F2Pool「魚池」🐟

[kano]

Not even an appology for causing the biggest bitcoin fuck up since the march 2013 hard fork, which was a bug and not caused by incompetence and greed.

Chun Wang, (F2Pool) has now not only proven the second time that he is clearly not compented and greedy but is also quite arrogant.

We will continue do SPV mining despite the incident, and I think so will AntPool and BTC China.

Chun Wang, Mark Kaperkles, Trendon Shavers -> The three worst people that ever happened to Bitcoin. The whole bitcoin network would much safer without F2Pools hashing power. Please find some other thing you can exploit/abuse and destroy.

I contacted someone I know at bitmain when it happened and got them to talk to the pool engineers to sort it out
(that's why antpool switched away from the v2 fork to the v3 fork)
antpool should hopefully have resolved the problem from recurring also.

But the idea for SPV came from eligius eloipool - blame them for starting it.
eligius has fixed it for the v3 change before it happened but even the eloipool git was very late updating the code to v3 so who knows what problems will occur like that in the future for pools using eloipool ...
The other pools using eloipool base have to fork it, so they probably have a lot of changes not in common with the git.

[kano]

RE: sloopy

First, we are not a pure SPV mining operation. We only do that within seconds after a new block is found on the network. We did have a safe guard in the beginning ...

Yes that was the cause of the problem and your safeguard didn't work.

[cloverme]

We will continue do SPV mining despite the incident, and I think so will AntPool and BTC China.

Another very good reason people should not mine on Chinese pools.  This is EXTREMELY bad for the Bitcoin network.

Agreed, these guys are greedy and shouldn't be used by the community. It's unfortunate that math can't defeat greed in this circumstance. They're bad news for Bitcoin.

[macbook-air]

Mining on block header is nothing wrong. It is not greed. It is strategy.

[p3yot33at3r]

Mining on block header is nothing wrong. It is not greed. It is strategy.

Then it is an incompetent, greedy & damaging strategy. For you, miners & the network in general.

As has been pointed out repeatedly.

[macbook-air]

Mining on block header is nothing wrong. It is not greed. It is strategy.

Then it is an incompetent, greedy & damaging strategy. For you, miners & the network in general.

As has been pointed out repeatedly.

Can you please prove me how it is greedy and damaging? Can you go to AntPool and BTC China's thread tell them the same? Can you go to Eligius's thread where the strategy was invented?

[kano]

Mining on block header is nothing wrong. It is not greed. It is strategy.

Then it is an incompetent, greedy & damaging strategy. For you, miners & the network in general.

As has been pointed out repeatedly.

Can you please prove me how it is greedy and damaging? Can you go to AntPool and BTC China's thread tell them the same? Can you go to Eligius's thread where the strategy was invented?

Just to save p3yot33at3r the trouble of doing it ...
Yes I have for both Eligius and AntPool

I have even run (and still am running) long running tests that show mining on Eligius eloipool is still slower block changes with their empty blocks every block change than mining on my pool that never gives out empty blocks (unless bitcoind has no transactions in it)
It is also easy to test yourself as I have explained.
https://bitcointalk.org/index.php?topic=1085800.msg11607167#msg11607167
https://bitcointalk.org/index.php?topic=1085800.msg11681429#msg11681429

But the point about empty blocks is at what point is it OK to mine empty blocks and not confirm new transactions?
1% of blocks? 10% of blocks? Obviously not 100% of blocks coz then no new transactions would be confirmed and bitcoin would die.
If it's 1% of blocks then you are saying it would be OK for my pool to mine ALL blocks empty - since my pool is currently less than 1% of the bitcoin network.
Of course I'd never do it.

The argument for mining empty blocks is that the pools are too slow to generate new work on a block change.
Well that argument is rubbish. My pool doesn't have that problem. So maybe you are just making an excuse for your poorly written pool software?

So now that's all done, what are you going to do about it?
I would have thought that after you lost 125BTC yesterday due to this you might stop doing it ... but I guess 125BTC means nothing to the pool?

