[ANN][LTC][Pool][PPLNS][STRATUM] - ltc.kattare.com - burnside's Mining Pool

[msm595]

As of right now, the pool efficiency is ~75%. Is there something that could be done to improve this? Does it have to do with share difficulties? I'm willing to dedicate some of my time to helping kattare add stratum if that is what is needed. Since it is based on open-source software, it shouldn't be too difficult.

[1713873938]

1713873938

[1713873938]

1713873938

[1713873938]

1713873938

[zenrith]

Adding Stratum to the pool would most likely reduce stale shares to less than 5%.

Granted, if the stales are being caused by DDOS attacks.... then it really doesn't matter if we have a dedicated glass tube laser comms device between us and the pool... the pool simply won't respond.   

[burnside]

As of right now, the pool efficiency is ~75%. Is there something that could be done to improve this? Does it have to do with share difficulties? I'm willing to dedicate some of my time to helping kattare add stratum if that is what is needed. Since it is based on open-source software, it shouldn't be too difficult.

On that particular number (overall pool efficiency) it's taking into account all the rookie bitcoin miners that point at the pool using sha256 and other bad configs.  You can get a feel for how many of these we get by looking at the "over 90% stales this round" display.  At one point today I saw that number at over 120,000 for someone.  Needless to say, that cuts into the efficiency pretty good.  Only way to really fix it I think would be to temp-ban people that run like that for more than an hour or something.

[burnside]

Adding Stratum to the pool would most likely reduce stale shares to less than 5%.

Granted, if the stales are being caused by DDOS attacks.... then it really doesn't matter if we have a dedicated glass tube laser comms device between us and the pool... the pool simply won't respond.   

It's a combination of:

- DDoS attacks themselves
- The DDoS service (which we have to relay all traffic through, so even when we're not under DDoS there is an efficiency hit.)
- Bad configs (sha256 miners) sucking up resources.

Without all of that in the mix the pool traditionally had ~3% stales.  Stratum will definitely help with that when we roll it out.

Cheers.

[chrcoe01]

stratum will make me happy

[seamusmcaffrey]

Question... why under pool stats where it shows "next difficulty estimate" does it always say "-75" ? this number is neither correct nor meaningful, really... Is this just a reject error from website code somewhere?

[burnside]

Question... why under pool stats where it shows "next difficulty estimate" does it always say "-75" ? this number is neither correct nor meaningful, really... Is this just a reject error from website code somewhere?

Nah, it's an error from litecoind.  That number is based on the network hashrate, and right now this is what litecoind gives for the networkhashps:

# litecoind getmininginfo
{
    "blocks" : 330544,
    "currentblocksize" : 0,
    "currentblocktx" : 0,
    "difficulty" : 205.77652398,
    "errors" : "",
    "generate" : false,
    "genproclimit" : -1,
    "hashespersec" : 0,
    "networkhashps" : -2147483648,
    "pooledtx" : 20,
    "testnet" : false
}

It'll probably get fixed in a future litecoind release.  The code for the site is fine though.

Cheers.

[seamusmcaffrey]

alright, cool, just curious really. I can get that info from any number of places

[talon146]

Burnside. Not sure if you've answered this previously but I'd like to know of your plans, if any to offer stratum support. I currently am losing around 30% to stales across 4 cards with what seems to be the best config.

[glendall]

Glad I'm not the only with high stales. Thought it might be just my config!

I'm going between 12% - 26% stale, which seems pretty crazy (though I'm a new miner though, haven't got a feel for what is normal).  1/4 of my production lost to stales is too much.

I'm going to setup automatic pool hopping next chance I get to mitigate high traffic / downtime / lag...

edit: been watching last 15 mins.  Been amazingly all over the place.  Steady at 15% for most of my cards and then whammo, goes nuts, even had a stale of 44% for one poll.

That kind of lost efficiency really gets me right here [points to heart]  

edit 2: its possible the high stale rate is in part due to over estimation of hashing rates. I don't think my 7950 is doing 740-odd k/hashes as the portal reports, so perhaps it is a software reporting error which would account for like 8% of the stale value.. .

[creativex]

WOW. Stales are now under 3% for me. Awesome sauce burnside! Great work!

[talon146]

WOW. Stales are now under 3% for me. Awesome sauce burnside! Great work!

What cards/miners/config are you running?

[BKM]

WOW. Stales are now under 3% for me. Awesome sauce burnside! Great work!

Just started with you a couple of hours ago and am seeing around 18% stales using cgminer. I'll be back when you sort stratum out.

[creativex]

WOW. Stales are now under 3% for me. Awesome sauce burnside! Great work!

What cards/miners/config are you running?

A hodge + podge assortment of cards + CGMiner.

MSI 6950@6970
Powercolor 6870
HIS 5830
Asus 6850
Sapphire 6770 #1
Sapphire 6770 #2
Sapphire 7770 #1
Sapphire 7770 #2

Note: As soon as I posted that 3% stale rate above my stale rate rose to around 7%. I should've shut up, but still happy.

[Strahlex]

How can I remove my account?

[dyseac]

we dead again?? why is this so hard

[mattz0r]

How come the site is down

Only started mining last night!

[chrcoe01]

Mining works, but website is down....  :S

That's what I thought too, but the website was up for a brief time this morning and NONE of the work I did yesterday showed up. Also zero auto payouts and I have it set to 1LTC, there should have been several by now. It appears to be working fine, but I think the shares are going into the void.

Sorry about the late update.  We are definitely under DDoS.  I had shifted the site away from the DDoS service to see if stales improved and within a few hours we were under attack and down again.

The attack caused some crons to fail.  The crons lock and unlock as they start and finish so when the ddos interrupted them they stopped running until I manually cleared the locks.  Once I did that, all the block payouts caught up and paid out, though because of the ddos, payouts today sucked anyway.

I did make some progress today toward shifting to the new server.  I think by tomorrow i'll be able to move all the DB services over to it.

A couple people have pointed me at the elgius pool software for stratum.  It looks promising. I'll give it a whirl soon!

Cheers.

mattz0r, probably still being ddos'd see above quote.  It would be nice if it were down because of stratum implementation though 

[burnside]

Sorry about the late update.  We are definitely under DDoS.  I had shifted the site away from the DDoS service to see if stales improved and within a few hours we were under attack and down again.

The attack caused some crons to fail.  The crons lock and unlock as they start and finish so when the ddos interrupted them they stopped running until I manually cleared the locks.  Once I did that, all the block payouts caught up and paid out, though because of the ddos, payouts today sucked anyway.

I did make some progress today toward shifting to the new server.  I think by tomorrow i'll be able to move all the DB services over to it.

A couple people have pointed me at the elgius pool software for stratum.  It looks promising. I'll give it a whirl soon!

Cheers.

mattz0r, probably still being ddos'd see above quote.  It would be nice if it were down because of stratum implementation though 

Yes, still under DDoS.  (~3.5Gb/s)

I did make progress last night on the stratum side of things.  I have a test stratum server up and running, I need to convert some of my reaper miners to cgminer to test it now. It's fairly complicated though, I'm not going to get my hopes up too much quite yet. 

[mattz0r]

Thanks guys,

I'm still new to all this, can't get my GPU miner working properly though :/

But, I've only got a single GTX260 at the moment...

So resorting to CPU mining on my AMD 6 core, and a server which is doing nothing atm with an E3-1230v2

EDIT:

This is what I get on reaper

http://screencloud.net/img/screenshots/6b917fe82880f64138d7ee389dfcb0d3.png

Never get any shares, yet my CPU's that run slower... get more shares :/