Topic: [1200 TH] EMC: 0 Fee DGM. Anonymous PPS. US & EU servers. No Registration! - page 124. (Read 499602 times)

Pool down?

They crazy. Not a single stale is received by using EMC for a week!

There was a problem with block processing again last night, so two blocks are queued up and are about to be processed.  Different problem this time, though same result.  Slowly getting all the kinks worked out ... only two more steps to go.

*EDIT*

I should clarify that - by that, I mean there was a block sitting in an unconfirmed state which has now been confirmed and paid (two of them actually).

Confirmed.

Payouts should be fixed now... yet another casualty of the changes!   All payouts should be processed now.

Thank you Inaba for your fast support!! Really like it!! ^^

Yeah its broken. I will have it fixed soon and all payments will proccess.

Sorry about that... may be up to 45 minutes.

Hi!

 Anyone having issues with auto payment from EclipseMC in the latest hour or less?!
 My latest auto-payment never arrived. 40 minutes and still waiting...
 Just sent a PM to Inaba...  :\

Thanks,
Thiago

That makes sense--thanks for the info!

Jeez, I had to dig through a lot of different places to find out what the hell happened with those two blocks and why there appears to be an error.

Without getting into too much depth on the inner workings of how EMC works as far as block found detection goes, here's what happened:

Last night I made some changes to how block processing works.  I did this to accommodate a new method of block generation and wallet security - last night was a major change over from the old method to the new method (though we are still not done yet) - due to this, some of the code I was using was erroring out because it was not finding the data it wanted on the old system.  As you saw earlier today, there were two blocks that were stuck in an unconfirmed state - this was due to the same bit of code erroring out. 

Anyway, last night, after those two blocks actually confirmed, but the system did not register it, the bit of code that deals with block detection started working again (since it wasn't looking for the erroneous two blocks anymore), and then processed the two queued blocks that apparently existed at the time, so the 4 seconds, which now makes a bit of sense, is the amount of time it took to process the first block.  If you'll scroll back a couple pages, you'll see I reduced block processing time from 6 minutes to 4 - 6 seconds... well, there's your 4 seconds.

Anyway, long story short, the block was solved sometime after block 680, more specifically, 15 blocks afterwards, but the system did not register it until those two stuck blocks were no longer a factor to that bit of code.  The system then processed both blocks at once.

This does have a small effect on payouts in so far as it's considered a "short" round.  I have no way of determining how many shares were actually submitted to the false short round, though, so there's no way I can go back and retroactively fix it... the shares for the false short round were erroneously credited to the previous round, so no one lost submitted shares and you were paid "extra" for the previous round.  This will affect a small subset of people (mostly those with exceptionally low hashrates) and I want to sincerely apologize for this, as it was purely an error on my part and I am happy to entertain ways to correct it for those affected, but I can't think of any way to verify the shares for that given round, beyond what's already there.

I do not want to give out specific details as to what I have been working on lately as far as is related to this, but it's all related to security of the wallets and there had to be some major changes to the back end to accommodate it... this is unfortunately a casualty of the process.  I try to thoroughly test everything before making it live, but sometimes some things slip through and I work to correct them as quickly as possible, so I apologize to anyone who might have been affected, but I believe it is limited to a very small number of people and I will be glad to discuss with anyone affected individually on compensation.

Something seems odd in the Block History for block #680 (170290), though it's probably just something I'm misunderstanding.

It says block #680 finished at 2:52:10, and corresponds to block 170290.  Block #681 finished 4 seconds later at 2:52:14, and corresponds to block 170305.  Were there really 15 blocks found on the network in that time period?  Blockchain.info shows 170305 as found by EclipseMC (as expected), but 170290 is showing as found by Bitclockers.  Any idea what's going on?

Order of payout history should be fixed now.

I will be announcing the project within the next week or two.

I didn't get any shares on the 4 second block and very few on the other really short blocks. That along with the increased pool size probably caused it.

WOW! Our luck sure has come back after those horrible blocks. This variance is INSANE! Broke our lowest share count 2x last night! 753 in 4 seconds then an astounding 600 shares! Holy crap, I think the record previous was something like 2500 or so shares.

Also, as far as the DGM stuff, I think of it in the capacitance sense, but unlike prop where the rewards is 0 after the current block, the capacitor simply allows you to potentially get rewarded across several blocks, till its drained.

Whats this new feature Inaba? Tell us! Tell us! =P haha
Anyways, this pool is sure exciting!!

Lets hope none of those quick ones are invalids!
I still need to find my first block damnit! LOL

OH HEY! Inaba, can you look into the payout transaction history list...the BTC payments , for me are no longer in order by date...they are all screwy, since a few days ago.

Should be fixed now.  I have been making some major change behind the scenes to beef up security and pave the way for the new project/addition to the pool that I will be announcing soon and I inadvertently took out the validation processing for one particular server last night.  

Should be good to go now though, and going forward.  Still more changes to come, but they are mostly behind the scenes so nothing should be noticeable for awhile yet.

Quite the run of good luck.

As far as scamcoins, I'm not saying no, but it's unlikely we'll be doing merged mining for any of the scamcoins in the near future, unless I make some major break through in merged mining... NMC is already a giagantic pain to deal with.

That's happened once before.  I'm sure Inaba will fix it real soon. He's good like that.

I can't believe I solved 3 blocks in under 2 hours. That's highly unusual at this difficulty given my hashing power.

Wow!!! The luck the last six hours is amazing.

What is up with blocks 170170 and 170203 not confirming though??? 170170 was found 27+ hours ago.

You... got... me....

The more different blockchains you mine, the more longpolls you are going to get, which is no small issue..........

There's also some variance (caused by luck in future finding of blocks), not just delay. Of course, over the long run it should not be significant.

That's all you can say! That's all you can say, that's all you can say...