Restarted server

[Steve Sokolowski]

It's pretty bad.
So I ran 4 L3+ machines on 1 static coin and got 90% of my hash power to be recognized and paid out.
The other 4 L3+ machines I ran on auto switching. I'm not sure they ever recorded shares. just constantly "no valid shares yet"

The profits were over 4:1

As it is now I can pick a coin like bitconnect and sit on it all day and make 4 TIMES more than the auto switching which isnt giving credit for even a third of the hashrates I put on this system. It's great I can connect in almost all of the time now! And now auto switching takes soo long I never get shares recognized for hours on end...AND it doesnt cause a failover so i could ever go to a failover pool and get paid for the work Im not getting credit for.


Also, I put some miners on SOLO mining, and when the coin wasn't available it pushed me to auto-switch coin mode.
I thought if the coin you were solo mining wasnt up it listed it as DEAD so you could go to another failover option.

I hate to say it but maybe revive the code from 3 weeks and try it again, cause it looks like the last 3 or 4 fixes have put more holes in the boat than they have fixed.

Since last night's release by Chris, have you experienced any additional instances of miners getting stuck on coins?

[hashingpro]

I havent seen it hang up on a coin and not report anything, but I still think bitconnect coin had something to do with it.
Like that network had instable issues or was up down up down up down, intermittant and it was just causing everyone to hang on it.
It never kicked, never went to another coin, and the profits never went up for 10-20 mins until after the reset.
I havent seen any hangups, but I havent seen any/much of BCC coin so far today on the site to get hung up on.

Either BCC are down/offline or we are blocking them now as a way around the hangups in which case its putting a hit on the daily profits not having that coin up to mine so far. But if the choices are mine a lesser coin and get credit or mine a high coin and get stuck for hours with no credit, ill take the lower coins with a payout.

might need another day to check on it once BCC is up, see if thats the one everyone gets stuck on.
I know it has a huge hashing capacity and a long block time but 20 mins no hash, no credits, still calculating is a bit too long i think.

[Steve Sokolowski]

I havent seen it hang up on a coin and not report anything, but I still think bitconnect coin had something to do with it.
Like that network had instable issues or was up down up down up down, intermittant and it was just causing everyone to hang on it.
It never kicked, never went to another coin, and the profits never went up for 10-20 mins until after the reset.
I havent seen any hangups, but I havent seen any/much of BCC coin so far today on the site to get hung up on.

Either BCC are down/offline or we are blocking them now as a way around the hangups in which case its putting a hit on the daily profits not having that coin up to mine so far. But if the choices are mine a lesser coin and get credit or mine a high coin and get stuck for hours with no credit, ill take the lower coins with a payout.

might need another day to check on it once BCC is up, see if thats the one everyone gets stuck on.
I know it has a huge hashing capacity and a long block time but 20 mins no hash, no credits, still calculating is a bit too long i think.

I don't think this issue has anything to do with bitconnectcoin. The problem was some code that had been added to not reassign some miners in cases where we thought it was safe to do so. It turns out that the code was incorrect and those miners were never reassigned in any case, even when it was indeed safe.

That should solve the problem with miners getting stuck. There may be other issues that are separate from that one.

[Kfedorek]

Tha auto switching methods today were really bad...i had the worst day of profit(as in coins mined) in a long time today. Should i switch it to mine only LTC?

[hashingpro]

With one of the exchanges offline they use for pricing they cant mine certain coins as they have no clue what they are worth vs others they do know.
And they cant get stuck with coins holding them waiting to see what the prices will be once it comes back up before they can exchange and do a payout.

I'm sure the payouts will be lower with a lot of places today not just PH.

If BCC goes offline and nicehash cant sell power to mine it, im sure their profits drop.

[Steve Sokolowski]

With one of the exchanges offline they use for pricing they cant mine certain coins as they have no clue what they are worth vs others they do know.
And they cant get stuck with coins holding them waiting to see what the prices will be once it comes back up before they can exchange and do a payout.

I'm sure the payouts will be lower with a lot of places today not just PH.

If BCC goes offline and nicehash cant sell power to mine it, im sure their profits drop.

Chris says that three exchanges are now offline, making things even worse.

[GregoryGHarding]

with nova, they are being attacked via their API ports, not sure about the other two, but nova is working on a fix.

[mycide]

That was a huge drop 40% down over night.
But i think something happend, my hashrate history show pool dropping my registered hashrate from 10 AM yesterday from 2.6 to 1.5gh/s
earnings reflect that drop. hashrate history shows me as still only having 1.5gh but live worker and antminers ui shows all working fine, so missing shares again?

[piet]

so missing shares again?

Totally agree ! something is wrong

[Steve Sokolowski]

There are definitely not any issues with missed shares - that was investigated last night and has not occurred since September 10 when we made a fix for that.

It's possible there is some other issue, but that issue would be caused by hashrate recording, not by missed shares, because profit is at breakeven today. If money had actually been lost to miners, then we would show a large profit, which isn't the case.

Chris can investigate again later today when he awakens, and I can look into this problem in 12 hours.