Node p2pool version 	13.5-73-g12ca9d8-dirty 	Protocol version 	1300
Node efficiency 	60.00% 	Node getwork latency 	305 ms
Node fee 	0% 	Node donation 	0%

Bitcoin Status
Network rate 	303.41 PH/s 	Blocks 	339,372
Current block 	998,775 bytes / 1,600 txs 	Block difficulty 	43,971,662,056.08958
Hashes to win 	9.22 EH 	Block probability per GH 	5.295015761671106e-12%
Total Bitcoins 		Expected time to block 	0h 0m 30 s
Exchange rate 		Block interval 	0h 10m 30 s
Connections 	[in] 12 / 523 [out] 	Node uptime 	2d 6h 37m 53 s
Node bitcoind version 	99900 	Protocol version 	70002
Relay fee 		Peer latency 	675 ms

I assume my Bitcoind GetBlockTemplate Latency is 675ms, right? From what I've read, the latencies are OK and they are stable when the "payload too long" error is not occurring. Only when the error comes up in the console of p2pool then those latencies are indeed skyrocketing. As I understand from the logs and statistics this payload error does not look like a cause from latencies, but latencies are the symptoms of the payload error. So where all the sudden arises this payload error from? Is it the "maxblocksize" of Bitcoind Core which is 1MB? Because the payload error is always in combination when the blocktime between blocks were +10mins (and +2000 transactions). Get the idea the problem comes from that direction.

Any thoughts?

Yes.
1) Switch to latest p2pool version.
git clone https://github.com/forrestv/p2pool
2) to limit outgoing/incomming connections of p2pool use this settings:
--max-conns 6 \
--outgoing-conns 6 \
3) use 0.9.4.0 bitcoin core version with bitcoin.conf like this (nothing more is necessary):
server=1
rpcuser=your_rpc_user
rpcpassword=your_rpc_password

Thanks for the tips. I have the latest p2pool and bitcoind code and have limited my incoming and outgoing p2pool connections. However, my bitcoin.conf file does set blockmaxsize, minrelaytxfee, maxconnections, blockminsize, blockprioritysize, and mintxfee. Do you believe that one of those settings could cause this issue within p2pool?

windpath

legendary

Activity: 1258

Merit: 1027

Quote from: mahrens917 on January 17, 2015, 03:44:21 PM

Reddit has an up voted thread asking if p2pool is worth it compared to other pools. Take a look and you can add your comments: http://www.reddit.com/r/BitcoinMining/comments/2sqi3l/miners_of_reddit_do_you_have_statistics_on/

Thx for posting, also a thread on main Bitcoin Reddit:

http://www.reddit.com/r/Bitcoin/comments/2sqh0a/decentralising_mining_making_p2pool_more/

mdude77

legendary

Activity: 1540

Merit: 1001

Quote from: sEpuLchEr on January 17, 2015, 03:34:20 PM

oooiii... what happen to the hashrate??

Are you guys all stuck trying to get pypy to work or merge mining? LOL.

Don't forget the hashrate isn't really the hashrate. It's a calculated value based on how many shares have been submitted in the last X minutes.

Could easily be bad luck across the board.

M

mahrens917

member

Activity: 78

Merit: 10

Reddit has an up voted thread asking if p2pool is worth it compared to other pools. Take a look and you can add your comments: http://www.reddit.com/r/BitcoinMining/comments/2sqi3l/miners_of_reddit_do_you_have_statistics_on/

sEpuLchEr

sr. member

Activity: 248

Merit: 250

Are we there yet?

oooiii... what happen to the hashrate??

