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

[rav3n_pl]

Limit connections to 30 or 20:
--max-conns 30

But there were about only 20 connections on his chart (10 outgoing and 10 incoming).

It's got to be something else.

Ahh, missed numbers. Maybe it is some OS limitation?

[zvs]

i encountered the bizarre issue with all the DOA shares again after a peer joined:

I've observed this twice now as well, but I never figured out that it was related to a peer joining.

The first time, it caused one my cgminer instances to go beserk and submit endless stales.
Last time, just this morning, my phoenix instance did it.  
Like you, restarting the miners and p2pool fixed it.

p2pool bug or attack?

M

Since it's the second time I've seen it and both times it occurred directly after a peer joined, I was thinking something along the lines of the peer passing along 'wrong' information.  I don't think it was a deliberate attack, though.... since I imagine if that were to be the case, it'd happen far more often (twice in about 2 weeks so far for me -- I've got another screenshot posted farther up this thread, I believe).

I'm 99% sure it doesn't have to do with the # of connections, since at the time I was running with maxconn 900 in bitcoind, and I've had that up to >1000 before.

found it (was posted in the other p2pool thread):  https://bitcointalk.org/index.php?topic=66182.msg992288#msg992288

both times after a peer disconnected then it (or a different one) connected shortly after

p2pool bug or attack?

i'm thinking bug, since like up above, i'd think it'd occur more often if it was deliberate

although I suppose someone could eventually find out how to 'make' it occur deliberately

[mdude77]

Since it's the second time I've seen it and both times it occurred directly after a peer joined, I was thinking something along the lines of the peer passing along 'wrong' information.  I don't think it was a deliberate attack, though.... since I imagine if that were to be the case, it'd happen far more often (twice in about 2 weeks so far for me -- I've got another screenshot posted farther up this thread, I believe).

I'm 99% sure it doesn't have to do with the # of connections, since at the time I was running with maxconn 900 in bitcoind, and I've had that up to >1000 before.

found it (was posted in the other p2pool thread):  https://bitcointalk.org/index.php?topic=66182.msg992288#msg992288

both times after a peer disconnected then it (or a different one) connected shortly after

p2pool bug or attack?

i'm thinking bug, since like up above, i'd think it'd occur more often if it was deliberate

although I suppose someone could eventually find out how to 'make' it occur deliberately

Aren't there only 2 reasons why a peer would pass the 'wrong' info?  A bug, or an attack? 

I have 3 miners.  Why when it happens does it only affect one miner?  And why wouldn't restarting that miner fix it, instead I have to restart p2pool?

Very perplexing.

M

[twmz]

I have had an annoying hardware failure of one of the machines that provides data for p2pool.info behind the scenes.  Until I can rebuild this machine, it means p2pool.info may be intermittently out of action or inaccurate.

[zvs]

blocks solved by:

1PoNvnRhbvFE4xebEXjcVWfA3f2CGQkrCw
1MUfvW9osTXZgZdVSXyHSj7E4gousvnDyF
1PoNvnRhbvFE4xebEXjcVWfA3f2CGQkrCw  (yes, two)
12AEGVu3PNJ37fVbMBnhL6mCv1YmTjF74z

huzzah

[rav3n_pl]

blocks solved by:

1PoNvnRhbvFE4xebEXjcVWfA3f2CGQkrCw
1MUfvW9osTXZgZdVSXyHSj7E4gousvnDyF
1PoNvnRhbvFE4xebEXjcVWfA3f2CGQkrCw  (yes, two)
12AEGVu3PNJ37fVbMBnhL6mCv1YmTjF74z

huzzah

And 30days finally reach 100%

[twmz]

I have had an annoying hardware failure of one of the machines that provides data for p2pool.info behind the scenes.  Until I can rebuild this machine, it means p2pool.info may be intermittently out of action or inaccurate.

It's fixed.  

And no comment...

[check_status]

I have had an annoying hardware failure of one of the machines that provides data for p2pool.info behind the scenes.  Until I can rebuild this machine, it means p2pool.info may be intermittently out of action or inaccurate.

Your not the only one to have a hardware failure related to p2pool.

Bitpop --Had a motherboard problem.
twmz --What was your issue?
JayCoin --Attacker formatted HDD and stole ⊅ from his p2pool subpool.
check_status (me) --Was connected to Jaycoins pool, after the attack my motherboard wouldn't POST with the HDD connected. Removed the HDD and the motherboard POSTs fine.

The fact that p2pool.info is experiencing an issue, makes it difficult to prove how profitable p2pool is compared to other pools. How convenient.

[TravisE]

JayCoin --Attacker formatted HDD and stole ⊅ from his p2pool subpool.

That's not a hardware failure.

[twmz]

I have had an annoying hardware failure of one of the machines that provides data for p2pool.info behind the scenes.  Until I can rebuild this machine, it means p2pool.info may be intermittently out of action or inaccurate.

