@poolwaffle
Mooncoin wallet upgrade to 1.862.1 to fix exploit of KGW ... PLZ upgrade to new wallet, hardfork at block 67500 I think it was. From what I read this exploit affects all KGW coins.
I appreciate the notice, but we haven't been mining mooncoin for a while, they've been delisted from most major exchanges. They're currently disabled, if something changes I'll upgrade the daemon, resync, and enable
1) Its been a whole 48hrs, and I've been on IRC a decent bit of that
I was on a 10hr flight, and had some sleep-catchup to do.
2) About half our hashrate comes from one user (see any of the 30+ posts in this thread), he can choose which pool he wants to be on.
3) Yep, we had a really shitty luck-day. Really can't say much more than that...
Onward to the more exciting stuff:
I've added a few new coins, and some logic for handling new coins. I've noticed in the past, as soon as we add a new coin, other pools jump on it almost immediately, as we publish data about which coins we're mining. And as soon as all the other pools jump on it, that coin gets a bit diluted, and while it still might be profitable, its
less profitable. So, in an attempt to squeeze the most out of "fresh" coins, they'll be listed across the site as "hidden" for the first [undetermined] number of days. As soon as their profitability drops (natural or added to other pools), I'll set them to non-hidden, and all the data will be available. We've always tried to provide the most transparency of all of the pools, and while this hides some information in the short term, ideally it boosts profitability during that time, with full transparency afterwards. As of now, we're mining 2 new "hidden" coins
We've added support for an (almost) fully-automated exchange, which should let us mine new coins (see above), and also split some of our selling of major coins (LTC, DOGE, etc) to multiple exchanges, to get the best prices. While each exchange is slightly different, almost none of them allow for automated withdrawals (understandable), so for now we need to manually withdraw funds, which can delay/impact payments. The good news is that we're doing a relatively small amount of our exchanging through the new exchange, and the bump to 2% fees lets us be significantly more resistant to any delays (hooray!).
We've tweaked our switcher a good bit. For smaller coins (read: under ~20 difficulty), we switch segments of the pool for mining, rather than the entire pool in an attempt both to not blow-out the coin, and to not compete with ourselves. A good example of this, block-propagation time from USEast to SEAsia is about 600ms, and if we switched the whole pool (currently 24ghs) to a small coin (example 15 difficulty), we have an
average block time of 2500ms. We'd be finding blocks on our USE cluster before we knew we found one from our SEA cluster. Switching parts of the pool avoids this problem by limiting our hashrate, as well as limiting competing clusters. The first piece of that however (limiting hashrate) is based on selecting for
ideal hashrate per coin, saying "I want X hashrate on coin1, Y hashrate on coin2". Because difficulties of coins are always changing (KGW, et al), and hashrates per cluster/server are always changing, we could get in situations where it would "thrash" (keep switching between two) a set of servers between two coins. This should be fixed in our new switcher.
Slightly long-term (maybe a week or so?), we're spooling up a new box for our database, which should speed it up by an absurd amount (we've outgrown our current master DB). The cutover to the new database will entail a small amount of downtime (up to an hour), and will be announced significantly ahead of time (a day or so), both here, and on the header of the website.
Sorry for the long, drawn-out post, but figured I'd explain everything that went into the changes, rather than leave something out