Are you guys all stuck trying to get pypy to work or merge mining? LOL.

idonothave

full member

Activity: 238

Merit: 100

worthless coins renaissance Kiss

====================
small namecoin merged mining howto

(I suppose You are already running functional p2pool node... ehm... with linux)
download namecoin:
git clone https://github.com/namecoin/namecoin.git
go there and compile it:
cd namecoin
qmake namecoin-qt.pro
make
edit namecoin.conf file:
nano ~/.namecoin/namecoin.conf
this should be there:
server=1
rpcuser=your_nmc_user
rpcpassword=your_nmc_password
rpcport=7333
rpcallowip=*
then run namecoin-qt... and wait a little moment, minutes, hours, days... until synced (make coffee while waiting)
add to Your p2pool command this parameter:
--merged http://your_nmc_user:[email protected]:7333/
rerun p2pool node
check data/bitcoin/log for merged word:
grep merged data/bitcoin/log |wc -l
if You get non zero value, it works (probably)
make coffee and tell guys You got it!

idonothave

full member

Activity: 238

Merit: 100

Quote from: windpath on January 17, 2015, 01:39:03 PM

Quote from: IYFTech on January 17, 2015, 12:54:40 PM

Quote from: windpath on January 17, 2015, 12:42:39 PM

Quote from: IYFTech on January 17, 2015, 12:23:25 PM

Quote from: idonothave on January 17, 2015, 11:47:25 AM

I suppose daemons has to be compiled with wallet enabled and run at the same machine as bitcoind and p2pool if succeed will create there mining address?

All on the same machine, yes - but p2pool won't create an address on other daemons, payments will go to the default daemon wallet address.

How many of those coins is there still a market for?

All except FSC & GRP... Wink

Pretty cool, nice way to add a little to the bottom line

I suppose merged mined coins I will get the whole block, right? So for NMC 25 coins with my 8600GHs I will get once per 100 days, as I see.

idonothave

full member

Activity: 238

Merit: 100

Quote from: corrow on January 17, 2015, 02:03:21 PM

Quote from: idonothave on January 16, 2015, 09:08:26 PM

Quote from: corrow on January 16, 2015, 05:46:32 PM

Well when recoding p2pool, It would be Nice to solve the "payload too long" issue as well.

Does somebody know how to deal with this? Huh

Please check this post: https://bitcointalksearch.org/topic/m.10164796

Roll Eyes

What is Your Bitcoind GetBlockTemplate Latency? Has You p2pool node external IP to see stats?

Thanks for the reply!

Well my stats looks like this:

Code:

P2Pool Status
Local rate 	155.77 GH/s (DOA 3.67 GH/s / 2.35%) 	Shares 	Total: 2 (Orphan: 0, Dead: 1, Unknown: 1)
Global pool rate 	1.93 PH/s (DOA 321.61 TH/s / 16.67%) 	Share difficulty 	6,707,631.61
Current block value 		Expected time to share 	2d 3h 22m 33 s
Total node payout if block found NOW 		Expected time to block 	1d 3h 11m 10 s
Node peers 	[in] 2 / 30 [out] 	Node uptime 	2d 6h 35m 22 s
Node p2pool version 	13.5-73-g12ca9d8-dirty 	Protocol version 	1300
Node efficiency 	60.00% 	Node getwork latency 	305 ms
Node fee 	0% 	Node donation 	0%

Bitcoin Status
Network rate 	303.41 PH/s 	Blocks 	339,372
Current block 	998,775 bytes / 1,600 txs 	Block difficulty 	43,971,662,056.08958
Hashes to win 	9.22 EH 	Block probability per GH 	5.295015761671106e-12%
Total Bitcoins 		Expected time to block 	0h 0m 30 s
Exchange rate 		Block interval 	0h 10m 30 s
Connections 	[in] 12 / 523 [out] 	Node uptime 	2d 6h 37m 53 s
Node bitcoind version 	99900 	Protocol version 	70002
Relay fee 		Peer latency 	675 ms

I assume my Bitcoind GetBlockTemplate Latency is 675ms, right? From what I've read, the latencies are OK and they are stable when the "payload too long" error is not occurring. Only when the error comes up in the console of p2pool then those latencies are indeed skyrocketing. As I understand from the logs and statistics this payload error does not look like a cause from latencies, but latencies are the symptoms of the payload error. So where all the sudden arises this payload error from? Is it the "maxblocksize" of Bitcoind Core which is 1MB? Because the payload error is always in combination when the blocktime between blocks were +10mins (and +2000 transactions). Get the idea the problem comes from that direction.