[loshia]

Yeah it is very hard to explain to Chinese "engineers" that something is totally wrong. As long as it is working "somehow" they will do nothing about it and there is no force on earth that will make them think the opposite  
They are just copy pasting from each other and that is all.
And they do not care..
The usual excuse is please do understand...we will put more men's power.
You need to put more brain power folks and make things right!

[macbook-air]

Mining on block header is nothing wrong. It is not greed. It is strategy.

Then it is an incompetent, greedy & damaging strategy. For you, miners & the network in general.

As has been pointed out repeatedly.

Can you please prove me how it is greedy and damaging? Can you go to AntPool and BTC China's thread tell them the same? Can you go to Eligius's thread where the strategy was invented?

Just to save p3yot33at3r the trouble of doing it ...
Yes I have for both Eligius and AntPool

I have even run (and still am running) long running tests that show mining on Eligius eloipool is still slower block changes with their empty blocks every block change than mining on my pool that never gives out empty blocks (unless bitcoind has no transactions in it)
It is also easy to test yourself as I have explained.
https://bitcointalk.org/index.php?topic=1085800.msg11607167#msg11607167
https://bitcointalk.org/index.php?topic=1085800.msg11681429#msg11681429

But the point about empty blocks is at what point is it OK to mine empty blocks and not confirm new transactions?
1% of blocks? 10% of blocks? Obviously not 100% of blocks coz then no new transactions would be confirmed and bitcoin would die.
If it's 1% of blocks then you are saying it would be OK for my pool to mine ALL blocks empty - since my pool is currently less than 1% of the bitcoin network.
Of course I'd never do it.

The argument for mining empty blocks is that the pools are too slow to generate new work on a block change.
Well that argument is rubbish. My pool doesn't have that problem. So maybe you are just making an excuse for your poorly written pool software?

So now that's all done, what are you going to do about it?
I would have thought that after you lost 125BTC yesterday due to this you might stop doing it ... but I guess 125BTC means nothing to the pool?

I second this.

I have to make things clearly, as some people may not fully understand what is going on. We are NOT running the pool without a full node nor only mining on top of unverified blocks as some people suggested, at least it is not what we intended to do. The incident is only because the fall back safeguard mechanism was manually deactivated for upgrade at some point.

This should be able to easily corrected. And after that, our operation will be safe enough for everyone. I will also double check with AntPool and BTC China to ensure that their operations are as safe. It is not about greedy. F2Pool is running by a group of geeks. We always like to explore new technologies and possibilities. That is why we are among the first to introduce FSS RBF, why we chose to have a countdown clock to Pluto in our coinbase, and why we donated the Namecoin and Huntercoin community with 25000 NMC and more. We are trying our best to make the best profit to our customers. And we are also responsible to the security and usability of the Bitcoin network. And we believe the way we are doing bitcoin mining is safe yet profitable to our miners.

[p3yot33at3r]

And we believe the way we are doing bitcoin mining is safe yet profitable to our miners.

Unfortunately, recent events have proven that this is not the case.

[philipma1957]

Okay so how long will it take this to be fixed?

I mine here not much but some.   About 2th------my actual machines.

I rent large ph from miners at times and mine at:

 solo.ckpool

kanos pool

and mmpool.org

So the risk here is small to me personally as 2th does not earn much and these miners are costing me and my friend 3 cents a kwatt.

But F2pool lost 125 btc of a claimed 10,000 btc reserve. 
So as I read this losing 1.25% of all of F2pool's reserve is still not enough to get F2pool to stop doing this style of mining.
We all know that F2pool is the gold standard of pps mining and has more then 22% of the entire network.
Once you add in the merged coins they give they are pretty much the best pool to mine in to a miner that does not want variance.

Of course f2pool is subject to a withholding attack like no other pool.  I could direct 4ph a day to nice hash and west hash offer to rent it out at a bargain price of  0.0095  and my partner in the attack could go to nicehash and rent the gear for 0.0098 point it all at F2pool.

My losses would be about 3 to 4 btc a day and F2pool would lose more then 40 btc a day.

