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

[BookLover]

I think you are misunderstanding my situation.  I am limited to ~200MB per 24h or my connection will be grounded to a near halt.(This limit is not just on my miners and includes normal data consumption)

Here might be some insight...

Rough calculation:

Now:
--------
One request: 300 bytes of HTTP request, 700 bytes of data ==> ~1 kB of data every 5 seconds for each worker. It is 12kB per minute per worker.

Json over TCP:
--------
One request: (approx) 20 bytes of request, 300? bytes of response EVERY MINUTE ==> 320 bytes per minute per worker.

By very simple optimization, you cut bandwidth to 2.5% of original size. Without any binary fiddling and proprietary stuff. How many % will be the savings between Json over TCP and binary over TCP?[/list]

AFAIK, most pools are still using HTTP method.  So you're looking at about 17.3MB per worker, per 24 hours.

Possibly using some sort of compression-enabled tunnel *might* be able to shave some bandwidth, but I wouldn't expect anything more than 30% savings.

Cheers,
Kermee

Thanks for all the suggestions, I'll see what I can do on my end.

I noticed something odd on the performance charts page.  On the 30th the pool speed was 1917GH/s and the BTC earned was 900BTC and on the 31st the pool speed was 1912GH/s and the BTC earned was 900BTC also.  So if I had the same number of GH/s both days(In this case 1.86GH/s), then I should either get the same amount on both days or maybe slightly more on the 31st.  What I actually was rewarded was 0.88764BTC on the 30th and 0.83512BTC on the 31st.

I'm not accusing the pool of cheating me and I say this with much respect to all the work eleuthria has gone through, I just can't figure out what could have caused this and my only plausible theory is the pool is cheating me.

[kano]

...
I noticed something odd on the performance charts page.  On the 30th the pool speed was 1917GH/s and the BTC earned was 900BTC and on the 31st the pool speed was 1912GH/s and the BTC earned was 900BTC also.  So if I had the same number of GH/s both days(In this case 1.86GH/s), then I should either get the same amount on both days or maybe slightly more on the 31st.  What I actually was rewarded was 0.88764BTC on the 30th and 0.83512BTC on the 31st.

I'm not accusing the pool of cheating me and I say this with much respect to all the work eleuthria has gone through, I just can't figure out what could have caused this and my only plausible theory is the pool is cheating me.

Reward is based in the number of shares you find.

Shares are random as are blocks.

If you mine at X Mh/s every day - each day you will get a different number of shares - that can be almost the same or even vary a lot.

Even if the total number of blocks found those days is the same, your total number of shares can still be different each day.

[eleuthria]

BookLover: Connection issues on your end, using your miner for something other than mining if it isn't a dedicated rig, and as kano said, luck, can affect individual rewards.  You'll notice on the 'My Account' page I give you your 15 minute average speed based on your share submission (the only way a pool can determine your speed).  This can easily vary 10%, and sometimes as much as 25%, +/- from what your miner's actual speed is, due to the randomness of share production.  A 24 hour time frame -should- normally average out very close to your actual speed, so rewards generally don't vary more than a small percentage.  Based on your numbers, it was only ~6% difference between the two days.  If one day you were +3%, the other day -3%, there is your difference.

UPDATE:  Another potential reason are the short rounds, under 10 minutes.  Even more volatile are the ones that are under 1 minute, where your rewards can be anywhere from 0 to 3x your normal production due to your luck of finding shares in the short time frame.

[BookLover]

To kano:
I know my shares will vary.   My speed was pulled from the website, not my miner, so the 1.86GH/s is calculated FROM the number of shares submitted.

To eleuthria:
I had no unusual connection issues.  My mining rigs are dedicated.  I got my speed from the my performance page which as you know is the 24h average, so my average speed for both days are the same and my average speed should be based upon average share submission.  Thus my average share submission should be VERY close.  The Update answers my question, and I understand what happened now.  Thanks!

P.S. If someone happens to feel like it, could they please crunch the numbers to see if this is possible?  I'm not sure how to do it.

[BkkCoins]

P.S. If someone happens to feel like it, could they please crunch the numbers to see if this is possible?  I'm not sure how to do it.

Just go to the block history page and select each of the time ranges in question. You can visually look over the blocks to see if there are many short blocks, or count how many blocks in each range, or you can sum up the shares for each range and compare.

