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

[ancow]

The reason for this is because the pool pays transaction fees when it should to ensure they get processed by other pools, rather than 0 fees and force users to wait until BTC Guild generates a block to include them.

Do you think having an option to do payouts without fees would be worth it? I, for one, don't mind waiting to be included in the next btcguild block, and if it would save you some transaction fees, I'd gladly take it...

[eleuthria]

The reason for this is because the pool pays transaction fees when it should to ensure they get processed by other pools, rather than 0 fees and force users to wait until BTC Guild generates a block to include them.

Do you think having an option to do payouts without fees would be worth it? I, for one, don't mind waiting to be included in the next btcguild block, and if it would save you some transaction fees, I'd gladly take it...

Since BTC Guild finds a block roughly ever 1h 30m (subject to variance, could be 10 hours, could be 10 seconds), it's actually easier to just utilize a sendmany transaction and pay the small fee attached to it.  Automatic payouts are by far the most preferred way to have users get their BTC out of BTC Guild.  It keeps down block chain bloat, avoids paying extra tx fees, and keeps users from storing too much BTC on my servers.

[eleuthria]

Updated the Idle Email notifications for BTC Guild.  Everyone's idle timer has been turned off by default, so you will have to update it to enable the timers again.

The new notification system doesn't require a minimum speed to enable, and should be more reliable for users that barely met the speed requirements previously in place.  When one or more of your workers go idle, you will receive an email notification listing the ones currently idle.  You will not receive any additional emails until you reset your idle timer.

The original minimum speed requirement was due to how many users set their idle timers, and then would turn their miners off regularly, causing BTC Guild to have many emails marked as spam, and get rejected before ever reaching the users inbox.  This new setup should solve it, since there is no reason for a part-timer to constantly turn their idle email notifications back on.

[DutchBrat]

Eleuthria:

I have noticed for some time now that the pool becomes non-responsive after a Longpoll request (not all of them)

Code:

 cgminer version 2.3.1 - Started: [2012-04-07 02:18:56]
--------------------------------------------------------------------------------
 (5s):0.0 (avg):286.2 Mh/s | Q:473  A:173  R:1  HW:0  E:37%  U:4.10/m
 TQ: 2  ST: 0  SS: 0  DW: 25  NB: 4  LW: 0  GF: 5  RF: 4
 Connected to http://mine.btcguild.com:8332 with LP as user XXXXX
 Block: 00000688823fea93562f7c4b79414535...  Started: [02:50:12]
--------------------------------------------------------------------------------
 [P]ool management [G]PU management [S]ettings [D]isplay options [Q]uit
 GPU 0:  45.5C 2568RPM | 317.3/286.2Mh/s | A:173 R:1 HW:0 U:4.10/m I: 4
--------------------------------------------------------------------------------

[2012-04-07 02:58:48] Accepted 00000000.44cdeb58.9d798909 GPU 0 thread 0 pool 0
[2012-04-07 02:58:51] Accepted 00000000.1bcba3f7.4fec6391 GPU 0 thread 0 pool 0
[2012-04-07 02:59:03] Accepted 00000000.013035ea.9785cf58 GPU 0 thread 0 pool 0
[2012-04-07 02:59:04] Accepted 00000000.ad009ed1.34afe995 GPU 0 thread 0 pool 0
[2012-04-07 02:59:22] Accepted 00000000.243fcd6e.d6b010d0 GPU 0 thread 0 pool 0
[2012-04-07 02:59:23] Accepted 00000000.4f8b16e4.f04c1198 GPU 0 thread 0 pool 0
[2012-04-07 02:59:25] Accepted 00000000.6ca6067b.5c3fb1a7 GPU 0 thread 0 pool 0
[2012-04-07 02:59:44] Accepted 00000000.13adfaea.bae2d64d GPU 0 thread 0 pool 0
[2012-04-07 02:59:59] Accepted 00000000.c6816cb8.6a7ad67a GPU 0 thread 0 pool 0
[2012-04-07 03:00:14] LONGPOLL requested work restart, waiting on fresh work
[2012-04-07 03:00:26] Pool 0 not providing work fast enough
[2012-04-07 03:00:35] longpoll failed for http://mine.btcguild.com:8332/LP/, sle
eping for 30s

It doesn't matter if I mine on mine.btcguild.com /  mine2.btcguild.com or de.btcguild.com

Any idea what can cause this ? Been going on for 1- 2 weeks as far as I know, can't figure it out.

I ran a test by connecting to Deepbit for half a day and did not observe any problems there.

