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

[SpAcEDeViL]

Add a Pool Node to the P2Pool Network....

e.g. On a Pool Node all miners works on one p2pool share.

And make a new option wallet-address/+P0.02 for payouts with 0.02 BTC or auto payout after one week.

BTC payout from pool wallet to miner-wallet based on the shares to the node.

So can small miners connet to the p2pools.

Without that, a miner needs over 50 GHs to make a p2pool share... thats to high for small miners....

[ksenter]

Cheers for taking the time for to write that up zvs. It helps  to learn from others with experience.

--max-outgoing doesn't seem to exist on mine, is --outgoing-conns the same thing?  Running on Windows in case it matters.

[kano]

Cheers for taking the time for to write that up zvs. It helps  to learn from others with experience.

Now if people will do it... =p

Also, you might have high latency to those three p2pools I suggested to --addnode, just look at the list on http://p2pool-nodes.info/ and find some other well connected ones if so.  I wouldn't add any w/ an average getblocktemplate time of over .3s, since those are probably overloaded as is... well, either that, or they may be experiencing this issue (as referenced above):



immediate action is recommended in such cases

... and I thought I'd add that most orphans directly after a block is solved are unavoidable, as p2pool will (usually) orphan the share that was directly before the block solve (or it may have been a few seconds after, which is the reason for it doing that... but 95% of the time it's a legit share that just gets jacked). 

Amiga FTW

[zkfq123]

Code:

btcaddress+256

Should be ok.

thanks a lot,but i dont understand,why is

Code:

btcaddress+256

and
if i have another  82g avalon i also use btcaddress+256 ?
[/quote]

it a  fool idea

when i set btcaddress+256 i got a 99% reject!!!!!!!!!!!!!!!!!!11

[zvs]

Code:

btcaddress+256

Should be ok.

thanks a lot,but i dont understand,why is

Code:

btcaddress+256

and
if i have another  82g avalon i also use btcaddress+256 ?

it a  fool idea

when i set btcaddress+256 i got a 99% reject!!!!!!!!!!!!!!!!!!11

that'd be based off your hardware then,  i thought the avalons could handle variable difficulty.  maybe it's considering sub 256 stuff as rejects

[zvs]

Cheers for taking the time for to write that up zvs. It helps  to learn from others with experience.

--max-outgoing doesn't seem to exist on mine, is --outgoing-conns the same thing?  Running on Windows in case it matters.

Whoops.  No, I was just flat out wrong.  I just looked:

    p2pool_group.add_argument('--max-conns', metavar='CONNS',
        help='maximum incoming connections (default: 40)',
        type=int, action='store', default=20, dest='p2pool_conns')
    p2pool_group.add_argument('--outgoing-conns', metavar='CONNS',
        help='outgoing connections (default: 6)',
        type=int, action='store', default=10, dest='p2pool_outgoing_conns')


--outgoing-conns is the correct thing to use.  i've always just modified the source to change the settings.. not sure where I got --max-outgoing from...  and the --p2pool-node additions don't count towards that total

i'll edit the original message

[gyverlb]

Code:

btcaddress+256

Should be ok.

thanks a lot,but i dont understand,why is

Code:

btcaddress+256

and
if i have another  82g avalon i also use btcaddress+256 ?

it a  fool idea

when i set btcaddress+256 i got a 99% reject!!!!!!!!!!!!!!!!!!11

that'd be based off your hardware then,  i thought the avalons could handle variable difficulty.  maybe it's considering sub 256 stuff as rejects

Avalons use cgminer which handles variable difficulty fine. At least I can confirm mine does (I don't set difficulty manually but p2pool adjusts difficulty automatically anyway and I don't get any suspicious reject %).
zkfq123 probably has a problem specific to his/her setup, but his/her bug report isn't really helpful (google "how to report a bug").

[gnar1ta$]

it a  fool idea

when i set btcaddress+256 i got a 99% reject!!!!!!!!!!!!!!!!!!11

256 is too high for 82GH/s.  I don't know where the double your hash rate idea came from.  Read here:

https://bitcointalk.org/index.php?topic=274023.0

Or just let p2pool do it for you 

[ksenter]

Cheers for taking the time for to write that up zvs. It helps  to learn from others with experience.

--max-outgoing doesn't seem to exist on mine, is --outgoing-conns the same thing?  Running on Windows in case it matters.

