[CLOSED] BTC Guild - Pays TxFees+NMC, Stratum, VarDiff, Private Servers

[os2sam]

Unless I see a significant number of requests, there will not be a 'no vote' option.

I have yet to hear a valid argument for the need to implement this.  I'm not saying I'm for or against the modifying the block chain.  I'm just saying nobody is putting up a good reason to do it in the first place.

So I guess I would vote for a server which has a no vote capability.  Is that convoluted or what

Sam

[1713512171]

1713512171

[allinvain]

Pool is down!!! mine1.btcguild.com to be specific.

[eleuthria]

Yes, mine1.btcguild.com is down, and it looks like the bug is in PoolServerJ, possibly related to the February 20th protocol change [even though it was only supposed to affect old versions of bitcoind].  Working on tracing the problem back immediately.  Until I can figure out what is wrong, mine2 and mine3.btcguild.com are both still running normally!

[allinvain]

Yes, mine1.btcguild.com is down, and it looks like the bug is in PoolServerJ, possibly related to the February 20th protocol change [even though it was only supposed to affect old versions of bitcoind].  Working on tracing the problem back immediately.  Until I can figure out what is wrong, mine2 and mine3.btcguild.com are both still running normally!

Yep, you are correct. I immediately switched over to mine2. I'm staying away from mine3 for a reason I'm sure you can think of :p.

[eleuthria]

Warning to any pool using PoolServerJ

As far as I can tell, something is broken in PoolServerJ as of today, and its likely related to the February 20th protocol change that is affecting older versions of bitcoind.  I'm not positive, but this is the ONLY thing I can come up with.  I've tried 3 different bitcoind versions, and I've tried disabling namecoin merged mining as well.  I've tried two alternative servers and starting PSJ and the same problem happens.

DO NOT RESTART POOLSERVERJ IF YOU ARE CURRENTLY USING IT FOR YOUR POOL!

[allinvain]

Warning to any pool using PoolServerJ

As far as I can tell, something is broken in PoolServerJ as of today, and its likely related to the February 20th protocol change that is affecting older versions of bitcoind.  I'm not positive, but this is the ONLY thing I can come up with.  I've tried 3 different bitcoind versions, and I've tried disabling namecoin merged mining as well.  I've tried two alternative servers and starting PSJ and the same problem happens.

DO NOT RESTART POOLSERVERJ IF YOU ARE CURRENTLY USING IT FOR YOUR POOL!

Have you looked into ecoinpool?

[eleuthria]

Have you looked into ecoinpool?

I did, but it was one of the earlier releases.  I might take another look at it over the weekend now that it has matured a bit more.

[eleuthria]

Update on last night's mine1.btcguild.com [btcguild.com] pool server problems

Last night I tried to push a small configuration change to the pool, to reduce some of the extensive logging that has been enabled ever since we started using PoolServerJ.  When trying to restart, it was popping up an error I had never seen before.  I have traced it back and the following is what I believe has happened to PoolServerJ:

When the 0.4 workmaker edition was released, a lot of additions were made to the PoolServerJ code.  One of these was a sort of piggy-back function, where it connects to the actual bitcoin p2p network.  This was added as a way to try to push block solutions out on the network faster than by submitting it through bitcoind.  It was also a safety net, because if bitcoind doesn't respond when a winning share is submitted, it may end up being orphaned/lost.

I am not positive yet, but I believe shadders developed this portion mimicking how bitcoind connected to peers.  Then the February 20th protocol change hit, where version numbers have checksums.  This seems to have broken the implementation of PoolServerJ's p2p connection [I am 95% sure on this, shadders will confirm hopefully].

BTC Guild was one of (possibly the first) earliest pools to implement the workmaker version of PoolServerJ 0.4.  We had never had problems with it, and while some pool ops had bugs, ours ran very smooth.  However, our version had a fatal error in the p2p protocol handling.  In a later update to 0.4, the p2p connection handling would continue to run retries after a given time frame.  In the version we were running, a failure to connect after retrying would end up killing the PoolServerJ process.  This is what happened last night.

I have since updated to the absolute latest PoolServerJ 0.4, which no longer ends the process if the p2p connection fails.  shadders has been notified of the issue, and hopefully we'll see a fix to the error.  Until then, anybody using PoolServerJ can either ignore the errors (assuming they're on the latest code that won't crash if the p2p fails), or comment out the "source.local.1.p2p.hostport" line, which will disable the p2p connection.


I was able to get the pool back up last night, though there were a few hiccups in the middle of the night.  As of a few minutes ago, a restart was done and all errors have vanished.  Everything should be back to normal, if not better than before.

[rjk]

is this the place you want us to talk about luke jr?

No. STFU.


eleuthria - It would be really cool to see how well ecoinpool works out for you as a larger pool. Graet has been running it and loving it from what I hear, but I don't think his pool load is quite as high as BTC Guild. One of ecoinpool's selling points is its scalability, and it would be really cool to see it in action, if you decide to try it out.

[horgh]

I get this trying to access the website:

Secure Connection Failed
      

          An error occurred during a connection to btcguild.com.

SSL received a record that exceeded the maximum permissible length.

(Error code: ssl_error_rx_record_too_long)

  The page you are trying to view can not be shown because the authenticity of the received data could not be verified.
  Please contact the website owners to inform them of this problem. Alternatively, use the command found in the help menu to report this broken site.

Edit: Already working again.

[Joshwaa]

eleuthria any more thought about the NMC cash in option on your site?

[eleuthria]

eleuthria any more thought about the NMC cash in option on your site?

Sorry, I haven't had much time lately to consider how to handle on-site NMC exchanging.  I'm currently working on preparing the backend to support PPLNS as an option for miners (with fast PPS/PPLNS switching via the frontend).

[Joshwaa]

Very Nice!

[eleuthria]

Well, turns out shadders had posted a few updates to the web-based management side of PoolServerJ between 4.0rc1 and the repo.  I just went in and banned a handful of users that were 99% likely to be botnets.  Might give mine1 a hand, though right now it looks as though somebody was trying (and failing) to attack mine1.btcguild.com.

Bandwith graphs for the server went to ~6x normal for the last hour.  Already starting to die down, and based on the glimpses of my miner at home, they weren't very effective.

[allinvain]

The issue you were having with PoolServerJ on mine1, I'm curious to know if it's also been fixed on the other pool servers. I switched to mine2 when mine1 was having random connection issues and I'm still on mine2. Should I move?

[eleuthria]

The issue you were having with PoolServerJ on mine1, I'm curious to know if it's also been fixed on the other pool servers. I switched to mine2 when mine1 was having random connection issues and I'm still on mine2. Should I move?

Yes, it was fixed on all servers.  I still recommend mine2/mine3 over mine1 though [just because mining on the default is always going to have the worst performance of the 3].

[eleuthria]

Minor cosmetic (sort of) change to the website:  The block list now links to the blockchain.info transaction page, instead of Block Explorer.  More information in an easier to read format.

[allinvain]

Does anyone know what the heck is wrong with block explorer? Downtime upon downtime.

[eleuthria]

mine2/mine3 required a brief restart to make sure they were updated to the new date for the BIP16/p2sh.  Sorry for any connection issues that were caused during the restart.

[nbtcminer]

Hey has mining difficulty been uped this month? And if so, by how much? Seems like my 24 hour production has dropped a fair amount.