Your not the only one to have a hardware failure related to p2pool.

My hardware issue had nothing to do with p2pool.  The motherboard and/or CPU of the host machine for my home web server's VM died.  I haven't bothered to diagnose which failed--I just built a new host computer and moved the VM there. 

This was a Windows Server 2008 R2 virtual machine that runs the web server and SQL server I use at home for managing my farm of mining rigs.  It was also a convenient machine to run my internal caching DNS server on which meant when it died, none of my computers could resolve hostnames anymore.  That is why p2pool.info was slightly affected (my p2pool nodes could not communicate to p2pool.info to send it data because they couldn't resolve the hostname).  Neither p2pool nor bitcoin were running on this machine and could not have been associated with the hardware failure, so you'll have to look elsewhere for your conspiracy theories.

The fact that p2pool.info is experiencing an issue, makes it difficult to prove how profitable p2pool is compared to other pools. How convenient.

I knew there was a reason I ignored you.  I should have trusted my instinct and not unignored you to see what you were asking.  Fixed.

[check_status]

I'm trying to start a conspiracy that doesn't involve Zhou Tong, could you at least understand even if you don't unignore me.

[jackmaninov]

Now that our 90 day luck is above 100% (knock on wood), I propose we go on a p2pool membership drive.

I'd be willing to donate some coins through the pool in say 30 days if we could reach some sort of goal (increased # users? 500 GH/s?). I can't really decide on how the goal should be framed though, since it would be easy for a big miner to just split his hashing power across different addresses to simulate more users.

What do people think?

[randomguy7]

The only thing stopping me from mining on p2pool (except of some testing) is the constant bad luck. If that's solved and the 90 days luck factor keeps swinging around 100%, I'd love to switch to p2pool as p2pool basically is quite awesome .
I'm only a small miner but I think many people think the same way.

[Krak]

The only thing stopping me from mining on p2pool (except of some testing) is the constant bad luck. If that's solved and the 90 days luck factor keeps swinging around 100%, I'd love to switch to p2pool as p2pool basically is quite awesome .

Seconded. Only problem is that as the hashrate rises, it seems like the luck gets worse.

[RandomQ]

I have had an annoying hardware failure of one of the machines that provides data for p2pool.info behind the scenes.  Until I can rebuild this machine, it means p2pool.info may be intermittently out of action or inaccurate.

It's fixed.  

And no comment...

That is very nice first time I've seen the 90 day over 100% 

Thanks for Great Week P2Pool

[Ente]

Only problem is that as the hashrate rises, it seems like the luck gets worse.

Some changes have been made to p2pool recently. Hope is that this cured the luck for good!

Ente

[Krak]

Some changes have been made to p2pool recently. Hope is that this cured the luck for good!

Ente

Ahh, good to hear. I haven't been following the development very closely lately; wasn't getting email notifications for this topic for some reason. I may have to give it a try again soon.

[zvs]

The only thing stopping me from mining on p2pool (except of some testing) is the constant bad luck. If that's solved and the 90 days luck factor keeps swinging around 100%, I'd love to switch to p2pool as p2pool basically is quite awesome .

Seconded. Only problem is that as the hashrate rises, it seems like the luck gets worse.

well, as the hashrate rises, i know i get more stale/dead shares, because of all the new work messages.

i wish I could add some difficulty to my miners, but the /1000 or whatever doesn't work in phoenix

[louisBSAS]

i wish I could add some difficulty to my miners, but the /1000 or whatever doesn't work in phoenix

I was wondering who else is out there using Phoenix and what command line options are you using for best results on p2pool. Does worksize have anything to do with the 'difficulty' issue?

I have played with setting poclbm, phatk and phatk2 - the phatk kernel seems to give me the best hashrate.

I have tried with/without vectors and fastloop plus higher and lower Aggression - and given each change a couple days to see how it plays on stales, DOA's, etc.

Anyone want to share 'best configuration' for ATI cards?

* These two give me the best results so far.

... seems to increase local hash rate but payout drops over time - dead on arrival: between 1.9 - 4.0%
-k phatk DEVICE=0 WORKSIZE=128 VECTORS BFI_INT FASTLOOP AGGRESSION=9

... seems to generate a higher payout over time but slows reported hash rate.
-k phatk DEVICE=0 WORKSIZE=256 VECTORS BFI_INT FASTLOOP AGGRESSION=9


I am running 3 boxes with a mix of 6870's 6990's and 7970's - phoenix 1.7 - Ubuntu 10.04lts

[Frizz23]

I have had an annoying hardware failure of one of the machines that provides data for p2pool.info behind the scenes.  Until I can rebuild this machine, it means p2pool.info may be intermittently out of action or inaccurate.

It's fixed.  

And no comment...

That is very nice first time I've seen the 90 day over 100% 

Thanks for Great Week P2Pool

Did they change the algorithm to calculate the luck charts on p2pool.info?