KanoPool since 2014 🐈 - PPLNS and Solo 0.5% fee - Worldwide - 2438 blocks

[joel coin]

I taught Stat at Cal Poly during my university days, and indeed you get LOTS of false conclusions on these boards based on emotion instead of mathematical rigor.  Funny at first, then tiring, then irritating - especially when coming from longtime members who should know better...  
I'd hate to know how high kanosan's aggrivation level gets sometimes with all the noob du jours chiming in during a run of bad luck!  

give us a break
we like to learn also

I second @padrinogtr's sentiment -- with all due respect @VRobb, everyone used to be a noob at some point (including you and Kano), until they weren't.  I won't speculate on how high Kano's aggravation level gets with noobs 'chiming' in the forum, but I will say that without noobs mining in the pool, this pool wouldn't grow and would most likely be forgotten by the ever increasing level of difficulty as the network grows, causing the runs of bad luck to become longer and more frequent.  

You don't need statistics to understand that

I am a noob to this pool, having joined about a month ago, just in time to have the worst luck I have ever seen in my not so noobish 3.5 years of mining. 

I am a noob to this pool as well, a noob to hands on mining since 2 weeks ago, my BCT forum account says I'm a "newbie" but but I've been messing with bitcoin since 2012 (or earlier) so not a noob to btc... So really these forum labels don't say much in the way of real experience lol 

[voodooman]

Question @kano:
Is the blockfinding in any way connected to the core client?
The only thing changed in the last few weeks is core update to 15.x. Isn't this right?
Okay beside the split.

Sorry for this unqualified Question. This just came to my mind cause I updated my node to 15.0.1.


Come ON BLOCKS!!!!

I'll answer that...no, Core doesn't affect finding blocks. It's your wallet, or if you're running a full node, you're helping process the mempool (the way I understand it). That's not mining.

Anyway...I've been running a full node for a long time, and have been running 15.0.1 with no issues.

Mine on!

Sorry for not digging deeper into my question.

I meant:
Could it be "a problem" with the new 0.15 core (or maybe not just the core itself) or let's say the network ALL OVER THE WOLRD
with probably more than 50% hosted in china!!??
Somehow the new block must be provided to the network (which is build by many many full node core hosts).
So if we find a block or are about to find a block in a second and another pool in china
is finding the "same" block and provides it to the chinese nodes because of the lower latency ( and which are probably more than 50% of the network)
and they get their block committed in that second we find the "right" block,
we even didn't get a chance to provide our block to the network or something like this.

So my question is:
Could we do anything, maybe by providing more full nodes in the western hemisphere, to get "more luck"
or just an even playground for this?

I think much of this stuff is also connected to low latencies (so it's also connected to how well the network is spread all over the world or near kanosan's server :-D )

Maybe I'm completely wrong but just for the protocol my 2 cents.


Oh and never mind kano. I think you're doing a great job!!!
I don't think there is anything wrong with your system.
I know it's just bad bad luck actually.

[miguelmorales85]

I would like to add Kano's node(s) to my node's whitelist.

Where can I find this information? Maybe it is private to prevent DoS attacks.

if it is something public please post it to add it to the whitelist of my nodes.

peace

[VRobb]

...Snip
Oh and never mind kano. I think you're doing a great job!!!
I don't think there is anything wrong with your system.
I know it's just bad bad luck actually.

Exactly.  Glad to see not everyone failed Stat!
Mine on and let's crack one already!! 

[Sosoft]

This thread and situation are so depressing right now. I can't imagine what Kano goes through when stuff like this happens. All I have to offer is my hashing power and well wishes. Kano pool has both of those.
Good luck Kano pool (Kano). I'll be back in a few days of out of town.

[kano]

At the moment I'm still trying to get stats from certain large miners to find a single case of a miner finding a block that didn't show up on the pool.
This would show something wrong other than bad luck.
Unfortunately a lot of the big miners reset every so often and thus lose all the stats up to the reset point, so I've not been able to check that many.

Anyone else here of course if you have time to do it and can, please check your miners.
What I'm looking for is a miner with a "Best Share" above Network Difficulty (1196792694098.793)
Any shares found that should be blocks, I can convert the "Best Share" share difficulty into a block hash and check it matches one of the blocks on the pool.
There aren't hundreds of blocks in the last month, so it's not hard to check.

Meanwhile, if this continues on for much longer, I'll consider putting the pool into maintenance mode for a few days (i.e. stop all mining) and continue my investigation that way.
I'm considering going back to the old code I ran until segwit required the update, and re-implementing segwit and none of the other random changes in ckpool.
I've been running the same code non-stop since August-8 when I added the 🐈 to the coinbase, and of course we've not had anything like this before this week.
Thje  Aug-8 change was only a minor change to the earlier Jul-20 change for segwit. There's been no other changes I've implemented, in the past year, to the mining code.
The most difficult issue is the fact that we do find blocks, rather than none at all, thus if there was a problem, it wouldn't be as easy to reproduce.

Some large miners have already moved hash rate away, temporarily, due to the cost issue of paying very large electricity bills and other related costs.

I'm also considering completely blocking NiceHash, for any who use that, since their stats are pretty close to the 'unexpected' limit at the moment

[qrnode]

s9

up for 9d18h8m54s
gh/s (rt) 13,768.07
gh/s (avg) 13,777.77
found blocks 0
local work 42,733,327
utility 16.85
wu 191,989.27
best share 6106892878

Not going anywhere. We plan to add more hash power soon.

[xuy]

With 62PH now are there still 1300 users?
I recalled there were 700~800 users when the pool was at ~60PH.

[kano]

With 62PH now are there still 1300 users?
I recalled there were 700~800 users when the pool was at ~60PH.

The big drop hasn't been a lot of users.

[padrinogtr]

im still here
but just with 2TH

[philipma1957]

At the moment I'm still trying to get stats from certain large miners to find a single case of a miner finding a block that didn't show up on the pool.
This would show something wrong other than bad luck.
Unfortunately a lot of the big miners reset every so often and thus lose all the stats up to the reset point, so I've not been able to check that many.

Anyone else here of course if you have time to do it and can, please check your miners.
What I'm looking for is a miner with a "Best Share" above Network Difficulty (1196792694098.793)
Any shares found that should be blocks, I can convert the "Best Share" share difficulty into a block hash and check it matches one of the blocks on the pool.
There aren't hundreds of blocks in the last month, so it's not hard to check.

Meanwhile, if this continues on for much longer, I'll consider putting the pool into maintenance mode for a few days (i.e. stop all mining) and continue my investigation that way.
I'm considering going back to the old code I ran until segwit required the update, and re-implementing segwit and none of the other random changes in ckpool.
I've been running the same code non-stop since August-8 when I added the 🐈 to the coinbase, and of course we've not had anything like this before this week.
The  Aug-8 change was only a minor change to the earlier Jul-20 change for segwit. There's been no other changes I've implemented, in the past year, to the mining code.
The most difficult issue is the fact that we do find blocks, rather than none at all, thus if there was a problem, it wouldn't be as easy to reproduce.

Some large miners have already moved hash rate away, temporarily, due to the cost issue of paying very large electricity bills and other related costs.

I'm also considering completely blocking NiceHash, for any who use that, since their stats are pretty close to the 'unexpected' limit at the moment

I have drawn the conclusion the slowdown started in late July
I also do not rent much from nicehash as my luck is not very good with them.  Since most hash  was pointed at mmpool  where hitting the block means  you get the tx fees I keep track of the blocks I hit with nicehash.
I hit a few lower then I should have.  So my conclusions kind of mirror  kano's. I am okay at a few things  no where am I in Kano's league as a coder.

[kano]

I'll be putting in a few patches into the pool code and see if that has any unexpected positive results.
I'm just doing them now and testing them so will be about 15:00 UTC and I'll do a pool restart (in about an hour)

[rifleman74]

I'll be putting in a few patches into the pool code and see if that has any unexpected positive results.
I'm just doing them now and testing them so will be about 15:00 UTC and I'll do a pool restart (in about an hour)

will keep an eye on it kano-san.  Thanks for the hard work you do for this pool!

[mousehouse]

I'll be putting in a few patches into the pool code and see if that has any unexpected positive results.
I'm just doing them now and testing them so will be about 15:00 UTC and I'll do a pool restart (in about an hour)

Thanks for all the effort made. Maybe it's all just bad luck ...

My S9 is chugging along nicely,

Elapsed    GH/S(RT)    GH/S(avg)    FoundBlocks    LocalWork      Utility    WU    BestShare               
19d7h40m45s 13,554.04 13,580.33 0 83,444,083    16.66 189,789.13 166157797112

[hap0py]

Staying loyal to pool.  Thank you for your hard work kano.
Unfortunately restarted 33 hours ago, but best share since then is:  9,186,558,754
Please say if there's any other information or anything that could help you.

Question:  is maintenance mode completely offline, or testnet, or something else?  Will you need some miners to stay connected?

[realhayesreal]

All the chickens are dead rooster as well.  Wife hates me.  Thinking of going after the dog.........BTC

[miguelmorales85]

I'll be putting in a few patches into the pool code and see if that has any unexpected positive results.
I'm just doing them now and testing them so will be about 15:00 UTC and I'll do a pool restart (in about an hour)

Thank you, please keep us informed about the results.

good luck.

[smutboy420]

Anyone else here of course if you have time to do it and can, please check your miners.
What I'm looking for is a miner with a "Best Share" above Network Difficulty (1196792694098.793)
Any shares found that should be blocks, I can convert the "Best Share" share difficulty into a block hash and check it matches one of the blocks on the pool.
There aren't hundreds of blocks in the last month, so it's not hard to check.

For what its worth the best share numbers are much lower then 1196792694098.793

But here they are if it helps at all with any thing.  

S9 number 1 best share is 7959749381
S9 number 2 best share is 2344219225
S9 number 3 best share is 10718647319

[padrinogtr]

I'll be putting in a few patches into the pool code and see if that has any unexpected positive results.
I'm just doing them now and testing them so will be about 15:00 UTC and I'll do a pool restart (in about an hour)

thank you for being on top of this
kano FTW!!!

[padrinogtr]

All the chickens are dead rooster as well.  Wife hates me.  Thinking of going after the dog.........BTC

do it for the pool