Pages:
Author

Topic: [20 gh] NMCBit pool 3% fee prop 6.6% PPS - page 30. (Read 63693 times)

hero member
Activity: 780
Merit: 510
Bitcoin - helping to end bankster enslavement.
September 02, 2011, 09:17:06 PM
Hey Davinci,

I'm having a lot of RPC connection problems and rejected shares again on 4 miners, since we found the last block

Strange, since everything went great in the last round

Rebooted the miners, hope that it helps,

Brat


As you can see from our stats the total invalid shares are 3500 of 110,000 that's over 3% and although it's not a disaster it's not good.  If you have rejects greater than 3% let me know I will look into your shares and see what the reason is.

Davinci

BTW

I am working on the balances and I did have to reboot one of the servers that would end up having a lot of shares go bad.
hero member
Activity: 868
Merit: 1000
September 02, 2011, 08:58:52 PM
Hey Davinci,

I'm having a lot of RPC connection problems and rejected shares again on 4 miners, since we found the last block

Strange, since everything went great in the last round

Rebooted the miners, hope that it helps,

Brat
hero member
Activity: 780
Merit: 510
Bitcoin - helping to end bankster enslavement.
August 31, 2011, 11:34:19 AM
Hey Davinci,

looks like you fixed some bugs !

Stales are way down, I do have some RPC connection problems, but much less than over the weekend

Looking good !!!

Brat
Yeah well it's easy to have the system up and running when you don't have load. Smiley  The trick is to do it with huge load.
hero member
Activity: 868
Merit: 1000
August 31, 2011, 08:52:10 AM
Hey Davinci,

looks like you fixed some bugs !

Stales are way down, I do have some RPC connection problems, but much less than over the weekend

Looking good !!!

Brat
hero member
Activity: 780
Merit: 510
Bitcoin - helping to end bankster enslavement.
August 30, 2011, 11:33:19 PM
Working on the stats slowly so I am first going to fix speed up displaying blocks found and then work on resetting shares speed, then updating balances.  So it may take a few days if pushpool does not collapse or something else doesn't break.

A lot of work ahead of me need to fix the web site before the launch date.  After that I need to fix pushpool and namecoind to as it's sufficient right now but not perfect.
hero member
Activity: 780
Merit: 510
Bitcoin - helping to end bankster enslavement.
August 30, 2011, 10:36:38 PM
Stabilized the miner connection to nmcbit.com  with 3 severs round robin should be good enough to keep it from collapsing over and over again.  Lets just wait and see.
Anyhow decided to sacrifice everything else on the web site for this lets hope my work is not another learning experience.

Please let me know how your connections are so far I'm putting out a lot of shares but I get this as well...


