solve stats don't add up

[ericecook]

"MonitorAuxNetwork_19529360": {
    "block_stats": {
      "accepts": 2, 
      "last_solve_hash": "4e3676e541aca1d02d7b56d85c20c1046c150af05dd63b6c3e59a77ad762fd89", 
      "last_solve_height": null, 
      "last_solve_time": null, 
      "last_solve_worker": null, 
      "rejects": 0, 
      "solves": 3, 
      "stale": 0
    }, 

[icook]

Which server was this? I'll need to look at the logs for all block submissions to track it down.

[ericecook]

I don't recall. I'll provide more info + do some digging if/when I see it again.

[ericecook]

I haven't seen this issue again, so I'm closing it unless I can get more info/confirmation

[icook]

Right, I actually think I figured out what you saw the other day. Basically, solves get incremented at the beginning of block submission, and stale/reject gets incremented at the end. If it's a stale block it will try 5 times over 5 seconds, giving you a 5 second window of inconsistency.

[ericecook]

It appears to have happened again - and just to be sure I waited a couple minutes and refreshed

This is on the LTC vardiff server. I'll look into it more tonight if I have a chance

  "MonitorAuxNetwork_19525264": {
    "block_stats": {
      "accepts": 54, 
      "last_solve_hash": "af7ae9f25eadc9eb30df43f4e1bf7b282315ea06d023b0889a935ac0c20b2d61", 
      "last_solve_height": null, 
      "last_solve_time": null, 
      "last_solve_worker": null, 
      "rejects": 0, 
      "solves": 55, 
      "stale": 0
    }, 
    "currency": "DOGE", 
    "current_net": {
      "difficulty": 15281.465631037563, 
      "height": 397175, 
      "last_block": 1412080445.506851
    }, 
    "last_signal": 1412080445.463508
  }, 

[ericecook]

Doing some digging.... It appears this only happens on the Dogecoin daemon.

My best guess is that occasionally the block submit to the Dogecoin server is causing an uncaught exception, thus breaking out of the while loop and avoiding marking the block as stale or rejected - although I wasn't able to find anything in the logs to validate this theory.

[icook]

If that's the case the exception should be in the logs somewhere.