Whoops.  No, I was just flat out wrong.  I just looked:

    p2pool_group.add_argument('--max-conns', metavar='CONNS',
        help='maximum incoming connections (default: 40)',
        type=int, action='store', default=20, dest='p2pool_conns')
    p2pool_group.add_argument('--outgoing-conns', metavar='CONNS',
        help='outgoing connections (default: 6)',
        type=int, action='store', default=10, dest='p2pool_outgoing_conns')


--outgoing-conns is the correct thing to use.  i've always just modified the source to change the settings.. not sure where I got --max-outgoing from...  and the --p2pool-node additions don't count towards that total

i'll edit the original message

Awesome thanks for the clarification!

[zvs]

14iVxhdTHgog4nU6DmRzEmuXvJ2PxgqQ3w

7.8% orphans, 9.1% DOA over 375 shares on what looks like a local node, despite the doa % (not sure why else it'd be on a roadrunner cable line)

pm me if you want to fix it

1QKkhvTNVWbg3w1CFCVDUrgWJ8w5iRWaNF

8.5% orphans over 610 shares on a node in Germany (!)       (ed: there's also a secondary issue here)

pm me if you want to fix it

[phillipsjk]

tip #7: compile bitcoind with outgoing connections of 1, run bitcoind with listen=1, maxconnections=2, with one node on connect... well, that's what I do, since I can monitor it constantly.. you'd probably want to change maxconnections to 3 or 4, in case one of the nodes craps out, be it due to the Interpol, cult of the dead cow, or whatever else.  an example conf file,

While a low number of connections may save bandwidth, wouldn't it possibly have the side-effect of making the Bitcoin network less resilient?

I guess I will see how much bandwidth is used when I finally get my node up.

[zvs]

tip #7: compile bitcoind with outgoing connections of 1, run bitcoind with listen=1, maxconnections=2, with one node on connect... well, that's what I do, since I can monitor it constantly.. you'd probably want to change maxconnections to 3 or 4, in case one of the nodes craps out, be it due to the Interpol, cult of the dead cow, or whatever else.  an example conf file,

While a low number of connections may save bandwidth, wouldn't it possibly have the side-effect of making the Bitcoin network less resilient?

I guess I will see how much bandwidth is used when I finally get my node up.

yeah, but if you are trying to run a node locally and are bandwidth starved, that's something you should do.  

otherwise you'll either randomly connect to an outgoing peer that is missing part of the block chain or have one connect to you.  then your upstream is saturated

i think it's not so bad in other parts of the world, but in the US, you really get gimped on upstream

[jedimstr]

tip #7: compile bitcoind with outgoing connections of 1, run bitcoind with listen=1, maxconnections=2, with one node on connect... well, that's what I do, since I can monitor it constantly.. you'd probably want to change maxconnections to 3 or 4, in case one of the nodes craps out, be it due to the Interpol, cult of the dead cow, or whatever else.  an example conf file,

While a low number of connections may save bandwidth, wouldn't it possibly have the side-effect of making the Bitcoin network less resilient?

I guess I will see how much bandwidth is used when I finally get my node up.

yeah, but if you are trying to run a node locally and are bandwidth starved, that's something you should do.  

otherwise you'll either randomly connect to an outgoing peer that is missing part of the block chain or have one connect to you.  then your upstream is saturated

i think it's not so bad in other parts of the world, but in the US, you really get gimped on upstream

Not all of us... I've got 75down/35up rated (typically 86down/39up actual).  FiOS kicks a$$.

Running a local node at galactica.geekgalaxy.com:9332  
Any tips?

[maqifrnswa]

@zvs: When I'm mining on your node, and it goes down, my cgminer client totally locks up and stops working. Other nodes, however, don't have that problem and cgminer properly switches to failover nodes. Has anyone else seen this on your node? I don't remember the error it gave; I stopped mining on it after it happened a few times (3 times last week).

[matthewh3]

@zvs: When I'm mining on your node, and it goes down, my cgminer client totally locks up and stops working. Other nodes, however, don't have that problem and cgminer properly switches to failover nodes. Has anyone else seen this on your node? I don't remember the error it gave; I stopped mining on it after it happened a few times (3 times last week).

This keeps happening to us on OGnasty's node and we've had to take 1.1TH off P2Pool because of this plus we'll have 650GH/s Jupiter soon to hash on as well.  So that's 1.75TH/s for P2Pool if anyone can help us.  Does anyone have a list of the nodes you mention that enable a fallover?

Edit:  Although my friend hosting the miners says it's cgminer actually crashing and is a cgminer problem.

[maqifrnswa]

@zvs: When I'm mining on your node, and it goes down, my cgminer client totally locks up and stops working. Other nodes, however, don't have that problem and cgminer properly switches to failover nodes. Has anyone else seen this on your node? I don't remember the error it gave; I stopped mining on it after it happened a few times (3 times last week).