Any thoughts?

Yes.
1) Switch to latest p2pool version.
git clone https://github.com/forrestv/p2pool
2) to limit outgoing/incomming connections of p2pool use this settings:
--max-conns 6 \
--outgoing-conns 6 \
3) use 0.9.4.0 bitcoin core version with bitcoin.conf like this (nothing more is necessary):
server=1
rpcuser=your_rpc_user
rpcpassword=your_rpc_password

MissouriMiner

sr. member

Activity: 312

Merit: 250

Quote from: mdude77 on January 17, 2015, 02:25:27 PM

Quote from: idonothave on January 13, 2015, 05:19:25 AM

small pypy howto

... download pypy binary, I choosed 64 bit version to my ubuntu 14.04 LTS
wget https://bitbucket.org/pypy/pypy/downloads/pypy-2.4.0-linux64.tar.bz2
... untar it
tar xvf pypy-2.4.0-linux64.tar.bz2
... create symbolic link
sudo ln -sf /where_untared/pypy-2.4.0-linux64/bin/pypy /usr/bin/pypy
... download pip manager
wget https://bootstrap.pypa.io/get-pip.py
... install pip manager
pypy get-pip.py
... install twisted
pip install twisted
... done

use pypy instead python

optional

... download latest forrestv p2pool version
git clone https://github.com/forrestv/p2pool.git
... run it
... make coffee

I try just now with latest git p2pool version 13.4-62-g077b2c1, seems well

Anyone succeed in getting this to work under windows? I got pip installed, now I'm having a problem compiling twisted, complains about not finding rpc\rpc.h.

M

I've tested it in Windows 8.1 64-bit, but not running it now. Even though the OS is 64-bit, I had to use 32-bit for all the required packages. Also, I wouldn't bother compiling twisted since you can download the Windows package already compiled. Same goes for pywin, zope, WMI, and mingw. All 32-bit packages pre-compiled. This worked for me.

mdude77

legendary

Activity: 1540

Merit: 1001

Quote from: idonothave on January 13, 2015, 05:19:25 AM

small pypy howto

... download pypy binary, I choosed 64 bit version to my ubuntu 14.04 LTS
wget https://bitbucket.org/pypy/pypy/downloads/pypy-2.4.0-linux64.tar.bz2
... untar it
tar xvf pypy-2.4.0-linux64.tar.bz2
... create symbolic link
sudo ln -sf /where_untared/pypy-2.4.0-linux64/bin/pypy /usr/bin/pypy
... download pip manager
wget https://bootstrap.pypa.io/get-pip.py
... install pip manager
pypy get-pip.py
... install twisted
pip install twisted
... done

use pypy instead python

optional

... download latest forrestv p2pool version
git clone https://github.com/forrestv/p2pool.git
... run it
... make coffee

I try just now with latest git p2pool version 13.4-62-g077b2c1, seems well

Anyone succeed in getting this to work under windows? I got pip installed, now I'm having a problem compiling twisted, complains about not finding rpc\rpc.h.

M

corrow

member

Activity: 97

Merit: 10

Quote from: IYFTech on January 17, 2015, 02:09:24 PM

I see you're using an Rc version: v99900 - Maybe that's what's causing your errors. I only use the stable release versions for mining.......

OMG, Why I didn't saw that myself? Embarrassed

I'm gonna deploy a stable release of Bitcoin Core to factor this out! Thanks!

IYFTech

hero member

Activity: 686

Merit: 500

WANTED: Active dev to fix & re-write p2pool in C

I see you're using an Rc version: v99900 - Maybe that's what's causing your errors. I only use the stable release versions for mining.......

Topic: [1500 TH] p2pool: Decentralized, DoS-resistant, Hop-Proof pool - page 231. (Read 2591928 times)