[ Bonus PPS Pool - 130% PPS - OPEN ]

[Clipse]

UPDATE :

Ok it looks like I resolved the stack overflowing issues but will continue to monitor it. Please connect back so I can compare old results with new results using the same hashrate for comparison, thanks.

[1714109141]

1714109141

[1714109141]

1714109141

[1714109141]

1714109141

[DutchBrat]

Sent PM

2.5 - 3 GH ready to go

[Clipse]

Sent PM

2.5 - 3 GH ready to go

Yup, replied via email.

[MrTeal]

Hey Clipse, just FYI I think there might be something weird going on with how the hash rate it calculated. Not that I don't wish it was true, but I'm sending you about 1.7GH/s, and the latest 5min average is showing 11GH/s. There's nothing out of the ordinary in the number of shares submitted per minute in my cgminer log, so something else might be up there. The counting of the shares themselves don't seem to be having a problem, just the average hashrate.

[Clipse]

Hey Clipse, just FYI I think there might be something weird going on with how the hash rate it calculated. Not that I don't wish it was true, but I'm sending you about 1.7GH/s, and the latest 5min average is showing 11GH/s. There's nothing out of the ordinary in the number of shares submitted per minute in my cgminer log, so something else might be up there. The counting of the shares themselves don't seem to be having a problem, just the average hashrate.

Just a display bug on import resets(silly way of doing it atm but the next webpage rollout will be alot better just not getting to finalise it yet), it fixes itself after a 5-10min period of between 1-2 extra imports. I had to reboot the webserver it was sluggish as hell and made some sql/apache changes to handle request better.

The reason the hasrate ends up like that on import restart is because the calculation thinks you did xShares within import timeframe and not across 5mins as it would calculate after initial run.

[Clipse]

UPDATE :

I had to drop gpumax connections for some of the users who were directing to us from gpumax via private offline pool. I have no idea just yet why gpumax traffic is flooding the pool and it seems to spike randomly/not consistently thus was a real pain in the neck to pinpoint exactly whats causing the pool tcp connections to flood.

That said, pool is stable now and shouldnt crash anymore/die randomly when I am asleep however if you were directing to us via gpumax then you would have to set us directly if you want to mine with us while we figure out what is up with gpumax connection.

[DutchBrat]

Sent PM

2.5 - 3 GH ready to go

Yup, replied via email.

Never received it... 

[Clipse]

Sent PM

2.5 - 3 GH ready to go

Yup, replied via email.

Never received it...  

Arg my smtp server is failing me bigtime, let me send again.

edit: just send the mail again, please pm me once you received.

[Clipse]

UPDATE :

Website and email down for next 1-2hrs, mainly trying to sort out why the stats importing is so slow and hogging the system and also resolving my email debacle so dont panic if you cant access the website in this timeframe.

Good news though it seems I have resolved something(patent pending) and stales have plummeted nicely over the last 2hrs.

[DutchBrat]

Was running smoothly there for about an hour or 2, but in the last few mins communication failures followed by a batch of Rejected shares flew by

Long Poll seemed to work fine there for a bit....

[Clipse]

Was running smoothly there for about an hour or 2, but in the last few mins communication failures followed by a batch of Rejected shares flew by

Long Poll seemed to work fine there for a bit....

My bad, the recent blooper was me fatfingering. Its back up and hopefully it will be smooth without having to do any more changes to it.

[MrTeal]

Whatever you did, it's working. I restarted a couple miners to reset the stats in CGminer, and after 1000 shares I only see 4 rejects. So far it seems to be an order of magnitude better. Thanks Clipse.

[Clipse]

Whatever you did, it's working. I restarted a couple miners to reset the stats in CGminer, and after 1000 shares I only see 4 rejects. So far it seems to be an order of magnitude better. Thanks Clipse.

I must have cut the correct wire

Since the most recent restart across 11 LP's (thats the big benchmark point) the pool stale/reject rate combined is 0.08%

[DutchBrat]

Strange thing,

Anubis reports 4.9% stales (data from cgminer) and your pool website only reports 3.42% on 33.5K shares, which is a big difference in payout.

1146 on your site vs 1598 in cgminer.... can't be that big of a difference rejected shares leaking to my backup pool

I am more than happy to take the lower reject % which is reported by you, obviously  , but are you sure all is being calculated proper on your end ?

There also is a difference in total reported shares 34.900 in cgminer vs 33.500 on your side, which obviously is not in my favour....

Reported hashrate is about the same 2600 MH/s

Anyone else have this problem ?

P.S. sorry to be a pita, but I again / still see lots of rejects just after a longpoll  

That's why it's called 'Beta' !

[Clipse]

You sure you didnt check the webstats before it updated a new cycle ?

Its not realtime and will allways atleast for the most part be ~5mins behind your own stats.

ps. im back at the drawing board since the stale/reject problems obviously not completely gone yet, shrug.

[Clipse]

Just so you dont wonder why stats not updating , im moving the old shares to payment history and testing few more things, your new shares will be updated right after that.

Just a short while, sorry.

EDIT: Trying one more thing, if it still mess up like it is atm I will take pool down until I can install a proper patch hopefully within a few hours.

EDIT: Taking pool down for atleast a couple of hours, pointless to keep taking it up and down atm just to test, sorry about this.


EDIT-LATEST: Pool is up and receiving work, just monitoring first to see over next few blocks if it goes to hell again or not, in the meantime Im moving the old shares to payment history and will import the new shares(like those being submitted atm) right after this test.

[Clipse]

UPDATE :

Pool is running stable for now, need alot more blocks to hit to see how it performs after a ton of getwork push at LP. Webstats db cleaned up so that is atleast faster to access now. Anyhow lets see how this run goes.

[Zubilica]

UPDATE :

Pool is running stable for now, need alot more blocks to hit to see how it performs after a ton of getwork push at LP. Webstats db cleaned up so that is atleast faster to access now. Anyhow lets see how this run goes.

I have try it for a few hours , I'm disappointed. Donno if my connection is to blame or the pool , but most of the time my workers are not working.

[Clipse]

UPDATE :

Pool is running stable for now, need alot more blocks to hit to see how it performs after a ton of getwork push at LP. Webstats db cleaned up so that is atleast faster to access now. Anyhow lets see how this run goes.

I have try it for a few hours , I'm disappointed. Donno if my connection is to blame or the pool , but most of the time my workers are not working.

Pool has gone down a while ago, sorry been hard at it working on solution so for now I will keep it offline and will setting load balanced servers that should be a reasonable workaround for now and future load.

Atleast I can say I know whats causing it, its just the solution that I am working on and thats not so easy.

Its been a tough day, I need some sleep now and hopefully before end of today all will be well and running for good

[Zubilica]

I will try it on when is stable