[~1000 GH/sec] BTC Guild - 0% Fee Pool, LP, SSL, Full Precision, and More

[Miner-TE]

I'll take this opportunity to update my miners.  No worries.

[1715190365]

1715190365

[Miner-TE]

Of course .... it's back up now....  Oh well..

[tripper22]

Standing by.    Of course, you feel no pressure..  

Getting a huge flood right now, adding debug log printouts into the code to find out exactly where the crash is coming from.  It's a memory leak issue, looks like it's related to that hashing I mentioned previously.  Also checking network logs to see if this flood is legit or not (although it looks like something slush mentioned with his pool over the weekend).

UPDATE:  Found out which worker is triggering the crash, now trying to figure out why.
UPDATE 2:  Back up and not crashing at the moment.  Gotta package some items that sold today for shipment, we'll see if it's stable or not soon.  Sorry for the inconvenience, but some bugs just can't be found until you open up for the whole world to play with.

Hey eleuthria keep up the great work. I think it's going to be an awesome pool when all the bugs are worked out. I hope it wasn't one of my miners that caused the crash...

[SteveB]

Of course, the pool went down just minutes after I had to step away to do something else and came back just as I returned.
But I guess that is to be expected with a new pool.
Keep up the great work!

[eleuthria]

Thank you all for the continued support.  As Sukrim stated, it is a great idea to have a second miner running low priority on another pool in case BTC Guild crashes in this early period.

[Nythain]

Switching over now for a day to test out. Couldn't have happened at a better time since almost all the other major pools right now have been having some wonky issues too.

[eleuthria]

One particular worker has been causing the crashes.  I've blocked them from the pool temporarily while I try to figure out what makes them so special.

[SteveB]

I hope it is not mine. I now get a lot of:

Code:

09/05/2011 22:19:33, long poll exception:
Traceback (most recent call last):
  File "BitcoinMiner.pyo", line 259, in longPollThread
  File "BitcoinMiner.pyo", line 221, in request
  File "httplib.pyo", line 898, in request
  File "httplib.pyo", line 935, in _send_request
  File "httplib.pyo", line 892, in endheaders
  File "httplib.pyo", line 764, in _send_output
  File "httplib.pyo", line 723, in send
  File "httplib.pyo", line 704, in connect
  File "socket.pyo", line 514, in create_connection
error: [Errno 10061] No connection could be made because the target machine actively refused it

[eleuthria]

I sent a message to the owner of the worker, trying to find out what miner they're using and if there's anything strange in the settings.

[Miner-TE]

I'm also having issues.

alot of "failed to connect ..  retrying."  and rejected results.

using latest phoenix.

[eleuthria]

Are the errors occuring RIGHT NOW (as of about the last 30 seconds)?  My miners have resumed working since the block was put in place on the bad worker.

[SteveB]

It seems to be working again.

And I haven't received any message so I guess it wasn't mine that caused the problem.

[Miner-TE]

Last "failed to connect" was at 21:22:04  PST.   Appears to be resolved.

[SteveB]

Last "failed to connect" was at 21:22:04  PST.   Appears to be resolved.

That was the same time my miner started working again too.

[SteveB]

And thanks to Sukrim for the second low priority miner tip! It failed over to deepbit PPS beautifully.

[Nythain]

Still getting problems communicating with bitcoin prc errors sometime around 23:35~23:37 CST

[SteveB]

I was getting errors again too:
error: [Errno 10054] An existing connection was forcibly closed by the  remote host

But now it works again.

[eleuthria]

I need to make a button that auto-posts:  Pool crashed, bandaid applied and fingers are crossed!

My previous fix looked logical to me.  But I put it about 4 lines past the function call that crashed the process.  My brain/eyes didn't want me to read it properly after working about 18 hours straight .

UPDATE:  Got a response from the person who has the bad miner client crashing the poor pool.  Should be resolved shortly.

[Nythain]

Yeah, seems to have stopped at around 23:39 CST. Given the small nature of it, guess it was eleuthria fixing some stuff or rebooting something or whatnot. Either way, gonna stick with it. I really like the idea of this pool and its operation. I've already set my donation % to 3 just based on the overall awesomeness of the transparency and the site and eleuthria's openness to ideas and suggestions. Good work!

whoops, spoke too soon, acting back up at around 23:44 CST

[eleuthria]

I've applied a few more bandaids, including an auto restart script for the night, so if the pool crashes it's back up in about 1/10th a second.  I've got it creating a dump after each crash so when I wake up I can try to fix it while I'm at work.

It's still accepting shares, and submitting PROOF-OF-WORK to bitcoind, but you may lose some mining time during the brief pool restarts.  Consider this a 'caution' sign.  Hopefully after a good night's sleep I'll be able revive my old C debugging skills that have been dormant for so many years.