Much appreciated,

Brat

P.S.

Next Longpoll request it happened again:

Code:

 cgminer version 2.3.1 - Started: [2012-04-07 02:18:56]
--------------------------------------------------------------------------------
 (5s):336.3 (avg):285.6 Mh/s | Q:587  A:210  R:1  HW:0  E:36%  U:3.98/m
 TQ: 2  ST: 5  SS: 0  DW: 27  NB: 4  LW: 0  GF: 6  RF: 4
 Connected to http://pit.deepbit.net:8332 with LP as user XXXXX
 Block: 00000688823fea93562f7c4b79414535...  Started: [02:50:12]
--------------------------------------------------------------------------------
 [P]ool management [G]PU management [S]ettings [D]isplay options [Q]uit
 GPU 0:  60.0C 2551RPM | 313.2/285.6Mh/s | A:210 R:1 HW:0 U:3.98/m I: 6
--------------------------------------------------------------------------------

[2012-04-07 03:09:06] Accepted 00000000.cb7f704f.ac058b87 GPU 0 thread 0 pool 0
[2012-04-07 03:09:08] Accepted 00000000.6509be45.3900e3d1 GPU 0 thread 0 pool 0
[2012-04-07 03:09:22] Accepted 00000000.52efebb9.4cd89e68 GPU 0 thread 0 pool 0
[2012-04-07 03:09:31] Accepted 00000000.b0ed46f3.046c17f1 GPU 0 thread 0 pool 0
[2012-04-07 03:09:43] Accepted 00000000.51624697.6442c294 GPU 0 thread 0 pool 0
[2012-04-07 03:10:02] Accepted 00000000.2bec0ff4.ec1b6a68 GPU 0 thread 0 pool 0
[2012-04-07 03:10:12] Accepted 00000000.4e6865c6.66f5fe07 GPU 0 thread 0 pool 0
[2012-04-07 03:10:18] LONGPOLL requested work restart, waiting on fresh work
[2012-04-07 03:10:22] Pool 0 not providing work fast enough
[2012-04-07 03:11:22] Pool 0 http://mine.btcguild.com:8332 not responding!
[2012-04-07 03:11:22] Switching to http://pit.deepbit.net:8332
[2012-04-07 03:11:23] Accepted 00000000.27c89484.f2310589 GPU 0 thread 0 pool 1
[2012-04-07 03:11:27] Accepted 00000000.cc6c9705.c44ec2dc GPU 0 thread 0 pool 1

[eleuthria]

I'm going to talk with the cgminer developer.  There is definitely something wrong WITH CGMINER in how it acts at longpolls, and it's likely an incompatibility with PoolServerJ.  Other miners do not have the same issue cgminer has had.  It seems to occasionally refuse to talk with the pool, and blames the pool for it while other miners happily request work and get a response.

The temporary fix is to set your first backup pool to a different BTC Guild server.

Miners tested that DONT have the problem:  poclbm, MPBM, Ztex miner, Ufasoft.  Haven't tested phoenix/Diablo, but I get a feeling they also don't have the same issue.

Interestingly, the problem seems to be when cgminer gets a response TOO FAST after a longpoll.  Higher latency to the pool actually seems to make the problem disappear.

[DutchBrat]

Thanks Eleuthria!

[phorensic]

Hmm, I don't see that problem in my cgminer.  Although I'm not logging and I'm not watching it 24 hours a day.  Mine just seems to be very reliable.

[eleuthria]

Hmm, I don't see that problem in my cgminer.  Although I'm not logging and I'm not watching it 24 hours a day.  Mine just seems to be very reliable.

It's been an interesting problem that I've been aware of with a few people, but it doesn't seem to be widespread.  Latency seems to be a contributing factor.  One miner with a sub-20ms latency to the German pool gets this on almost every single longpoll.  Hopefully we can figure out what it is that cgminer is hanging up on in the next week.

If you do see the problem occurring, the workaround is to have your first backup set to a different BTC Guild pool server, which so far has seemed to keep people 100% mining on BTC Guild without any idle time and without leaking a few MH/s onto a different backup pool.

[eleuthria]

this is a very good pool

Thank you for the compliments, its refreshing to see a purely positive post in the thread once in a while.

[eleuthria]

A few performance adjustments were made to the database backend this evening.  As a result, the speed estimates on My Account and the API are being rebuilt.  Speed estimates should be back to normal after 15 minutes.

[Joshwaa]

Thanks for the heads up. Also thanks again for the great pool.

