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

[lenny_]

Hello guys.
I am still waiting with my Block Erupter Blades for some developer willing to improve these bugs (stratum proxy incompatible with p2pool and p2pool itself incompatible with BE Blade). I can donate machine worktime to developer.
Can someone is interested in debugging this?

https://bitcointalk.org/index.php?topic=18313.msg2166295#msg2166295

[Prattler]

FWIW, adding this to the bottom of my bitcoind.conf fixed the massive recent latency issues.  Now, I'm not sure if this is the best-for-the-network settings, but it certainly fixed my latency.   I increased the relay fee minimum some, and I think most importantly I doubled the time it takes for txns to become high priority.  Of course, that part is only a temporary fix...

EDIT: This only helped for a couple hours....

blockprioritysize=500000
mintxfee     =0.0002
minrelaytxfee=0.0002

You should lower blockprioritysize and increase mintxfee/minrelaytxfee a bit more. Blockprioritysize=500000 is way too much, you are allocating 500 kB to free transactions!

I suggest you do

Code:

blockprioritysize=50000
mintxfee     =0.0005
minrelaytxfee=0.0005

[gyverlb]

Naive?  My little world?  As that where we're at now - posting personal insults publicly?

I'm fed up with trolls and people ignoring reality on some recurrent subjects like bitcoind latency problems. I think you are naive and you can't get the big picture, if you are easily offended by my opinion and think it's a personal attack that's your problem and you are free to ignore me and my opinions. Don't expect me to ignore the misinformation you help spread though.

As for stratum problems, I don't have an opinion right now and IIRC my posts on the subject repeatedly asked for data to find out a pattern and didn't reject problem reports as PEBKAC out of hand like you seem to imply.
I don't have any problem with stratum even with hashrates equivalent or above some people having problems so if there is a bug it's in some configurations not all. If people don't report bugs with enough context for the devs to start investigating, they should not complain that their problem isn't fixed (especially when this work is done for free).

[PatMan]

Naive?  My little world?  As that where we're at now - posting personal insults publicly?

I'm fed up with trolls and people ignoring reality on some recurrent subjects like bitcoind latency problems. I think you are naive and you can't get the big picture, if you are easily offended by my opinion and think it's a personal attack that's your problem and you are free to ignore me and my opinions. Don't expect me to ignore the misinformation you help spread though.

As for stratum problems, I don't have an opinion right now and IIRC my posts on the subject repeatedly asked for data to find out a pattern and didn't reject problem reports as PEBKAC out of hand like you seem to imply.
I don't have any problem with stratum even with hashrates equivalent or above some people having problems so if there is a bug it's in some configurations not all. If people don't report bugs with enough context for the devs to start investigating, they should not complain that their problem isn't fixed (especially when this work is done for free).

Trolls........lol.

[centove]

You know, all this bickering back and forth about the tx stuff don't make a lot of sense to me. If it wasn't an issue why did the official client add the option in? Clearly there must be some motivation behind it.

When I looked:
Total Fees   1.36079892 BTC
Total Size   12163.966796875 (KB)

Then there are the ones that say your hardware is crap. Umm.. just what should a full node be running on?
Currently I have a Dual Xeon server with 16 cores and 32 Gig of memory and SAS 10k drives in it and a Gig connection to the internet. And I was getting latencies over 10 seconds. How much more power do I need to throw at BTC???

[crunchy]

i have seen enough , nice steal software this p2pool thing.

its to fucking easy to steal other miners money.

the whole setup of p2pool just sucks.

its designed not to make you money or at least as little as possible.

never install python with the windows executable either , its an easy way to hack your pc.

i've watched this p2pool software enough, how many f*** don't like to edit the source files to their liking?


months of p2pool got me 15 ltc, 4 days solos and i allready have a 50ltc block found.


Keep your shitty hacktool.

[gyverlb]

months of p2pool got me 15 ltc, 4 days solos and i allready have a 50ltc block found.

Keep your shitty hacktool.

So you don't understand shit about mining and you risk being sued for libel by people you don't know anything about. Nice move.

[zvs]

i have seen enough , nice steal software this p2pool thing.

its to fucking easy to steal other miners money.

the whole setup of p2pool just sucks.

its designed not to make you money or at least as little as possible.

never install python with the windows executable either , its an easy way to hack your pc.

i've watched this p2pool software enough, how many f*** don't like to edit the source files to their liking?


months of p2pool got me 15 ltc, 4 days solos and i allready have a 50ltc block found.