Also if someone choose to they could skip nicehash and direct withholding miners right at  F2pool .

They could also have the mining gear discard shares over 80 bill    with reduces  the  winning blocks the gear makes but the gear still wold make blocks.
  My point here is F2pool is inviting people to direct mining gear at them that will discard some blocks.  Many people do not like the style of mining being done here.   If I direct 100th here that has software discarding all 80+ bill  bill shares it hurts F2pool. Not me. I would get my 1 btc a day - 4% or about 0.96btc and the merged coins brings me near 100% payout.

 F2pool would simply think their luck is a bit bad.  That day.

Not that I would ask Kano or Ck to build me a cgminer that discards shares. But many people could write this into a  cgminer custom build.

I simply do not understand why F2pool chooses to ignore this risk and invites angry people to attack them this way by mining the way they do.

[jonnybravo0311]

As long as people keep pointing hash to this pool, there is no incentive for a change.  Since they've still got over 70PH/s, it doesn't seem too many people are doing much complaining about their practices.

[philipma1957]

As long as people keep pointing hash to this pool, there is no incentive for a change.  Since they've still got over 70PH/s, it doesn't seem too many people are doing much complaining about their practices.

Which is why I point my actual miners here.  My friend provided his office and 1.5 kwatts at 3 cents  a kwatt and I  provided the miners.

Pointing them here gives us simple easy predictability of income.  No other pool does it as well.

[loshia]

As long as people keep pointing hash to this pool, there is no incentive for a change.  Since they've still got over 70PH/s, it doesn't seem too many people are doing much complaining about their practices.

Which is why I point my actual miners here.  My friend provided his office and 1.5 kwatts at 3 cents  a kwatt and I  provided the miners.

Pointing them here gives us simple easy predictability of income.  No other pool does it as well.

REALY?
Having in mind that net diff stays more or less same each pool operated by honest people shal give same predictable income. The only thing you need is patience....and you can be sure that 100% avarage luck is not a science fiction...

[macbook-air]

Okay so how long will it take this to be fixed?

It has already been fixed.

But F2pool lost 125 btc of a claimed 10,000 btc reserve. 
So as I read this losing 1.25% of all of F2pool's reserve is still not enough to get F2pool to stop doing this style of mining.
We all know that F2pool is the gold standard of pps mining and has more then 22% of the entire network.
Once you add in the merged coins they give they are pretty much the best pool to mine in to a miner that does not want variance.

We did not lose 125 btc. We mined only 4 orphan blocks, which is worth 100 btc. AntPool switched to prevblkhash=363731 0000000000000000009cc829aa25b40b2cd4eb83dd498c12ad0d26d90c439d99 at 2015-07-04T02:09:32.802Z, btcchina and we followed 0.21/0.24 seconds later. We found invalid block 363732 0000000000000000155f2519d35cd5d2869900bcc5093594b27763a0315390b4 at 02:23:06.258Z, 2.54 seconds later AntPool started to mine on to our version of 363732. BTC China however ignored our block continue mining on the invalid 363731 from BTC Nuggets until 02:46:49.826Z when bitfury found the valid version of block 363731 00000000000000000c28e23330c29046f19e817fe8fe039f4044b2b2882aef53. AntPool did not switch back to the valid chain until 03:19:34.907Z. These pools started to recognize what had happened and discuss the matter on WeChat group. I was sleeping, but someone called me regarding the fork. We also managed to switch to the valid chain a few minutes later at 03:25:40.221Z.

So BTC China lost 37 minutes 17 seconds, AntPool lost 70 minutes 2 seconds, we lost 76 minutes 7 seconds. At each pool's current hashrate, AntPool lost 36.5 BTC, BTC China lost 9.7 BTC, we lost 34.7 BTC.

[macbook-air]

Of course f2pool is subject to a withholding attack like no other pool.  I could direct 4ph a day to nice hash and west hash offer to rent it out at a bargain price of  0.0095  and my partner in the attack could go to nicehash and rent the gear for 0.0098 point it all at F2pool.

