Status as of Tuesday, May 30, 2017

Steve Sokolowski avatar
Steve Sokolowski
Archived posts: 4559
Joined: Aug 27, 2014

Status as of Tuesday, May 30, 2017

by Steve Sokolowski · · Post #6018

Just a brief note for today:
  • Chris was able to get the mining server moved to the new processor last night, the first of the three performance improvements. CPU load has decreased to an average of 35%, down from 60% before, and there have been no spikes to 100% since. This should allow us to handle up to 4000 workers even before the software improvements go out and should eliminate periods of rejected shares.
  • We'll be releasing the "non-locking" payouts update tonight, which may delay payouts while Chris confirms its correctness. This update should prevent delayed shares in the early morning. There will be no impact on mining. If this update is unsuccessful, we'll pay out with the old code and try again tomorrow.
  • The WAMP server crashed with an "exhausted all open file descriptors" error earlier today. We're investigating that and hope to find out more information soon. This isn't our code, so troubleshooting it is difficult.
tmopar
Archived posts: 60
Joined: Apr 16, 2017

Re: Status as of Tuesday, May 30, 2017

by tmopar · · Post #6020

[quote="Steve Sokolowski"]Just a brief note for today:
  • Chris was able to get the mining server moved to the new processor last night, the first of the three performance improvements. CPU load has decreased to an average of 35%, down from 60% before, and there have been no spikes to 100% since. This should allow us to handle up to 4000 workers even before the software improvements go out and should eliminate periods of rejected shares.
  • We'll be releasing the "non-locking" payouts update tonight, which may delay payouts while Chris confirms its correctness. This update should prevent delayed shares in the early morning. There will be no impact on mining. If this update is unsuccessful, we'll pay out with the old code and try again tomorrow.

    What processor did you guys end up with?

    Also is this the non-locking fix I described earlier in the other thread?

    Thanks!
    TMopar