Keep your shitty hacktool.

yeah, i know

i got 1.2 bitcoins out of bitminter even though i found a 25 btc block

what the hell is that

you dirty ratbastards

[fehknt]

You should lower blockprioritysize and increase mintxfee/minrelaytxfee a bit more. Blockprioritysize=500000 is way too much, you are allocating 500 kB to free transactions!

Oops.  I misunderstood what blockprioritysize meant, thanks!

[maqifrnswa]

Naive?  My little world?  As that where we're at now - posting personal insults publicly?

Coming from the same person who has constantly denied that there is a problem with the stratum code implementation and still does, even though it's proven - that's quite rich, don't you think? It's a bit sad to dislike someone because they were right, but there's absolutely no excuse to resort to insulting them, weather they are correct or not.

You were not right, in fact you were very wrong. You need to understand why you're not getting anywhere and people have lost patience with you:

Order of events
1) We had a run of long blocks one month ago around the same time the stratum protocol was implemented in p2pool.
2) You said, "it's impossible to be this unlucky, it must be stratum since that was the most recent change. FIX STRATUM"
3) stratum has absolutely nothing to do with the rate of finding blocks per work. If stratum was 99% broken (so only 1% of the work actually gets accounted for), the rate of finding blocks per work would not have been affected at all - so you were ignored by everyone that knows how mining works since your data did not back up your claim. Also, it was shown that the rate of bad luck was likely to happen when we were at low p2pool network hash rates.
4) Suddenly, without "fixing anything" we had a run of luck close to 180% for a week or two, wiping out the bad luck from (1) above.
5) ASICs came online, no cgminer or p2pool dev had an avalon, but there were reports of p2pool not being able to keep up with avalons
6) actual, real reports with data came out showing that something isn't 100% right with ASICs and p2pool (they work, but need some tweaking) - but still no p2pool dev had one to fix it. We still don't know exactly what it is, but it has something to do with what avalon expects to see and what p2pool is serving. Aseras donated Avalon time to cgminer to finally get cgminer support for avalons.
7) BFL donated an ASIC to forrestv to make sure BFL SCs work with p2pool
8 ) you show up again, "see I was right - you should have listened to me although I had no data and no logical argument backing up my claim"

So in the end, you were wrong - but keep trying until you found someone with a real bug and then claimed credit for it, even though their bug had nothing to do with your evidence or what you are claiming originally.

Let's have some manners please......and some constructive criticism instead of childish remarks.

Demanding fixes without actually knowing what you are asking to fix is childish and naive. Developers don't know what to fix if an oracle shows up saying "fix it" - they need data, which you have never given. It's like saying, "My internet connection is slow. Microsoft, fix Windows!" Then later someone finds a problem with a printer driver and you say, "See, I told you Windows was broken and you all ignored me!" No one in the world will fix it and repeatedly demanding it will get you nowhere. That's why you feel like this lone crusader martyr, you don't know how to help and are frustrated that no one will do what you demand.

We've been begging you to give us data and bug reports for months, but you just lurk around to yell "fix it." That's why people are losing patience with you and is the behaviour of a troll.

[GrapeApe]

I noticed that some nodes have fees. I was wondering how they set it up that way? I am not planning on adding a fee to my node I'm just curious.

--fee lol, use --help

Wow don’t I feel like the idiot thx for the help.

[gyverlb]

[..long explanations..]
We've been begging you to give us data and bug reports for months, but you just lurk around to yell "fix it." That's why people are losing patience with you and is the behaviour of a troll.

Wow, you have the patience of an angel.

I'm not in any condition to be patient right now (health problems making my everyday life more difficult) but even in my best shape I don't think I would have it in me to write this.

[K1773R]

Naive?  My little world?  As that where we're at now - posting personal insults publicly?

Coming from the same person who has constantly denied that there is a problem with the stratum code implementation and still does, even though it's proven - that's quite rich, don't you think? It's a bit sad to dislike someone because they were right, but there's absolutely no excuse to resort to insulting them, weather they are correct or not.

You were not right, in fact you were very wrong. You need to understand why you're not getting anywhere and people have lost patience with you:

