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

[jtoomim]

Anyone, tell, please when it will be compiled on Windows.

That is a good question for jtoomim

Bugs first, binaries later.

[1713577437]

1713577437

[1713577437]

1713577437

[flameruk]

My NTP is working any ideas?

Try setting the clock in BIOS and system UTC + 0

I don't like that this is necessary. Flameruk, can you tell me what operating system you were using, what time zone you had your BIOS set to, and what time zone your OS was set to? I think I need to change the code that I use in p2pool's code when determining UTC time, and I'd like to be able to reproduce your issue.

Hi Jtoomim.

The Bios was almost surely set at current time and GMT 0, London.
I generally set this when provisioning a server.
I could verify but need to get to my offices as its racked up in the server room.
If you really need to know ill down the box tomorrow, get a console on it and find out for you. Its no issue really.

Im running Centos 7 without a GUI.
I had the timezone as GMT , we are currently in British Summer Time so we ride +1 in front.
I used timedatectl and changed the time zone to UTC then the errors went away immediately.

Ive also noticed after running for a few hours now :-

1) Without tempting fate, my number of Orphans and Dead Shares is way lower.
2) My DOA is about the same.
3) My traffic rate graph seems to be not so volatile and kind of peaks smoothly and not "spikey" if you get where im coming from.
4) My BGBT Latency is lower and seems smoother.
5) Im now using 1.2M of RAM, where the previous node (other fork) would NEVER use more than 800 Mb. Ive made no changes to Centos or the Hardware.

So far so good. Hope the info is useful.
Now we just need some blocks.

[flameruk]

Code:

2017-05-11 15:16:06.170906 Generating a share with 748972 bytes (2821 new) and 1701 transactions (12 new)

You should probably also add blockmaxsize=999000 to bitcoin.conf.
[/quote]

OK, with the old fork my server was never happy running for 1Mb blocks.
Ill adjust this now seeing as everything dare I say seems "more stable" for me at least.
With the BGBT Latency down, it might handle it better now.

[-ck]

May I suggest that if forrestv does not approve of jtoomim's fork and include the code in the main repository that a new thread be created for the alternative fork out of respect for the original thread and pool? Alternatively, if forrestv is happy for the jtoomim fork to be the main pool and take over management of the p2pool concept entirely this should also run in its own thread such that jtoomim can be the opening poster where announces/changes can be documented.

No comment on this? You are hijacking forrestv's thread and forking his pool away from him...

[flameruk]

OK, Im stable enough now to connect on the network.

Node IP: ukp2pool.uk (default ports 9332)
Connected to: ml.toom.im:9334
To connect to me: run_p2pool.py -n ukp2pool.uk:9332
CPU: 2 X 2.1 GHz Opteron (2373) Quad Core
Pipe: Leased Line 10/10
Using: Python 2.7.5

[Huckster]

Diff: 560G
It says my BFGMiner. 

[sawa]

Code:

...
ImportError: No module named zope.interface

I fixed and checked, download again https://github.com/ilsawa/p2pool-ltc/releases/download/16.0/p2pool_win32_4c59633.zip

[jtoomim]

No comment on this? You are hijacking forrestv's thread and forking his pool away from him...

As you may have noticed, forrestv isn't very active in this community any more. As such, I have been treating this more as a p2pool community thread rather than forrestv's thread. If he views it differently, then I will defer to that.

The branch that I made is still mostly his code, and still defaults to including a 0.5% donation to him. I'm not trying to steal p2pool away from him; I'm just trying to keep the project alive even though the official maintainer and sole github committer is inactive.

[-ck]

No comment on this? You are hijacking forrestv's thread and forking his pool away from him...

As you may have noticed, forrestv isn't very active in this community any more. As such, I have been treating this more as a p2pool community thread rather than forrestv's thread. If he views it differently, then I will defer to that.

The branch that I made is still mostly his code, and still defaults to including a 0.5% donation to him. I'm not trying to steal p2pool away from him; I'm just trying to keep the project alive even though the official maintainer and sole github committer is inactive.

I'm sure you would find your own thread far more flexible and useful anyway, but I'm not making any calls on moderating this unless forrestv does show up and declare his wishes.

[jtoomim]

OK, with the old fork my server was never happy running for 1Mb blocks.

This commit (https://github.com/jtoomim/p2pool/commit/e71535e4401d0e0b6d19b2c0741b1fff2d9770a1) significantly reduces the CPU load from p2pool dealing with large numbers of transactions. CPU load vs transaction volume was O(n^2) in the old version of p2pool, but should be O(n) in my branch.

[forrestv]

I don't see anything wrong with jtoomim's posts here.

With regard to development efforts, jtoomim: I'll merge any pull requests from you from now on that pass a quick review (I don't have means to test currently). If you want, we can discuss giving you push access to https://github.com/p2pool/p2pool.

ck-'s point about making a new thread may make sense if jtoomim takes over and creates new releases, as he'll be unable to directly edit the first post of this thread that links to them.

