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

[gyverlb]

Hi guys,
I have priority issue with bitcoind latency.
My bitcoind node is literally flooded by hoax transactions, like this:
https://blockchain.info/tx/7227c9d392166fdcd917f39da2def36bcf69b437d90e349d73a48952c0fbdd69
https://blockchain.info/tx/292d1d7ff7ece8eb67b4bf3c2f523736679c616fce8216bf37ac23edec6e0e03
1 single transaction, nice miner fee included (1.46 USD actually), but it have over two thousands outputs and 100 KB in size! It's like a fishhook, whoever will take it, will loose seconds of latency.

Bitcoind latency is at ~ 4s (I have dedicated server with AMD A10 APU + SSD), for months it was at 0.3s. About 18 hours ago my latency went extremely high up to 7s and till then, network is under attack of these hoax transactions. Exactly same problem has been reported by my collegues at forum.bitcoin.pl. They showed be their stats and same latency issues (bitcoind latency at 3s or worse, like in my case).

My bitcoin.conf:

Code:

(...)
maxconnections=30
blockmaxsize=500000
mintxfee=0.00005

How to filter these transactions? 

Is your efficiency suffering? I have the same symptoms but my efficiency is still good.

When bitcoind 0.8.2 is released it should clean this dust by default.

[lenny_]

Yes, it is suffering. It was showing me nice 110% (I had node running for 2 weeks, with 6 thousand shares found), but it doesn't seem to be right, when bitcoind latency was at 3s.
I restarted bitcoind and updated p2pool to newest version. Since then I am having:

Code:

Node uptime: 0.090 days Peers: 7 out, 1 in

Local rate: 20.5GH/s (9.6% DOA) Expected time to share: 0.0754 hours

Shares: 27 total (5 orphaned, 3 dead) Efficiency: 86.79%

Look there: lenny.dnsd.me:9332

[gyverlb]

Look there: lenny.dnsd.me:9332

Down. You might have other problems?

[lenny_]

Not only I have this problem. Just look at stale share ratio from my peers:

Code:

2013-05-17 02:44:35.450770 Punishing share for 'Block-stale detected! 689887341d7de716487a3d0f03f4bdef93d1717cdf399f7eef < 7483cd23042b95f0733f56b263231115a50d8c5d10e7f4cde9'! Jumping from 3244e6d0 to dc70dda5!
2013-05-17 02:44:35.485379 New work for worker! Difficulty: 10.000000 Share difficulty: 1354.504804 Total block value: 25.000000 BTC including 0 transactions
2013-05-17 02:44:36.197110 Punishing share for 'Block-stale detected! 689887341d7de716487a3d0f03f4bdef93d1717cdf399f7eef < 7483cd23042b95f0733f56b263231115a50d8c5d10e7f4cde9'! Jumping from 3244e6d0 to dc70dda5!
2013-05-17 02:44:36.202615 Punishing share for 'Block-stale detected! 689887341d7de716487a3d0f03f4bdef93d1717cdf399f7eef < 7483cd23042b95f0733f56b263231115a50d8c5d10e7f4cde9'! Jumping from 3244e6d0 to dc70dda5!
2013-05-17 02:44:36.636124 P2Pool: 17331 shares in chain (17336 verified/17336 total) Peers: 6 (0 incoming)
2013-05-17 02:44:36.636223  Local: 8518MH/s in last 10.0 minutes Local dead on arrival: ~2.5% (0-8%) Expected time to share: 11.4 minutes
2013-05-17 02:44:36.636254  Shares: 34 (11 orphan, 4 dead) Stale rate: ~44.1% (28-61%) Efficiency: ~68.6% (48-88%) Current payout: 0.6031 BTC
2013-05-17 02:44:36.636287  Pool: 712GH/s Stale rate: 18.6% Expected time to block: 18.7 hours
2013-05-17 02:44:38.195234 Punishing share for 'Block-stale detected! 689887341d7de716487a3d0f03f4bdef93d1717cdf399f7eef < 7483cd23042b95f0733f56b263231115a50d8c5d10e7f4cde9'! Jumping from 381c6c6d to 3244e6d0!
2013-05-17 02:44:38.229273 New work for worker! Difficulty: 10.000000 Share difficulty: 1354.631688 Total block value: 25.000000 BTC including 0 transactions
2013-05-17 02:44:39.643104 P2Pool: 17332 shares in chain (17337 verified/17337 total) Peers: 6 (0 incoming)
2013-05-17 02:44:39.643269  Local: 8303MH/s in last 10.0 minutes Local dead on arrival: ~2.6% (0-8%) Expected time to share: 11.7 minutes
2013-05-17 02:44:39.643395  Shares: 34 (11 orphan, 4 dead) Stale rate: ~44.1% (28-61%) Efficiency: ~68.8% (48-88%) Current payout: 0.6031 BTC
2013-05-17 02:44:39.643508  Pool: 710GH/s Stale rate: 18.7% Expected time to block: 18.8 hours
2013-05-17 02:44:41.184819 Punishing share for 'Block-stale detected! 689887341d7de716487a3d0f03f4bdef93d1717cdf399f7eef < 7483cd23042b95f0733f56b263231115a50d8c5d10e7f4cde9'! Jumping from 381c6c6d to 3244e6d0!
2013-05-17 02:44:41.193301 Punishing share for 'Block-stale detected! 689887341d7de716487a3d0f03f4bdef93d1717cdf399f7eef < 7483cd23042b95f0733f56b263231115a50d8c5d10e7f4cde9'! Jumping from 381c6c6d to 3244e6d0!
2013-05-17 02:44:43.730638 Peer sent entire transaction 2a8ae4b892d5de8873163bf28db985ed0ed140144426a66b5cd8107382d3c542 that was already received
2013-05-17 02:44:46.193085 Punishing share for 'Block-stale detected! 689887341d7de716487a3d0f03f4bdef93d1717cdf399f7eef < 7483cd23042b95f0733f56b263231115a50d8c5d10e7f4cde9'! Jumping from 381c6c6d to 3244e6d0!
2013-05-17 02:44:46.198400 Punishing share for 'Block-stale detected! 689887341d7de716487a3d0f03f4bdef93d1717cdf399f7eef < 7483cd23042b95f0733f56b263231115a50d8c5d10e7f4cde9'! Jumping from 381c6c6d to 3244e6d0!
2013-05-17 02:44:47.374998 Punishing share for 'Block-stale detected! 689887341d7de716487a3d0f03f4bdef93d1717cdf399f7eef < 7483cd23042b95f0733f56b263231115a50d8c5d10e7f4cde9'! Jumping from c51fd7a0 to 381c6c6d!