[eleuthria]

Fixed a bug in the pool hash rate and hall of fame calculations which was caused by the modifications to the database last night.

[Joshwaa]

Eleuthria having to reset the idle miner warning each time it goes off is a big hassle for me is there anything you can do to get it back to the old way?

[bitlane]

Hey Eleuthria, I currently use 'btcguild.com:8332' for CGMiner.

What is the difference between using the above (simply 'btcguild.com:8332), 'mine2.btcguild.com:8332' or 'mine3.btcguild.com:8332' ?

Which address(es) should I be using as Primary ? and which for Backup ? (as you suggested for CGMiner/Longpoll problems) ?

Chers,
bitlane.

[eleuthria]

Hey Eleuthria, I currently use 'btcguild.com:8332' for CGMiner.

What is the difference between using the above (simply 'btcguild.com:8332), 'mine2.btcguild.com:8332' or 'mine3.btcguild.com:8332' ?

Which address(es) should I be using as Primary ? and which for Backup ? (as you suggested for CGMiner/Longpoll problems) ?

Chers,
bitlane.

I've always recommended mine2 and mine3 (no preference on order) over "btcguild.com".  They're 3 separate IPs/instances of the pool, and mine2/mine3 are significantly less loaded than the primary one since that's the default.

I'd personally recommend:  mine3.btcguild.com as primary, mine2.btcguild.com as a backup [or even de.btcguild.com as backup/second backup].

[eleuthria]

Eleuthria having to reset the idle miner warning each time it goes off is a big hassle for me is there anything you can do to get it back to the old way?

I'm not sure why you even turn them on if you're tripping them often enough for it to become a problem.  Either your setup is unreliable and should be fixed, or you're constantly changing pools in which case you're causing my domain to use up its daily email quota for idle warnings which have no purpose.

[Joshwaa]

Eleuthria having to reset the idle miner warning each time it goes off is a big hassle for me is there anything you can do to get it back to the old way?

I'm not sure why you even turn them on if you're tripping them often enough for it to become a problem.  Either your setup is unreliable and should be fixed, or you're constantly changing pools in which case you're causing my domain to use up its daily email quota for idle warnings which have no purpose.

I live in an area where we have frequent power problems. I have battery backups on all rigs but some times they dont last long enough. Plus I am no where near them during the day.

I am sorry for not knowing your mail server could not handle sending me 2 or 3 emails a day. 

I never change pools I have been consistently with you and I do not set back up pools(cause your service is always up).  Trust me have had offers to join many others but choose to stay here.

Thanks for the reply and consideration.

[eleuthria]

Eleuthria having to reset the idle miner warning each time it goes off is a big hassle for me is there anything you can do to get it back to the old way?

I'm not sure why you even turn them on if you're tripping them often enough for it to become a problem.  Either your setup is unreliable and should be fixed, or you're constantly changing pools in which case you're causing my domain to use up its daily email quota for idle warnings which have no purpose.

I live in an area where we have frequent power problems. I have battery backups on all rigs but some times they dont last long enough. Plus I am no where near them during the day.

I am sorry for not knowing your mail server could not handle sending me 2 or 3 emails a day.  

I never change pools I have been consistently with you and I do not set back up pools(cause your service is always up).  Trust me have had offers to join many others but choose to stay here.

Thanks for the reply and consideration.

It's not that the mail server can't handle it, its that I had to switch to using Google Apps for mailservers to avoid the CONSTANT problems we had with rejected emails.  As a result, we can only send a few hundred emails per day, and we were constantly hitting the cap on emails being sent per day, meaning emails just got dropped.  The biggest offenders were idle emails, and people who received them multiple times per day.

[Olly_K]

Is anyone else seeing they worker speeds in their account be much lower than their miners are actually generating ?

one of my miners in BAMT is hitting 1500mh/s whilst the speed of it under My Account is more like 1100mh/s. Flicked over to my backup pool and its straight on at 1500mh/s.

thanks

[os2sam]

Is anyone else seeing they worker speeds in their account be much lower than their miners are actually generating ?

one of my miners in BAMT is hitting 1500mh/s whilst the speed of it under My Account is more like 1100mh/s. Flicked over to my backup pool and its straight on at 1500mh/s.

thanks

Well, BTC Guild uses a 15 minute averaging window whereas other pools use 30 or so.  Mine fluctuates quite a bit, above and below what my miner says.  Your never going to get a completely accurate rate from a poolside hash rate estimating program.

But, if its consistently lower I would check for communications problems with the pool and test using the other BTCG servers for comparison.
Sam