[jtoomim]

With regard to development efforts, jtoomim: I'll merge any pull requests from you from now on that pass a quick review (I don't have means to test currently). If you want, we can discuss giving you push access to https://github.com/p2pool/p2pool.

Sounds good. 1mb_hardforked isn't quite ready for master yet, as it hasn't been tested with any altcoins -- this commit (https://github.com/jtoomim/p2pool/commit/fe9179e4a81543aadfcc840c0697df0aa475748f) in particular worries me -- but once I've tested it at least with one altcoin and have investigated the timezone issue a bit better, I'll put in a PR.

[windpath]

With regard to development efforts, jtoomim: I'll merge any pull requests from you from now on that pass a quick review (I don't have means to test currently). If you want, we can discuss giving you push access to https://github.com/p2pool/p2pool.

Sounds good. 1mb_hardforked isn't quite ready for master yet, as it hasn't been tested with any altcoins -- this commit (https://github.com/jtoomim/p2pool/commit/fe9179e4a81543aadfcc840c0697df0aa475748f) in particular worries me -- but once I've tested it at least with one altcoin and have investigated the timezone issue a bit better, I'll put in a PR.

Great! Thank you both.

[Huckster]

Code:

...
ImportError: No module named zope.interface

I fixed and checked, download again https://github.com/ilsawa/p2pool-ltc/releases/download/16.0/p2pool_win32_4c59633.zip

Works fine! Thanks a lot, sir!

UPD: But... not

http://huckster.ru/err.png

And this ignore this divisor: /2+0.000244

[sawa]

Code:

...
ImportError: No module named zope.interface

I fixed and checked, download again https://github.com/ilsawa/p2pool-ltc/releases/download/16.0/p2pool_win32_4c59633.zip

Works fine! Thanks a lot, sir!

UPD: But... not

A little longer dig, 2-3 hours. Then look at what the value will show the difficulty of the pool.

And this ignore this divisor: /2+0.000244

https://bitcointalk.org/index.php?topic=18313.msg18881770#msg18881770

[Huckster]

Probably because the pool will not give shares with difficulty less than pool.hws.ru:9327/difficulty

For testing purposes, i found GC3355, people called this a "hedgehog"

Same settings, same device, pool is different:

http://huckster.ru/good.png

[sawa]

Probably because the pool will not give shares with difficulty less than pool.hws.ru:9327/difficulty

For testing purposes, i found GC3355, people called this a "hedgehog"

Same settings, same device, pool is different:

I understand, I edited my message. Read it.

[Dropwise]

Hello,
I successfully got jtoomim's fork up and running with the below instructions.
However, it appears it is running Version: 15.0-43-gcc6fd7d rather than Version: 16.0-7-gb0d472e-dirty at ml.toom.im:9332

Can anyone please clarify:
1)  is this correct?
2)  is it possible to upgrade to 16?   if so, can you please provide instructions?

Many thanks,
d

https://github.com/jtoomim/p2pool/tree/1mb_hardforked has new code. This should fix the sync issues that people have been having. It also includes a few bootstrap nodes to make it easier for people to connect to the jtoomimnet p2pool.

To set up and run my fork, you will need to do the regular steps for installing p2pool, except that instead of getting the regular github.com/p2pool/p2pool repository, you will do:

Code:

git clone https://github.com/jtoomim/p2pool
cd p2pool
git checkout 1mb_hardforked

If anyone continues to have trouble connecting to jtoomimnet, please let me know.

[jtoomim]

I successfully got jtoomim's fork up and running with the below instructions.
However, it appears it is running Version: 15.0-43-gcc6fd7d rather than Version: 16.0-7-gb0d472e-dirty at ml.toom.im:9332

Ignore that. The version that p2pool advertises is not based on the code itself, but instead on the git history of the branch that you're logged into. Specifically, p2pool runs "git describe --dirty --always" in order to determine its version number. git describe will look for the most recent tag that is reachable from a commit. As I haven't tagged any new versions yet (though I totally should have), git describe doesn't say anything helpful. For some unknown reason, it shows up as 16.0-7-* on my nodes but 15.0-43-* on the nodes that other people have made from cloned repositories, even though the code is exactly the same.

If you run "git log", that will list the most recent changes that were made on the branch you're using. As of today, the most recent commit should be:

commit cc6fd7d227f869aa02cdf9f89e8349aa9d2a4344
Author: Jonathan Toomim
Date:   Wed May 10 06:11:49 2017 -0700

    Don't attempt retransmission of shares when transactions are missing (avoid hang on sync)

[jtoomim]

2017-05-10 20:50:23.585000 >   File "C:\Users\ppif\Documents\GitHub\p2pool\p2pool\data.py", line 323, in check
2017-05-10 20:50:23.585000 >     if self.timestamp > int(time.mktime(time.gmtime()) - time.mktime(time.gmtime(0))) + 300:
2017-05-10 20:50:23.585000 > exceptions.OverflowError: mktime argument out of range

I did some testing, and I was unable to reproduce this error. Is it still occurring? If so, do you think you might be able to give me remote access to this machine (e.g. via showmypc.com or teamviewer) so I could do some debugging and testing?