gyverlb: I PM you my private IP, check there, if dnsd.me doesn't resolve for you.

[daemondazz]

My getwork latency has also increased from ~0.5s to ~6s approx 18 hours ago.

[gyverlb]

Maybe using bitcoind from current git could solve this for people with low efficiencies. Could be worth a try. The Gavin's patch that stirred controversy in the past weeks should refuse to relay this kind of dust by default.

[Prattler]

My bitcoind node is literally flooded by hoax transactions, like this:
https://blockchain.info/tx/7227c9d392166fdcd917f39da2def36bcf69b437d90e349d73a48952c0fbdd69
My bitcoin.conf:

Code:

(...)
mintxfee=0.00005

How to filter these transactions?  

Upgrade your bitcoind to the latest git. That should solve it.

I'd also recommend changing the config to

Code:

(...)
mintxfee=0.0001

or a bit higher. Your setting is way too low.

[lenny_]

Code:

time bitcoind getblocktemplate > /dev/null

real	0m4.595s
user	0m0.044s
sys	0m0.004s

Nice, isn't? 
I increased mintxfee=0.0001 as you said. Will try to upgrade bitcoind to that one from git.

[Subo1977]

Code:

time bitcoind getblocktemplate > /dev/null

real	0m4.595s
user	0m0.044s
sys	0m0.004s

Nice, isn't?  
I increased mintxfee=0.0001 as you said. Will try to upgrade bitcoind to that one from git.

mintxfee=

from source:

/** Fees smaller than this (in satoshi) are considered zero fee (for transaction creation) */
int64 CTransaction::nMinTxFee = 10000;  // Override with -mintxfee


so is'nt mintxfee=0.0001  a little to low?

 

[Prattler]

Yes, mintxfee=0.0001 is still very low. It's also the new bitcoind default as you correctly state.

I'd go for mintxfee=0.0002 or higher.

[Subo1977]

Yes, mintxfee=0.0001 is still very low. It's also the new bitcoind default as you correctly state.

I'd go for mintxfee=0.0002 or higher.

 0.00000001 BTC = 1 Satoshi

 0.0001 Satoshi  = 0.00000000001 BTC
 
I think it must be mintxfee =100000 or 200000

[zvs]

Yes, mintxfee=0.0001 is still very low. It's also the new bitcoind default as you correctly state.

I'd go for mintxfee=0.0002 or higher.

 0.00000001 BTC = 1 Satoshi

 0.0001 Satoshi  = 0.00000000001 BTC
 
I think it must be mintxfee =100000 or 200000

mintxfee 10000 is the default

that's 0.0001

it was changed in latest version.  the relay fee has been 0.0001 for a long time (year+?)... the creation fee was 0.0005...  so, was a bit odd.

i wasnt going to at first, but too many transactions filled with.. not much.  i set the relay fee to 0.0005 (the old creation fee),  or 50000

if it's really a problem, if you set it to something like 50001, you'll see much less transactions

[PatMan]

As discussed in the previous few pages, there are clearly some stratum issues still to resolve, and scalability issues to cope with ASIC sized workloads. BFL are sending (?have sent) forrestv a 5GH unit which should help drive development of both of these.

