long deadline confirm times.
I've been mining for about 2 weeks now, and I've been having issues with long confirm times. I have about 25tb plotted and get a fair number of deadlines, but the deadlines themselves often take > 30 seconds to confirm with the pool (burst-team.us), and the program will often have to re-send the deadlines.
I've pinged the pool, and I get 100% confirm with about 55-60ms latency, and network quality is listed at 99-100%, has anyone else run into this issue?
Only once has it errored to a gateway timeout.
@bsattin1 Oh yeah part of the problem is the pool takes too high of DL's from everyone. They get swamped.
I almost lost a block and just happened to walk by my miner and restarted the scan to have the pool finally accept it and I won. Who knows how many times this happened where I didn't see it. Hate to think..
I have also lost blocks due to the DDoS attacks. We checked the actual log and our best deadlines were not propagating out to the network. Most likely some of this is still going on.
this may not help the DDoS, but do you know if there's a way to run the miner and tell it to re-send unconfirmed DL's every few seconds?
How much traffic does the deadline take? I would hope the deadlines could be sent in 1 packet...
@bsattin1 well you would think on an error like that it would attempt a send again. I do know sometimes it will not confirm but still show up on the pool so it goes both ways.