Code:
2011-08-30 23:18:38: Listener for "testbitcoinpool": 30/08/2011 23:18:38, Problems communicating with bitcoin RPC 0 2
2011-08-30 23:19:13: Listener for "nmcbit test user": 30/08/2011 23:19:13, Problems communicating with bitcoin RPC 0 2
2011-08-30 23:20:09: Listener for "nmcbit test user": 30/08/2011 23:20:09, Problems communicating with bitcoin RPC 0 2
2011-08-30 23:20:58: Listener for "nmcbit test user": 30/08/2011 23:20:58, long poll exception:
2011-08-30 23:20:59: Listener for "nmcbit test user": 30/08/2011 23:20:59, Using new LP URL /LP
2011-08-30 23:20:59: Listener for "nmcbit test user": 30/08/2011 23:20:59, LP connected to 107.20.229.40:8332
2011-08-30 23:21:05: Listener for "nmcbit test user": 30/08/2011 23:21:05, Problems communicating with bitcoin RPC 0 2
2011-08-30 23:21:23: Listener for "testbitcoinpool": 30/08/2011 23:21:23, Problems communicating with bitcoin RPC 0 2
2011-08-30 23:23:51: Listener for "nmcbit test user": 30/08/2011 23:23:51, Problems communicating with bitcoin RPC 0 2
2011-08-30 23:24:47: Listener for "nmcbit test user": 30/08/2011 23:24:47, Problems communicating with bitcoin RPC 0 2
2011-08-30 23:25:52: Listener for "testbitcoinpool": 30/08/2011 23:25:52, long poll exception:
2011-08-30 23:25:53: Listener for "testbitcoinpool": 30/08/2011 23:25:53, Using new LP URL /LP
2011-08-30 23:25:53: Listener for "testbitcoinpool": 30/08/2011 23:25:53, LP connected to coinserver2.nmcbit.com:8332
2011-08-30 23:26:38: Listener for "nmcbit test user": 30/08/2011 23:26:38, Problems communicating with bitcoin RPC 0 2
2011-08-30 23:27:35: Listener for "nmcbit test user": 30/08/2011 23:27:35, Problems communicating with bitcoin RPC 0 2
2011-08-30 23:28:31: Listener for "nmcbit test user": 30/08/2011 23:28:31, Problems communicating with bitcoin RPC 0 2
2011-08-30 23:29:27: Listener for "nmcbit test user": 30/08/2011 23:29:27, Problems communicating with bitcoin RPC 0 2
2011-08-30 23:30:23: Listener for "nmcbit test user": 30/08/2011 23:30:23, Problems communicating with bitcoin RPC 0 2
2011-08-30 23:31:19: Listener for "nmcbit test user": 30/08/2011 23:31:19, Problems communicating with bitcoin RPC 0 2
2011-08-30 23:32:15: Listener for "nmcbit test user": 30/08/2011 23:32:15, Problems communicating with bitcoin RPC 0 2
2011-08-30 23:33:11: Listener for "nmcbit test user": 30/08/2011 23:33:11, Problems communicating with bitcoin RPC 0 2
2011-08-30 23:33:22: Listener for "testbitcoinpool": 30/08/2011 23:33:22, Problems communicating with bitcoin RPC 0 2
2011-08-30 23:34:18: Listener for "testbitcoinpool": 30/08/2011 23:34:18, Problems communicating with bitcoin RPC 0 2
2011-08-30 23:35:14: Listener for "testbitcoinpool": 30/08/2011 23:35:14, Problems communicating with bitcoin RPC 0 2

Is anyone else getting the same thing?

Davinci
hero member
Activity: 780
Merit: 510
Bitcoin - helping to end bankster enslavement.
August 30, 2011, 03:29:48 PM
Stats are frozen for now we found a block but I need to fix the DB update also I think I fixed some of the problems with pushpool I am working on it still right now then I will get to fixing the stored proc to update the DB faster or fail if the time gets to long.

Thanks again everyone lets hope I got it all straight this time.
donator
Activity: 1731
Merit: 1008
August 30, 2011, 01:03:25 PM
I got paid by Davinci for participating in the 48hrs testing of nmcbit pool.

Thanks,   count me in for future testing.
hero member
Activity: 481
Merit: 529
August 30, 2011, 07:53:05 AM
I have not seen anything like this has anyone else?

My miners are hosted, I don't have access to their logs.  But I am seeing less than half the expected hash rate (3.25G) and a good deal of traffic to my failover pool.

Can you give me any kind of details? Settings miner software etc?

Relaying information:
Quote
It's poclbm with some minimal modifications which do not affect
communication with pools much. But you might see rather high (2%)
level of rejects (dupes) on some pools but it is ok, the miner simply
resend some solutions if it thinks that the pool could have not
received it. It does not affect the overall performance.
full member
Activity: 180
Merit: 100
August 30, 2011, 12:25:53 AM
I, too, participated in this endeavor (the 5+ GH/s for 48 hours) and was well rewarded by Davinci, as promised. In a world where so little can be trusted, I have found Davinci to always honor his word.
hero member
Activity: 602
Merit: 500
August 29, 2011, 10:02:54 PM
I apologize that I had been running my clients with GUIMiner rather than a console client, so I couldn't really keep track of warning messages. My stales were fairly low throughout though, I just got a lot of random inability to connecto. Possibly the database overloaded or something, because I could usually access the website (but it was slow) even when my miners were all dropped off. Just from the temperature of my mining room I could tell that it was a systemic problem though, the temperature was significantly cooler than usual, suggesting that the cards were idling up and down a lot.

