One thing I noticed yesterday was that profit was up by about 6 bitcoins, which was a ridiculous figure. That prompted me to tell Chris to look into the issue. Because we don't have the ticketing system set up yet, our communication is not yet ideal, but since there was a loss of 2 bitcoins today, I assume that he paid back at least 2 from that huge surplus plus all of yesterday's profit.AvPro wrote:Steve Sokolowski wrote:What about the issues that Vanessa was talking about yesterday where she and FRISKIE believed that balances were not being credited fairly? Do you believe that's true, or do you think that bitconnectcoins caused lower profitability that may have been mistaken for that?AvPro wrote:No worries, I'm happy its back online and already our profit is back on par with Z and H2C pools
Tricky one, for a couple reasons because there's a triple prong of reasons that arose all at the same time
i) downtime of BCC, a very profitable coin reduced the profit figure for PH against other pools and consequently dropped the earnings for a period of at least 15hours by my count. I think this is the major reason for losses on 23 June. Can you clarify with this type of daemon error - can the live profit figure be reported incorrectly? It's not trying to calculate it using the last available diff it obtained from the BCC blockchain as part of the profit algorithm? Hence, throwing off the figure.
ii) addition of the DDOS protection has (extra latency perhaps) pushed up the "negative delta" for any users of NH. Last time I used it i noticed it was about 3% more delta, another reason for a reduction of profit if anyone using NH and comparing apples to apples from before DDOS protection.
iii) i think some investigation into balances being credited is merited, I've noticed a reduction recently but because the hashrate charts are likely correct I don't exactly have a good evidence. Now that BCC is enabled I will compare my earnings over the next 24 hours to an average day 2 weeks ago and will send a support email if it's way out.
In summary, perhaps lets see how everyone earnings are like over next 48hours and see if shares are still out.
The latency increase is about 15ms, but I don't think that it should increase "delta." "Delta" is a measure of work restarts. Stale shares might go up. Fortunately, we are having a meeting with Comcast on Monday and we hope to eliminate all this next week by getting the DDoS protection on the same network. But it's also possible that the WAMP server, which was overloading the connection for months (it just wasn't noticed until we examined usage after the attacks) was starving other services for bandwidth. Could you pay attention to the NiceHash figures starting at midnight?
The live profitability figure is the average earnings being recorded to the database, after mining. If something is wrong in mining itself, it will not affect that figure. The only way that it would be affected is if something other than the calculated earnings are being reported to the database. Chris spent a few hours last night and was not able to determine why that would be the case, but the timing of the complaints matches the time that we made changes to record fewer shares to the system. I need to talk with Chris to find out whether he fixed something and how that was related to the reduction in profit, because a lot of money went out to miners since last night for some reason and perhaps something changed.