My losses would be about 3 to 4 btc a day and F2pool would lose more then 40 btc a day.

Also if someone choose to they could skip nicehash and direct withholding miners right at  F2pool .

They could also have the mining gear discard shares over 80 bill    with reduces  the  winning blocks the gear makes but the gear still wold make blocks.
  My point here is F2pool is inviting people to direct mining gear at them that will discard some blocks.  Many people do not like the style of mining being done here.   If I direct 100th here that has software discarding all 80+ bill  bill shares it hurts F2pool. Not me. I would get my 1 btc a day - 4% or about 0.96btc and the merged coins brings me near 100% payout.

 F2pool would simply think their luck is a bit bad.  That day.

Not that I would ask Kano or Ck to build me a cgminer that discards shares. But many people could write this into a  cgminer custom build.

I simply do not understand why F2pool chooses to ignore this risk and invites angry people to attack them this way by mining the way they do.

I have already replied you in the ck thread last month. Please don't do that. We are a real pool without a single TH of bitcoin hashing power on our own. While half of AntPool's hashrate comes from their own equipments and mining farm. Let AntPool become a monopoly is no good to the Bitcoin ecology. Block withholding is steal, which is crime. We have already suffered badly from DDoS these days. Please don't do that. If you don't agree with us, simply leave. Thanks.

[philipma1957]

Of course f2pool is subject to a withholding attack like no other pool.  I could direct 4ph a day to nice hash and west hash offer to rent it out at a bargain price of  0.0095  and my partner in the attack could go to nicehash and rent the gear for 0.0098 point it all at F2pool.

My losses would be about 3 to 4 btc a day and F2pool would lose more then 40 btc a day.

Also if someone choose to they could skip nicehash and direct withholding miners right at  F2pool .

They could also have the mining gear discard shares over 80 bill    with reduces  the  winning blocks the gear makes but the gear still wold make blocks.
  My point here is F2pool is inviting people to direct mining gear at them that will discard some blocks.  Many people do not like the style of mining being done here.   If I direct 100th here that has software discarding all 80+ bill  bill shares it hurts F2pool. Not me. I would get my 1 btc a day - 4% or about 0.96btc and the merged coins brings me near 100% payout.

 F2pool would simply think their luck is a bit bad.  That day.

Not that I would ask Kano or Ck to build me a cgminer that discards shares. But many people could write this into a  cgminer custom build.

I simply do not understand why F2pool chooses to ignore this risk and invites angry people to attack them this way by mining the way they do.

I have already replied you in the ck thread last month. Please don't do that. We are a real pool without a single TH of bitcoin hashing power on our own. While half of AntPool's hashrate comes from their own equipments and mining farm. Let AntPool become a monopoly is no good to the Bitcoin ecology. Block withholding is steal, which is crime. We have already suffered badly from DDoS these days. Please don't do that. If you don't agree with us, simply leave. Thanks.

I like your pool.  I would never do what I said can be done to you. I am trying to understand why you want to mine the way people are complaining to about.

Frankly I think you are the best pool there is for pps. Bar none.  Which is why you have 20% of the network mining with you. 

[kano]

(All times below are as they specify UTC+00)

Yes, I stand corrected it was 4 f2pool blocks, not 5.
The main 'V2 fork was':

The first V2 in the longer fork (yes this wasn't actually the first V2 block after the change over)
2015-07-04 02:09:28+00 0000000000000000009cc829aa25b40b2cd4eb83dd498c12ad0d26d90c439d99 BTC Nuggets

Then F2Pool caused the extended fork by mining on that block:

All the below were V3 layered on top of the V2:
2015-07-04 02:23:06+00 0000000000000000155f2519d35cd5d2869900bcc5093594b27763a0315390b4 F2Pool
2015-07-04 02:49:11+00 00000000000000000cb7a20ee4e199e347ad7369936abae53a1518efa531ec61 F2Pool
2015-07-04 02:52:18+00 00000000000000000966d65e0fd87d1d5a8f154a2c955816c28e2006e381aa18 AntPool (shortly after this I contacted Bitmain)
2015-07-04 02:56:54+00 00000000000000001301bfd6f566a421c7eeba103d09b312032ca065cb185de7 F2Pool
2015-07-04 03:05:59+00 000000000000000013fe26675faa8f7dccd55ce5485bb6d0373fa66345901436 F2Pool

