I just lost a days of mining.

[Doggo]

I just noticed that my miner that was working on a ethereum pplns pool over the last few days just got kicked off while ethereum was under some sort of errors, then once it let me back on, it reset days of work. which was about 30% of the 2 eth block. I wanna know whats going on?

[Steve Sokolowski]

I just noticed that my miner that was working on a ethereum pplns pool over the last few days just got kicked off while ethereum was under some sort of errors, then once it let me back on, it reset days of work. which was about 30% of the 2 eth block. I wanna know whats going on?

I'm not aware of any issues with ETH that were persistent over several days.

This seems like it could be related to your specific configuration. Please submit a support ticket with your username and we'll be glad to assist!

[Doggo]

It wasn't a persistent issue, this only happened after I was unable to mine ethereum today.

[Steve Sokolowski]

This issue was also reported by other customers, but it wasn't caused by yesterday's system downtime. I assigned it to Steven and Chris for research.

This is a complex issue that is expected to require significant investigation, so I apologize for how long it is expected to take to get an answer for you.

[Doggo]

Alright, thanks. I'm all open ears if you guys find anything.

[Steve Sokolowski]

This issue has been resolved, and Chris released the changes.

It turns out that the problem was that the Postgres database doesn't enforce NULLs in unique constraints, so the system kept inserting repeated "first round" PPLNS rows every time it was restarted. We restarted the system earlier in the week, and while the restart didn't cause the problem, it was the first restart since PPLNS mining has been widely used.

Chris added together all of the rows and mining has resumed as normal. No money was lost.