This keeps happening to us on OGnasty's node and we've had to take 1.1TH off P2Pool because of this plus we'll have 650GH/s Jupiter soon to hash on as well.  So that's 1.75TH/s for P2Pool if anyone can help us.  Does anyone have a list of the nodes you mention that enable a fallover?

Edit:  Although my friend hosting the miners says it's cgminer actually crashing and is a cgminer problem.

it is a cgminer problem, I think I found some info here:
https://bitcointalk.org/index.php?topic=188533.0

funny thing is that OgNasty's node failsover fine for me! I have much smaller hashrate, maybe that has something to do with it too.

[matthewh3]

@zvs: When I'm mining on your node, and it goes down, my cgminer client totally locks up and stops working. Other nodes, however, don't have that problem and cgminer properly switches to failover nodes. Has anyone else seen this on your node? I don't remember the error it gave; I stopped mining on it after it happened a few times (3 times last week).

This keeps happening to us on OGnasty's node and we've had to take 1.1TH off P2Pool because of this plus we'll have 650GH/s Jupiter soon to hash on as well.  So that's 1.75TH/s for P2Pool if anyone can help us.  Does anyone have a list of the nodes you mention that enable a fallover?

Edit:  Although my friend hosting the miners says it's cgminer actually crashing and is a cgminer problem.

it is a cgminer problem, I think I found some info here:
https://bitcointalk.org/index.php?topic=188533.0

funny thing is that OgNasty's node failsover fine for me! I have much smaller hashrate, maybe that has something to do with it too.

KNC have released a new firmware update today.  If anyone finds out this stops the cgminer P2Pool for KNC Jupiters bug let me know and the pool will gain 1.75TH.

[zvs]

@zvs: When I'm mining on your node, and it goes down, my cgminer client totally locks up and stops working. Other nodes, however, don't have that problem and cgminer properly switches to failover nodes. Has anyone else seen this on your node? I don't remember the error it gave; I stopped mining on it after it happened a few times (3 times last week).

I think this is caused by the stratum interrupts when the p2pool doesn't find a new share for 90 seconds?

It says something about it when I have my video card pointed there too, stops for about 1/2 a second also, but i'm using some .. 6 month old version of cgminer

maybe it works properly now, i dunno

but that's when it happens.  when there is no new block or no new share for 90 seconds

ed: it should be common to any p2pool?  my video card starts back up right away on cgminer 3.3.4, it just says stratum connection was interrupted and then restarts

[zvs]

tip #7: compile bitcoind with outgoing connections of 1, run bitcoind with listen=1, maxconnections=2, with one node on connect... well, that's what I do, since I can monitor it constantly.. you'd probably want to change maxconnections to 3 or 4, in case one of the nodes craps out, be it due to the Interpol, cult of the dead cow, or whatever else.  an example conf file,

While a low number of connections may save bandwidth, wouldn't it possibly have the side-effect of making the Bitcoin network less resilient?

I guess I will see how much bandwidth is used when I finally get my node up.

yeah, but if you are trying to run a node locally and are bandwidth starved, that's something you should do.  

otherwise you'll either randomly connect to an outgoing peer that is missing part of the block chain or have one connect to you.  then your upstream is saturated

i think it's not so bad in other parts of the world, but in the US, you really get gimped on upstream

Not all of us... I've got 75down/35up rated (typically 86down/39up actual).  FiOS kicks a$$.

Running a local node at galactica.geekgalaxy.com:9332  
Any tips?

no, i like your outgoing connections =p

[gnar1ta$]

@zvs: When I'm mining on your node, and it goes down, my cgminer client totally locks up and stops working. Other nodes, however, don't have that problem and cgminer properly switches to failover nodes. Has anyone else seen this on your node? I don't remember the error it gave; I stopped mining on it after it happened a few times (3 times last week).

This keeps happening to us on OGnasty's node and we've had to take 1.1TH off P2Pool because of this plus we'll have 650GH/s Jupiter soon to hash on as well.  So that's 1.75TH/s for P2Pool if anyone can help us.  Does anyone have a list of the nodes you mention that enable a fallover?

Edit:  Although my friend hosting the miners says it's cgminer actually crashing and is a cgminer problem.

it is a cgminer problem, I think I found some info here:
https://bitcointalk.org/index.php?topic=188533.0

funny thing is that OgNasty's node failsover fine for me! I have much smaller hashrate, maybe that has something to do with it too.

KNC have released a new firmware update today.  If anyone finds out this stops the cgminer P2Pool for KNC Jupiters bug let me know and the pool will gain 1.75TH.

What flags are you using for cgminer?  I have had a dozen types of miners including a Jupiter on my node and they always fail over fine to my backup pool when I take the node down for maintenance. failover-only is the only thing I add to my conf for ASICs.