So AntPool switched some time after 2015-07-04 02:52:18+00
and F2Pool switched some time after 2015-07-04 03:05:59+00

we also know the latest they could have switched by their next block on the 'V3 fork'

F2Pool:
2015-07-04 04:33:01+00 00000000000000000d0d362a2fbb3c3e7c8c748cc169ea90663cd18a354bc09b

AntPool: (earlier)
2015-07-04 04:09:31+00 00000000000000000b9d7006b0a893302e02becbc2faf78e79e000bb51721db0

Both pools were obviously mining on the 'v2 fork' since 2015-07-04 02:09:28+00

[macbook-air]

(All times below are as they specify UTC+00)

Yes, I stand corrected it was 4 f2pool blocks, not 5.
The main 'V2 fork was':

The first V2 in the longer fork (yes this wasn't actually the first V2 block after the change over)
2015-07-04 02:09:28+00 0000000000000000009cc829aa25b40b2cd4eb83dd498c12ad0d26d90c439d99 BTC Nuggets

Then F2Pool caused the extended fork by mining on that block:

All the below were V3 layered on top of the V2:
2015-07-04 02:23:06+00 0000000000000000155f2519d35cd5d2869900bcc5093594b27763a0315390b4 F2Pool
2015-07-04 02:49:11+00 00000000000000000cb7a20ee4e199e347ad7369936abae53a1518efa531ec61 F2Pool
2015-07-04 02:52:18+00 00000000000000000966d65e0fd87d1d5a8f154a2c955816c28e2006e381aa18 AntPool (shortly after this I contacted Bitmain)
2015-07-04 02:56:54+00 00000000000000001301bfd6f566a421c7eeba103d09b312032ca065cb185de7 F2Pool
2015-07-04 03:05:59+00 000000000000000013fe26675faa8f7dccd55ce5485bb6d0373fa66345901436 F2Pool

So AntPool switched some time after 2015-07-04 02:52:18+00
and F2Pool switched some time after 2015-07-04 03:05:59+00

we also know the latest they could have switched by their next block on the 'V3 fork'

F2Pool:
2015-07-04 04:33:01+00 00000000000000000d0d362a2fbb3c3e7c8c748cc169ea90663cd18a354bc09b

AntPool: (earlier)
2015-07-04 04:09:31+00 00000000000000000b9d7006b0a893302e02becbc2faf78e79e000bb51721db0

Both pools were obviously mining on the 'v2 fork' since 2015-07-04 02:09:28+00

There was another one:

BTC China started to mine on BTC Nuggets' invalid block 363726 0000000000000000032527aa796d3672e32e5f85a452d3a584a28fc7efbcd5d0 at 01:56:19.25Z, we followed 0.23 seconds later, and AntPool 1.58 seconds later, until 01:59:44.61Z when BW found block 000000000000000002279afdc82b23535938b34a0cb4aaa04dcb85dc1768608e of the same height. All three pools started to mine on 363727 000000000000000002279afdc82b23535938b34a0cb4aaa04dcb85dc1768608e. 3 minutes, 25 seconds lost in total, no blocks were found during this period.

[djeZo]

As long as people keep pointing hash to this pool, there is no incentive for a change.  Since they've still got over 70PH/s, it doesn't seem too many people are doing much complaining about their practices.

Which is why I point my actual miners here.  My friend provided his office and 1.5 kwatts at 3 cents  a kwatt and I  provided the miners.

Pointing them here gives us simple easy predictability of income.  No other pool does it as well.

Not true.

According to this: https://bitcointalk.org/index.php?topic=1059040.0

High stale shares will suck all the possible profit.

No point mining there for EU/USA. China network lines are rubbish.