Update on shares issue
Posted: Thu Mar 05, 2015 8:33 am
As Chris reported, there is a bug in the server that sometimes causes spikes of hashrate. We corrected the invalid data and added 15% to everyone's balances to make sure that we were the most profitable pool during the days the issue occurred.
At the same time, we also implemented a failsafe that will kill the server in the case that the bug happens again, and then restart within five minutes. We have a solution to the problem but the solution introduced a regression and therefore we can't release it today until testing is complete.
Based on the occurrence of this bug in the past, there are likely to be two or three five-minute intervals where the server goes offline due to this new failsafe to prevent losses. If you add a "backup pool" to your mining software, then you should be able to direct your miners to the other pool for those times and return here when the pool comes back online. We are rushing a solution to the problem and hope to get a permanent fix in tonight. This weekend, we will be running a series of tests to make sure there aren't any additional issues and to improve stability.
If you were mining on March 2 when this problem started happening, then congratulations on your luck! We won't be requesting repayment or reducing balances for that day, so you probably earned about twice what you should have.
I'm going to ask Chris to post the following policy on balance errors to the help section:
At the same time, we also implemented a failsafe that will kill the server in the case that the bug happens again, and then restart within five minutes. We have a solution to the problem but the solution introduced a regression and therefore we can't release it today until testing is complete.
Based on the occurrence of this bug in the past, there are likely to be two or three five-minute intervals where the server goes offline due to this new failsafe to prevent losses. If you add a "backup pool" to your mining software, then you should be able to direct your miners to the other pool for those times and return here when the pool comes back online. We are rushing a solution to the problem and hope to get a permanent fix in tonight. This weekend, we will be running a series of tests to make sure there aren't any additional issues and to improve stability.
If you were mining on March 2 when this problem started happening, then congratulations on your luck! We won't be requesting repayment or reducing balances for that day, so you probably earned about twice what you should have.
I'm going to ask Chris to post the following policy on balance errors to the help section:
- If you are mining correctly and the pool's software incorrectly credits you with too much money, we will pay you the average earnings for that day plus an additional 15% as an apology for the times that the earnings were incorrect.
- If you are mining correctly and the pool's software incorrectly credits you with too little money, you will be paid the actual earnings of the day during that period.
- If the pool goes offline, then we will pay for the time the pool is online only (add a backup pool). This includes planned server maintenance.
- If there is ever a theft or hacking that causes us to lose funds, but you were mining correctly, then you will be paid the amount you are owed, even if the pool is forced to shut down afterward due to the losses. We retain additional funds for this purpose, and this is why we mandate payouts every two days, to reduce the amount available to steal.