I find that on blocks less than 1 minute I often either get 0 or 50% more shares than average. As a new miner I'm quite surprised by how much variance there is in luck.

[eleuthria]

Server move was completed early (~8:35 PM, PDT).  So far everything seems to be running smooth.  If you see anything weird (most likely things related to time due to the time zone change), please report it to webmaster@btcguild.com

[BkkCoins]

Great. Looks fine except the time since last share in the worker summary table now shows a 9 hour offset.
eg. 9:00:23 when should just be 0:23

[eleuthria]

Fixed the last share time problem.

[iLLmAtiCiAn]

Is anyone else having a hard time both connecting to the site and to the pool through mining?

[eleuthria]

Had 3-4 minutes of downtime, routing issues at the host.  By the time I was notified of the issue and had a chance to look it was already recovering.

[PcChip]

Strange - 1 of my miners recovered, 4 didn't and had transferred 100% of the hashing power to Arsbitcoin

I wonder what makes a percentage of them hang up like that? (phoenix 1.50)

[jjiimm_64]

E,  I need to share something with you that sharky spoke of last month.  I think it may be contributing to the pools 'luck' or lack there of.

I have mined millions of shares with your pool since June.  Sometime after you migrated to the new servers  I started getting miner idle messages.

Code:

2011-09-07 22:34:28: Listener for "guild.2.p":            [07/09/2011 22:34:28] LP: New work pushed
2011-09-07 22:34:28: Listener for "guild.1.p":            [07/09/2011 22:34:28] LP: New work pushed
2011-09-07 22:34:28: Listener for "guild.4.p":            [07/09/2011 22:34:28] LP: New work pushed
2011-09-07 22:35:25: Listener for "guild.3.p":		  [07/09/2011 22:35:25] LP: New work pushed
2011-09-07 22:35:25: Listener for "guild.2.p":            [07/09/2011 22:35:25] LP: New work pushed
2011-09-07 22:35:25: Listener for "guild.1.p":            [07/09/2011 22:35:25] LP: New work pushed
2011-09-07 22:35:25: Listener for "guild.4.p":            [07/09/2011 22:35:25] LP: New work pushed
2011-09-07 22:36:54: Listener for "guild.3.p":            [07/09/2011 22:36:54] LP: New work pushed
2011-09-07 22:36:54: Listener for "guild.2.p":            [07/09/2011 22:36:54] LP: New work pushed
2011-09-07 22:36:54: Listener for "guild.1.p":            [07/09/2011 22:36:54] LP: New work pushed
2011-09-07 22:36:54: Listener for "guild.4.p":            [07/09/2011 22:36:54] LP: New work pushed
2011-09-07 22:48:49: Listener for "guild.4.p":            [07/09/2011 22:48:49] LP: New work pushed
2011-09-07 22:48:49: Listener for "guild.3.p":            [07/09/2011 22:48:49] LP: New work pushed
2011-09-07 22:48:49: Listener for "guild.1.p":            [07/09/2011 22:48:49] LP: New work pushed
2011-09-07 22:48:49: Listener for "guild.2.p":            [07/09/2011 22:48:49] LP: New work pushed
2011-09-07 22:56:05: Listener for "guild.4.p":            [07/09/2011 22:56:05] LP: New work pushed
2011-09-07 22:56:05: Listener for "guild.1.p":            [07/09/2011 22:56:05] LP: New work pushed
2011-09-07 22:56:05: Listener for "guild.2.p":            [07/09/2011 22:56:05] LP: New work pushed
2011-09-07 22:56:05: Listener for "guild.3.p":            [07/09/2011 22:56:05] LP: New work pushed
2011-09-07 23:09:28: Listener for "guild.2.p":            [07/09/2011 23:09:28] Warning: work queue empty, miner is idle
2011-09-07 23:09:38: Listener for "guild.3.p":            [07/09/2011 23:09:38] Warning: work queue empty, miner is idle
2011-09-07 23:09:51: Listener for "guild.4.p":            [07/09/2011 23:09:51] Warning: work queue empty, miner is idle
2011-09-07 23:10:09: Listener for "guild.2.p":            [07/09/2011 23:10:09] Warning: work queue empty, miner is idle
2011-09-07 23:10:12: Listener for "guild.4.p":            [07/09/2011 23:10:12] Warning: work queue empty, miner is idle
2011-09-07 23:10:30: Listener for "guild.2.p":            [07/09/2011 23:10:30] Warning: work queue empty, miner is idle
2011-09-07 23:10:33: Listener for "guild.3.p":            [07/09/2011 23:10:33] Warning: work queue empty, miner is idle
2011-09-07 23:10:36: Listener for "guild.4.p":            [07/09/2011 23:10:36] Warning: work queue empty, miner is idle
2011-09-07 23:10:51: Listener for "guild.2.p":            [07/09/2011 23:10:51] Warning: work queue empty, miner is idle
2011-09-07 23:18:02: Listener for "guild.2.p":            [07/09/2011 23:18:02] LP: New work pushed
2011-09-07 23:18:02: Listener for "guild.4.p":            [07/09/2011 23:18:02] LP: New work pushed
2011-09-07 23:18:02: Listener for "guild.3.p":            [07/09/2011 23:18:02] LP: New work pushed
2011-09-07 23:18:02: Listener for "guild.1.p":            [07/09/2011 23:18:02] LP: New work pushed
2011-09-07 23:30:17: Listener for "guild.4.p":            [07/09/2011 23:30:17] LP: New work pushed
2011-09-07 23:30:17: Listener for "guild.2.p":            [07/09/2011 23:30:17] LP: New work pushed
2011-09-07 23:30:17: Listener for "guild.1.p":            [07/09/2011 23:30:17] LP: New work pushed
2011-09-07 23:30:17: Listener for "guild.3.p":            [07/09/2011 23:30:17] LP: New work pushed
2011-09-07 23:40:45: Listener for "guild.1.p":            [07/09/2011 23:40:45] LP: New work pushed
2011-09-07 23:40:45: Listener for "guild.2.p":            [07/09/2011 23:40:45] LP: New work pushed
2011-09-07 23:40:45: Listener for "guild.3.p":            [07/09/2011 23:40:45] LP: New work pushed
2011-09-07 23:40:45: Listener for "guild.4.p":            [07/09/2011 23:40:45] LP: New work pushed