If there is another round I'll run a console version next time to give more detailed feedback.


In the meantime, thought it would be prudent to point out that Davinci did indeed pay out for my/our time as he promised, his good word continues to hold merit Smiley
hero member
Activity: 780
Merit: 510
Bitcoin - helping to end bankster enslavement.
August 29, 2011, 09:26:00 PM
[30/08/2011 02:10:37] Result: 26458332 accepted
[30/08/2011 02:10:43] Warning: work queue empty, miner is idle
[30/08/2011 02:10:45] Result: 0bdb0062 accepted
[30/08/2011 02:11:00] Warning: work queue empty, miner is idle
[30/08/2011 02:11:01] Result: 83c8e921 accepted

lots of trouble connecting lately

Another new bug when 2 block are found right after each other the DB goes crazy.

I would shut it down but there is no point I will let it run until it sets all the values.  Then see if its correct and fix that bug if the values are not correct.  Oh well no one said programming is easy. Smiley
I feel a bit embarrassed as I thought I had it all working fine. 
donator
Activity: 1731
Merit: 1008
August 29, 2011, 09:11:45 PM
[30/08/2011 02:10:37] Result: 26458332 accepted
[30/08/2011 02:10:43] Warning: work queue empty, miner is idle
[30/08/2011 02:10:45] Result: 0bdb0062 accepted
[30/08/2011 02:11:00] Warning: work queue empty, miner is idle
[30/08/2011 02:11:01] Result: 83c8e921 accepted

lots of trouble connecting lately
hero member
Activity: 780
Merit: 510
Bitcoin - helping to end bankster enslavement.
August 29, 2011, 09:04:17 PM
No Problem Davinci, as always: Thanks for your hardwork

It's fun to be in on this from the beginning

That being said: More bad news

Looks like all my miners have lost connection at the same time now (9 miners hasing at a total 273 MH rate indicates something is wrong)

Sorry to have to be the one to tell you Wink

Brat

Yeah with the block found the DB is gone nuts trying to update the tables don't know why yet as I am trying to contain other fires ah well I will get it all fixed by the 5th one way or another.

Davinci
hero member
Activity: 868
Merit: 1000
August 29, 2011, 01:04:36 PM
No Problem Davinci, as always: Thanks for your hardwork

It's fun to be in on this from the beginning

That being said: More bad news

Looks like all my miners have lost connection at the same time now (9 miners hasing at a total 273 MH rate indicates something is wrong)

Sorry to have to be the one to tell you Wink

Brat
hero member
Activity: 780
Merit: 510
Bitcoin - helping to end bankster enslavement.
August 29, 2011, 08:50:54 AM
hero member
Activity: 868
Merit: 1000
August 29, 2011, 07:08:12 AM
Davinci,

Sorry for being the bearer of bad news again, but that's why I love testing:
It's  not allowed to shoot the messenger  Wink

