[ANN][YAC][POOL]p2pool for yacoin!!

[eule]

Can anybody built an windows binary version for the latest e294b72? the " c265896-dirty" seems to work wrong

Since my node and most others run on that version, what is wrong with it?

[microxp]

Can anybody built an windows binary version for the latest e294b72? the " c265896-dirty" seems to work wrong

Since my node and most others run on that version, what is wrong with it?

you are using the python or windows binary version?

[cryptrol]

UPDATE8:
I have made a few changes, If you want to get involved:
1.download the latest source from github;
2.extract the zip ball to a new dir;
3.stop the old version;
4.start the new one as soon as possible, you can use the "run.sh" script provided;
5.go back to your website to test if it works well.

Notice:
1.I'm not 100% sure that the debug version will work, so it's just a test.
2.If it doesn't work, just stop the debug version, and roll back to the old one, miners connected to your server will reconnect  in 30 seconds.
3.If your server is busy as a result of changing version( got a lot of dead shares), restart the p2pool.

I installed HEAD version from github and so far so good.

There is one thing I would like to know, what are the values under Stats > Payouts ? The amount payed for the last block found ?
Also what are the reference values I should be aiming for ? Pool efficiency and number of DOA 's ?

Here are my numbers so far :

[rav3n_pl]

Can anybody built an windows binary version for the latest e294b72? the " c265896-dirty" seems to work wrong

Done win32 version, should be ok now.

[cryptrol]

For those who wonder if it will suffice with a 512Mb RAM VPS to run this, it will not.
Could be because of some memory leaks since p2pool starts and runs for some hours, but it eventually gets killed by the OS (running on debian).

If someone has some tricks or suggestions on how to run this with lower resources, that would be great.

[caish5]

I have mine on a 512mb VPS.
Admittedly it has chewed up all the ram half the swap.
But it has been running fine.
It's on Ubuntu 12.04 if that helps.
I didn't do anything special when i set it up.

[cryptrol]

And you don't have stability issues ?
Well, I guess I will have to try Ubuntu then.
Thanks for the feedback.

[cryptrol]

One more thing, are you using the yacoind on the same VPS ? It also eats some ram.

[rav3n_pl]

For those who wonder if it will suffice with a 512Mb RAM VPS to run this, it will not.
Could be because of some memory leaks since p2pool starts and runs for some hours, but it eventually gets killed by the OS (running on debian).

If someone has some tricks or suggestions on how to run this with lower resources, that would be great.

Use 32bit system.

[eule]

Also on a 512 mb VPS, running debian 64 bit, no problems so far. Memory usage IS very high though, thanks for the warning, if it crashes i'll try 32 bit.

edit: Most workers have nice and low DOA, some have pretty high, guess it depends on latency?

[caish5]

One more thing, are you using the yacoind on the same VPS ? It also eats some ram.

Yes they are both on there.

[caish5]

Also on a 512 mb VPS, running debian 64 bit, no problems so far. Memory usage IS very high though, thanks for the warning, if it crashes i'll try 32 bit.

edit: Most workers have nice and low DOA, some have pretty high, guess it depends on latency?

Yeah latency is the main factor I think.

[procrypto]

I've got this on a 2GB VPS at the moment, memory use is 750+MB and still steadily rising.

edit: Most workers have nice and low DOA, some have pretty high, guess it depends on latency?

This guide explains p2pool latency issues pretty well. https://bitcointalk.org/index.php?topic=153232.0

[rav3n_pl]

I've got this on a 2GB VPS at the moment, memory use is 750+MB and still steadily rising.

Restart p2pool. You step into memory leak.
My node eats up to 260MB after 24hrs.

[procrypto]

Hmm ok.. I have restarted the pool, and wiped the data directory for good measure. Has caused an insane amount of dead shares and 99% CPU load, sorry about that! Let's see what happens once it straightens itself out..

edit: seems to be better again now, efficiency will take a little while to recover though   Memory use back on the rise too, into mid 300MB already.

Thanks to whoever donated 50 YAC to the server fund too, really do appreciate it.  

[procrypto]

Someone from the CPUminer thread has an issue which I'm trying to get to the bottom of. It seemed to make more sense to continue the discussion here. See below for the issue.

Is anyone mining on http://yac.procrypto.com:8336 ?

I've been doing that all day, and yet no payout of any kind. My target address is not listed on the payout page either.

Is that normal with p2pool type mining pools? What should I expect?

-Michael

I checked the logs, and the address keeps cropping up with this error message..

Code:

2013-05-30 20:13:08.190739 Worker Y8iGmHkz1FBv8yzB3Cd227a6n5Hu6MxTZ6 submitted share with hash > target:
2013-05-30 20:13:08.190897     Hash:   56a339458c6793155260403e8cda63a00212eb09b5e55e2c20605a26804f372a
2013-05-30 20:13:08.190967     Target: 5a698417da16a4000000000000000000000000000000000000000000000

Are these orphaned shares? Is it possible that he is getting literally 100% orphans?

Is this potentially a latency issue?

Also, I've seen recommendations in other P2Pool discussions for appending a share difficulty (e.g. +0.0025) to the worker address/username. Is that something that could help here? There is a list I've seen with recommendations for different hashrates, but I'm not sure how useful it is. Could this be a low hashrate issue? I've PMd the user and asked him to check into the thread and let us know what hashrate he is getting.

[forsetifox]

The pool died! 

I suspect it has something to do with the no payouts in the last 5 hours.

 

[mberg2007]

Are these orphaned shares? Is it possible that he is getting literally 100% orphans?

I noticed the server was restarted a few hours ago. I'm still mining with the same address - perhaps you could check the logs and see if the problem still exists?

From your analysis it looks like I am submitting work that is simply incorrectly computed. Could this be a problem with the miner I am using? Overall I'm seeing a 70-75% acceptance rate, so I wasn't expecting there would be a problem of that kind. I'm hashing from multiple machines with an average rate of 6-700 kh/s per machine, so I'm not dragging my feet computing wise. Also there should be no network lagging issues worth speaking of.

-Michael

PS ... why is my target and hash not of the same length? I kind of expected that, otherwise it makes little sense to compare them?

[schnebi]

I have mine on a 512mb VPS.
Admittedly it has chewed up all the ram half the swap.
But it has been running fine.
It's on Ubuntu 12.04 if that helps.
I didn't do anything special when i set it up.

I had a look, at how much a VPS costs in the USA, and i was pretty shocked. In Germany you get a VPS with 2 Cores (4 GHZ), 2 GB RAM, 150 MB HD an a 100 MBIT Connection for 9.99€ per month, what is approximately 13 $

I searched for a VPS in the United States, but for the same price I just found VPS with a maximum of 512 MB RAM

[mberg2007]

I searched for a VPS in the United States, but for the same price I just found VPS with a maximum of 512 MB RAM

I've been using these guys for some vps testing.

Very fast VPS, works like a charm. The VNC console provides a virtual console view of the server if you need to "ctrl-alt-delete" it or watch the boot messages.

Only drawback is they don't like CPU hogs and will stop any process that runs at 100% after 24 hours or so. So they might not be suitable if mining is your purpose, but for all other projects they are cheap, they are fast and they answer their support mails in a timely manner. I recommend them.

-Michael