Order of events
1) We had a run of long blocks one month ago around the same time the stratum protocol was implemented in p2pool.
2) You said, "it's impossible to be this unlucky, it must be stratum since that was the most recent change. FIX STRATUM"
3) stratum has absolutely nothing to do with the rate of finding blocks per work. If stratum was 99% broken (so only 1% of the work actually gets accounted for), the rate of finding blocks per work would not have been affected at all - so you were ignored by everyone that knows how mining works since your data did not back up your claim. Also, it was shown that the rate of bad luck was likely to happen when we were at low p2pool network hash rates.
4) Suddenly, without "fixing anything" we had a run of luck close to 180% for a week or two, wiping out the bad luck from (1) above.
5) ASICs came online, no cgminer or p2pool dev had an avalon, but there were reports of p2pool not being able to keep up with avalons
6) actual, real reports with data came out showing that something isn't 100% right with ASICs and p2pool (they work, but need some tweaking) - but still no p2pool dev had one to fix it. We still don't know exactly what it is, but it has something to do with what avalon expects to see and what p2pool is serving. Aseras donated Avalon time to cgminer to finally get cgminer support for avalons.
7) BFL donated an ASIC to forrestv to make sure BFL SCs work with p2pool
8 ) you show up again, "see I was right - you should have listened to me although I had no data and no logical argument backing up my claim"

So in the end, you were wrong - but keep trying until you found someone with a real bug and then claimed credit for it, even though their bug had nothing to do with your evidence or what you are claiming originally.

Let's have some manners please......and some constructive criticism instead of childish remarks.

Demanding fixes without actually knowing what you are asking to fix is childish and naive. Developers don't know what to fix if an oracle shows up saying "fix it" - they need data, which you have never given. It's like saying, "My internet connection is slow. Microsoft, fix Windows!" Then later someone finds a problem with a printer driver and you say, "See, I told you Windows was broken and you all ignored me!" No one in the world will fix it and repeatedly demanding it will get you nowhere. That's why you feel like this lone crusader martyr, you don't know how to help and are frustrated that no one will do what you demand.

We've been begging you to give us data and bug reports for months, but you just lurk around to yell "fix it." That's why people are losing patience with you and is the behaviour of a troll.

you know there is a ignore button for spammers/trolls/morons/retards and also for ppl who dont understand anything but trying to argue about it (ie, him) lol

[PatMan]

Same people still living in denial. Please, do press the ignore button......

[tiktoc]

Sipa(one of the bitcoin developrs) did a pull request which may help with this https://github.com/bitcoin/bitcoin/pull/2677

Slush had similar problems on his pool and it was causing high invalids.

[gyverlb]

Sipa(one of the bitcoin developrs) did a pull request which may help with this https://github.com/bitcoin/bitcoin/pull/2677

Slush had similar problems on his pool and it was causing high invalids.

Oh look, people with a clue do what we wished for! How surprising...

It clearly will help: a 10x improvement in getblocktemplate will bring fresh air.

I still think they should implement this call on top of an asynchronous update of the block template to make the RPC call o(1) instead of o(nb_tx) or worse though. In the worst case, even if some params of getblocktemplate don't allow to precompute most of the template at least the common case scenario with no parameter can be sped up.

[gyverlb]

I still think they should implement this call on top of an asynchronous update of the block template to make the RPC call o(1) instead of o(nb_tx) or worse though. In the worst case, even if some params of getblocktemplate don't allow to precompute most of the template at least the common case scenario with no parameter can be sped up.

Suggested it, hopefully sipa will be interested enough to either explain why it can't work or make a mental note of the idea for the next performance roadblock.

[GrapeApe]

WOW! How?

1) actual bitcoin from git ( moves  variables ...Fees.. from main.h to main.cpp)
 2)  set in src/main.cpp the parameters to
int64 CTransaction::nMinTxFee = 1000000000;    # Override with -mintxfee
int64 CTransaction::nMinRelayTxFee = 1000000000;



3) compile bitcoin

 
4) in bitcoin.conf:
blockmaxsize=5000
blockprioritysize=0
blockminsize=0

Greets

I did what is stated above to the latest 0.8.2rc1 bitcoin-qt and my GBT latency is 3.00 ms (.003 s). Make these changes along with the changes Sipa is recommending should speed things up no?

[gyverlb]

I still think they should implement this call on top of an asynchronous update of the block template to make the RPC call o(1) instead of o(nb_tx) or worse though. In the worst case, even if some params of getblocktemplate don't allow to precompute most of the template at least the common case scenario with no parameter can be sped up.

Suggested it, hopefully sipa will be interested enough to either explain why it can't work or make a mental note of the idea for the next performance roadblock.

gmaxwell's comment makes me think the discussion about async processing of templates should be taken here. I'm wondering if forrestv implemented a p2pool template cache with the same approach I suggested could be provided by bitcoind. Especially just after a new block found on the network (see the comment here). There's no explicit configuration of bitcoind done to make signal p2pool new blocks so is it detecting them by calling a fast RPC method of bitcoind or is there a delay that could be shortened by using the "blocknotify" bitcoind option for example?

[Amph]

mintxfee=0.002
minrelaytxfee=0.002

those setting are the best for me, latency is now stable at 0.223