Now, about 3 or 4 days ago. I put one of my rigs on the pool for a day.  I only saw 1 miner idle message all day.  so I put all 14G there yesterday.  and we had a good day.  did not see hardly any miner idle messages...  now tonight I noticed that our luck was back above 2 million..  so I went into the console and this is what I found.  3 of my workers on one machine was idle for 10 minutes.  I had just restarted this box about an hour ago, so this is all the console I have to go on. This behavior does NOT happen on the other pools.. 

Now I know you'll say just go to them,  and I might. but I wanted to let you know my observation.  I really like your pool, and especially your website.  but it is hard to justify about 1-2 btc aday just for the kool site.....

[eleuthria]

I see ~90 seconds of idles in your log.  Whether or not that was on my side or yours I can't say, it was a very short time frame.  There was a 4 minute window in the middle of the day with idles today due to an ISP routing issue that affected about half of the servers including the website.

However, idles have _NOTHING_ to do with luck.  Luck is determined by shares in a round, which idles have absolutely no affect on.

[jjiimm_64]

I see ~90 seconds of idles in your log.  Whether or not that was on my side or yours I can't say, it was a very short time frame.  There was a 4 minute window in the middle of the day with idles today due to an ISP routing issue that affected about half of the servers including the website.

However, idles have _NOTHING_ to do with luck.  Luck is determined by shares in a round, which idles have absolutely no affect on.

from 23:10 to 23:18 until the worker got the new push pool I dont think it was working.  I can also notice this as a drop in my hashrate

I just noticed it was dropping..  went from 16G down to 14.6  I thought one of the rigs went offline... then I looked at the console again and I saw this.

Code:

2011-09-08 01:08:41: Listener for "guild.2.p":                           [08/09/2011 01:08:41] Disconnected from server
2011-09-08 01:08:51: Listener for "guild.2.p":                           [08/09/2011 01:08:51] Warning: work queue empty, miner is idle
2011-09-08 01:08:58: Listener for "guild.2.p":                           [08/09/2011 01:08:58] Connected to server
2011-09-08 01:13:42: Listener for "guild.3.p":                           [08/09/2011 01:13:42] LP: New work pushed
2011-09-08 01:13:42: Listener for "guild.4.p":                           [08/09/2011 01:13:42] LP: New work pushed
2011-09-08 01:13:42: Listener for "guild.1.p":                           [08/09/2011 01:13:42] LP: New work pushed
2011-09-08 01:13:42: Listener for "guild.2.p":                           [08/09/2011 01:13:42] LP: New work pushed
2011-09-08 01:18:07: Listener for "guild.3.p":                           [08/09/2011 01:18:07] Disconnected from server
2011-09-08 01:18:16: Listener for "guild.3.p":                           [08/09/2011 01:18:16] Warning: work queue empty, miner is idle
2011-09-08 01:18:18: Listener for "guild.3.p":                           [08/09/2011 01:18:18] Failed to connect, retrying...
2011-09-08 01:18:34: Listener for "guild.3.p":                           [08/09/2011 01:18:34] Failed to connect, retrying...
2011-09-08 01:18:51: Listener for "guild.3.p":                           [08/09/2011 01:18:51] Failed to connect, retrying...
2011-09-08 01:19:00: Listener for "guild.2.p":                           [08/09/2011 01:19:00] Disconnected from server
2011-09-08 01:19:06: Listener for "guild.3.p":                           [08/09/2011 01:19:06] Connected to server
2011-09-08 01:19:09: Listener for "guild.2.p":                           [08/09/2011 01:19:09] Warning: work queue empty, miner is idle
2011-09-08 01:19:10: Listener for "guild.2.p":                           [08/09/2011 01:19:10] Connected to server

I know the idles dont directly correlate to luck.  but if alot of miners are idle some of the time, then that could account for not finding the blocks as fast as we should.. NO?

edit:  I just checked on of the rigs that have been pointed to the guild for about 4-5 days straight. the console looks pretty bad with idles

I know you dont think it is on your side, but sharky was reporting the same thing...  it is not just me.  I just wanted to let you know. I think I will go with ars for a while and see what the logs look like later.  will let you know.
Jim

[Coolhwip]

WTF is this?. I can connect to other pools just fine so don't tell me it's a problem on my end.

Code:

[08/09/2011 03:54:40] ConnectionRefusedError: 10061: No connection could be made because the target machine actively refused it..
[08/09/2011 03:54:40] Failed to connect, retrying...
[08/09/2011 03:54:57] ConnectionRefusedError: 10061: No connection could be made because the target machine actively refused it..
[08/09/2011 03:54:57] Failed to connect, retrying...
[08/09/2011 03:55:13] ConnectionRefusedError: 10061: No connection could be made because the target machine actively refused it..
[08/09/2011 03:55:13] Failed to connect, retrying...

Edit: Fixed it myself, I had to flush the dns cache and reboot the modem a couple of times before it finaly updated. What kind of ass backwards server did you move the pool to :/

[BkkCoins]

Eeek. Last 4 blocks have followed doubling progession:

0:30:02, 1:11:02, 2:03:50, 4:01:16 (so far).

I sure hope this doesn't continue   

[eleuthria]

from 23:10 to 23:18 until the worker got the new push pool I dont think it was working.  I can also notice this as a drop in my hashrate

8 minutes between long polls is absolutely normal.  You were idle for under 2 minutes.

The second set of disconnects/reconnects in your log were me restarting the load balancers, I added an extra server to the list [the old 2nd server for SC Guild].  You'll notice the first disconnect is less than 30 seconds before the last reconnect. 

I also banned 17 IPs last night that were showing very high getwork requests and low/non-existent share submissions, which should help as well.  I'm still manually performing these bans until I'm confident my script wont ban legit users that have clients keeping a connection open with the pool for failover/load balancing in cgminer/poclbm.

[Iyeman]

Jesus


Shares per Block (24H)      3,304,038   

2872  144489   2011-09-08 13:53:06   7:12:33   11743359

[eleuthria]

Yeah, during the final hour of the block (I was asleep for most of it), the only thought going through my mind was:  "Please don't have us set the new record for most shares in a single block."

[eleuthria]

I've identified the problem people are having connecting.  The first two servers in the round robin list are receiving far greater load than the other 8.  If you're having issues with your connection try changing to any of the following IPs directly:

78.46.186.29
78.46.186.30
78.46.186.50
78.46.186.51
78.46.186.52
78.46.186.53
78.46.186.54


I'm provisioning a pair of two small servers to put at the top of the round robin list to alleviate the problem.  Hopefully they will be online in just a few hours.