Having LongPolling problems now

2011-08-29 14:04:34: Listener for "HennyNMC" started
2011-08-29 14:04:34: Listener for "HennyNMC": 29/08/2011 14:04:34, Setting pool [email protected]_Meyer @ nmcbit.com:8332
2011-08-29 14:04:35: Listener for "HennyNMC": 29/08/2011 14:04:35, Using new LP URL /LP
2011-08-29 14:04:35: Listener for "HennyNMC": 29/08/2011 14:04:35, LP connected to nmcbit.com:8332
2011-08-29 14:05:09: Listener for "HennyNMC": 29/08/2011 14:05:09, long poll exception:
2011-08-29 14:05:10: Listener for "HennyNMC": 29/08/2011 14:05:10, Using new LP URL /LP
2011-08-29 14:05:10: Listener for "HennyNMC": 29/08/2011 14:05:10, LP connected to nmcbit.com:8332
2011-08-29 14:05:11: Listener for "HennyNMC": 29/08/2011 14:05:11, long poll exception:
2011-08-29 14:05:12: Listener for "HennyNMC": 29/08/2011 14:05:12, Using new LP URL /LP
2011-08-29 14:05:12: Listener for "HennyNMC": 29/08/2011 14:05:12, LP connected to nmcbit.com:8332
2011-08-29 14:05:13: Listener for "HennyNMC": 29/08/2011 14:05:13, long poll exception:
2011-08-29 14:05:14: Listener for "HennyNMC": 29/08/2011 14:05:14, Using new LP URL /LP
2011-08-29 14:05:14: Listener for "HennyNMC": 29/08/2011 14:05:14, LP connected to nmcbit.com:8332
2011-08-29 14:05:15: Listener for "HennyNMC": 29/08/2011 14:05:15, long poll exception:
2011-08-29 14:05:16: Listener for "HennyNMC": 29/08/2011 14:05:16, Using new LP URL /LP
2011-08-29 14:05:16: Listener for "HennyNMC": 29/08/2011 14:05:16, LP connected to nmcbit.com:8332
2011-08-29 14:05:18: Listener for "HennyNMC": 29/08/2011 14:05:18, long poll exception:
2011-08-29 14:05:19: Listener for "HennyNMC": 29/08/2011 14:05:19, Using new LP URL /LP
2011-08-29 14:05:19: Listener for "HennyNMC": 29/08/2011 14:05:19, LP connected to nmcbit.com:8332
2011-08-29 14:05:31: Listener for "HennyNMC": 29/08/2011 14:05:31, Problems communicating with bitcoin RPC 0 2

Still using GUI miner with t he same settings

Brat
hero member
Activity: 868
Merit: 1000
August 29, 2011, 03:00:58 AM
Hey Davinci,

Just my 2 cents worth: could it be that MySQL locks-up during the rounds as well ? I am still experiencing lots and lots of RPC communication errors, across all my miners which causes a high % of stales

I can see that I am not the only one since we are running a total stale % of close to 8%

First I thought that was caused by the database problems at the beginning of each new round, but it seems the problems remain during the entire round

Mind you, I have no knowledge of pushpool or the database, so I might be way off

Brat
hero member
Activity: 780
Merit: 510
Bitcoin - helping to end bankster enslavement.
August 28, 2011, 10:32:40 PM
Hi Everyone.

Thanks for all you hashs I am working hard right now in my new house trying to fix all the problems I've observed from the past 24 hours.   We have had several outages and many bugs show up this is what I need to do in order to create a solid pool.  Thanks everyone it is bitcoins well spent. 


BTW
Right now my back is killing me, I'm surrounded by boxes but I am going to work on nmcbit.com for most of the night and all day everyday for rest of the week.  My wife will be pissed.   Grin  Oh well.
hero member
Activity: 780
Merit: 510
Bitcoin - helping to end bankster enslavement.
August 28, 2011, 05:45:29 PM
Here we go again,

i assume we found a block since our total shares turned negative and all my miners are failing to the backup-pool since they can't connect anymore

Database problem again me thinks  Grin

At least if it's the same bug over and over Davinci will know what to do after his move

Good Luck !

Brat
Yeah there must be another DB bug as soon as you find a block it slows down the MySQL but not as bad as the last time.  The last time it would of taken hours to update the DB right not it took 5 minutes but that's not acceptable.
Pages:
Jump to: