[ Bonus PPS Pool - 130% PPS - OPEN ]

[Aseras]

The pool comes and goes for me as well.  When it is up it is MUCH better, but I still see a ridiculously high number of rejects and discards.

pool            share        accepted      rejects     discards     stales   

bonuspoool 10582    19035    752/4 %    834/4.4 %    59/0.3 %    

otherpool  19819             52138         2/0%         141/0%       1/0%

[Clipse]

The pool comes and goes for me as well.  When it is up it is MUCH better, but I still see a ridiculously high number of rejects and discards.

pool            share        accepted      rejects     discards     stales    

bonuspoool 10582    19035    752/4 %    834/4.4 %    59/0.3 %    

otherpool  19819             52138         2/0%         141/0%       1/0%

Its by no means as it should be yet but excluding stales generated when I have to reset the pool, I have been running <1% stales since my last post here across a ton of LP's so I can see improvements looking at those results.

My first goal is just to make it worthwhile for everyone and have them earn above 100% thus lowering the stales to where it has been running since my last post is moving into the right direction.

[Clipse]

UPDATE :

Seems the changes Ive made improved the post-LP situations drasticly, only 3 rejects (1 per LP) after the last 12 blocks occured as opposed to the old mass LP dumps so its looking very good now. The 3 rejects I experienced could also be due to my latency to the server since its between 400-500ms from where I am connecting from.

[Clipse]

WEBSTATS UPDATE COMING :

By Sunday or likely earlier I will have the next step to the webstats made available which would include the following:

- user login
- Seperation for userstats from payment history, so its a bit cleaner.
- Correct difficulty changes reflected ie. finally getting rid of the current bug displaying new difficulties wrong way around.
- First steps to account details managed by user ie. btc wallet, email address which would be used hopefully soon for automated payments.

More features will follow after this including worker json, seperately managed workers and whatever else I can find time for to add.

[Clipse]

Can the person who is using the following IP please PM me asap, IP: 68.54.174.227

Opening nearly 1000 connections isnt helping anyone

Also, whoever connects with username: 123456 , not sure what you are aiming at but if you want to earn then use your actual workername. Workernames will be enforced soon but until then its your responsibility to actually use the correct details.

Thanks

[Clipse]

Finally seems I got the pool stable and not slamming rejects LP after LP

The most recent spree was a very good test where 5 blocks came within span of 10minutes, it produced 1 reject only which is a fair value.

I will be moving current payment period over to payment history in about 4hrs so then we will have fresh stats for new payment cycle to look at, hopefully the start of stability and reliable low stales.

[DutchBrat]

I have seen my reject% dropping all day.... Enjoyed it !!!! Good job 

[DutchBrat]

Doing really great for the last hours, now the ool is down... communication and LP errors

[Clipse]

Doing really great for the last hours, now the ool is down... communication and LP errors

Im looking into it, been good day and now these random issues, gah losing my mind.

[DutchBrat]

It has been a really good day !!!!!

[Aseras]

Stales are way better. Rejects too. Still seeing 15-20% discards though from LP changes ?

[Clipse]

Stales are way better. Rejects too. Still seeing 15-20% discards though from LP changes ?

The last 2hrs had very frequent blocks so the DW results will look higher than they should simply because of the amount of times new blocks popped up while working on old work. DW should matter much its only work discarded from your queue and not actual completed worked that is being dropped

Was a crazy 2hrs of blocks o_0

I dont know yet how to cope with these luck spree periods but it doesnt happen often so for now we just have to deal with higher rejects/stales during such periods for the time being, other than that the post LP rejects look relatively in check now.

[Clipse]

Im looking into what caused the rejects over the night to shoot up to 2%, something went bellyup while I was sleeping.

For now I will just keep my sanity and make statement in OP that users could expect between 0.5%-3.5% stales/rejects until I can ensure everything is fixed, continously thinking the problems is gone but then having them pop up again is a huge headache right now. At the very least users would be making above 100% PPS and at the very best close to 105% PPS.

I need some Charlie Sheen tiger blood for more Winning.

[DutchBrat]

Now at 4.49% and rising..... lol

But might I add.... still making more than with any other pool, so no complaints 

[Clipse]

Now at 4.49% and rising..... lol

But might I add.... still making more than with any other pool, so no complaints  

Its still over small sample, ill get this to drop again over bigger sample, bare with me while I figure this out. Its not making much sense atm why this is happening at all and its not affecting everyone which is even stranger. There is still a bunch of users showing 0.5% or even lower stales and then there is a bunch of users with 3-4%, madness.

That said, I can try one more thing to try remedy the rejectdump for some of the users, will report back on that later tonight and see if it helps.

[Snowpea]

have you tried measuring the average latency across those users? maybe that's causing their rejects.

[DutchBrat]

I'm on 20ms.... Maybe I'm too fast 

[Clipse]

I'm on 20ms.... Maybe I'm too fast 

Im looking into the latency and right now there is a possibility its related to very low latency, not confirmed yet though.

[DutchBrat]

That's what the earlier issue was at BTCGuild.... miners with very low latency had problems...

According to Ckolivas it had to do with the Java backend of PoolservJ.... the only solution he suggested was to include --net-delay, but that didn't help any

I have no clue though how much the 'íntroduced' delay is with --net-delay

Weird thing is it runs so well for a few hours but the longer the pool is up, the bigger the issues get.... for some users....

As I said before... I don't like it, but even with 5% stales I'm still getting way more than at most pools.... and then there's always the upside of lower rejects 

[Clipse]

That's what the earlier issue was at BTCGuild.... miners with very low latency had problems...

According to Ckolivas it had to do with the Java backend of PoolservJ.... the only solution he suggested was to include --net-delay, but that didn't help any

I have no clue though how much the 'íntroduced' delay is with --net-delay

Weird thing is it runs so well for a few hours but the longer the pool is up, the bigger the issues get.... for some users....

As I said before... I don't like it, but even with 5% stales I'm still getting way more than at most pools.... and then there's always the upside of lower rejects  

Yeh its definitely not how I will leave things but on sidenote due to small samplesize your stales allready nearing <4% from earlier ~5% , Im just getting tired running in circles so really hope the solution I want to look into tonight will cope with the different miners and the issues experienced.