Excellent. At last, it seems some progress is finally being made.  

[Subo1977]

As discussed in the previous few pages, there are clearly some stratum issues still to resolve, and scalability issues to cope with ASIC sized workloads. BFL are sending (?have sent) forrestv a 5GH unit which should help drive development of both of these.

Excellent. At last, it seems some progress is finally being made.  

Which?  I Hope bugfixes for Stratum 

[PatMan]

As discussed in the previous few pages, there are clearly some stratum issues still to resolve, and scalability issues to cope with ASIC sized workloads. BFL are sending (?have sent) forrestv a 5GH unit which should help drive development of both of these.

Excellent. At last, it seems some progress is finally being made.   

Which?  I Hope bugfixes for Stratum 

Exactly, that's the number one priority eh? 

[azw409]

I've been mining on P2Pool for a week and I'm very impressed. I had a lot of problems with Phoenix originally but since switching to CGMiner all the stats are now good and I'm earning more than I ever did with 50BTC.

The only remaining issue is that the bitcoind getblocktemplate latency constantly increases over time in quite large jumps. It starts around 0.5s and creeps up slowly to around 2s then jumps to 10s and stays there.

Restarting everything: Bitcoin-qt, p2pool and CGMiner fixes it again but within 24 hours it'll be back up again.

Anybody have any ideas what could be going on ?

Thanks

[zvs]

I've been mining on P2Pool for a week and I'm very impressed. I had a lot of problems with Phoenix originally but since switching to CGMiner all the stats are now good and I'm earning more than I ever did with 50BTC.

The only remaining issue is that the bitcoind getblocktemplate latency constantly increases over time in quite large jumps. It starts around 0.5s and creeps up slowly to around 2s then jumps to 10s and stays there.

Restarting everything: Bitcoin-qt, p2pool and CGMiner fixes it again but within 24 hours it'll be back up again.

Anybody have any ideas what could be going on ?

Thanks

not sure about large jumps, but if your bitcoind is accepting transactions into memory that no pool will ever include then it'll gradually get slower

try bitcoind getmininginfo and see if it's much higher when it starts getting slow (or bitcoin-qt or whatever)

[Searinox]

I've been mining on P2Pool for a week and I'm very impressed. I had a lot of problems with Phoenix originally but since switching to CGMiner all the stats are now good and I'm earning more than I ever did with 50BTC.

The only remaining issue is that the bitcoind getblocktemplate latency constantly increases over time in quite large jumps. It starts around 0.5s and creeps up slowly to around 2s then jumps to 10s and stays there.

Restarting everything: Bitcoin-qt, p2pool and CGMiner fixes it again but within 24 hours it'll be back up again.

Anybody have any ideas what could be going on ?

Thanks

Version 11.4 introduced a bug where the latency creeps up over time. The only fix I found was reverting to 11.3 which is located here http://u.forre.st/u/drxlzowj/p2pool_win32_11.3.zip

[Richy_T]

I'm having an issue with p2pool and my router. It's not a show-stopper but I was wondering if someone might have an idea what's going on or be able to work with me to work out what is going on.

Basically, anytime I turn on QOS, I appear to stop doing any useful work. cgminer trundles along, fans spin, heat generated, Unfortunately, it appears that that's all wasted. the debug output of p2pool says that it's punishing shares for being stale. On the pools graph page, the incoming part of peers and the p2p in on traffic rate drop to zero. For whatever reason, it looks like peers are no longer supplying me with useful information (?)

I've even set things so that p2pool and bitcoind are QOSed to the highest speed category just in case they were getting squashed that way but it still gives the same result.

I can leave QOS off for now but I'd like to get this working properly. I'm using Tomato so the configuration is fairly straightforward and easy. I'll post some pictures in a little bit for illustration. I can perform packet captures and other good stuff but I'm not up on the p2pool networking.

[zvs]

I'm having an issue with p2pool and my router. It's not a show-stopper but I was wondering if someone might have an idea what's going on or be able to work with me to work out what is going on.

Basically, anytime I turn on QOS, I appear to stop doing any useful work. cgminer trundles along, fans spin, heat generated, Unfortunately, it appears that that's all wasted. the debug output of p2pool says that it's punishing shares for being stale. On the pools graph page, the incoming part of peers and the p2p in on traffic rate drop to zero. For whatever reason, it looks like peers are no longer supplying me with useful information (?)

I've even set things so that p2pool and bitcoind are QOSed to the highest speed category just in case they were getting squashed that way but it still gives the same result.

I can leave QOS off for now but I'd like to get this working properly. I'm using Tomato so the configuration is fairly straightforward and easy. I'll post some pictures in a little bit for illustration. I can perform packet captures and other good stuff but I'm not up on the p2pool networking.

the only thing i've ever used in windows that i've found did a proper job of managing bandwidth was netlimiter