How to solve "Pool 0 not providing work fast enough" when Solo mining ?

[28nmASIC]

Hi

I try to do solo mine on Mercury using the following configure in bitcoin.conf on Window 7(Virtual machine) and Bitcoin-qt v0.8.5beta

listen=1
gen=0
rpcallowip=192.168.*.*
rpcuser=user
rpcpassword=password
rpcport=9999
server=1
daemon=1

My Mercury status still only has 0 difficulty accepted
CGMiner Status   Running (pid=4409)
Last Checked   Tue Nov 12 06:41:10 UTC 2013
Avg. Hash Rate   134 Gh/s
WU   1891
Difficulty Accepted   0



When I do ssh root@192.168.1.10 and run command "screen -r ", I found this message

--------------------------------------------------------------------------------
cgminer version 3.6.6 - Started: [2013-11-12 06:00:22]
--------------------------------------------------------------------------------
(5s):144.3G (avg):134.9Gh/s | A:0 R:0 HW:1221 WU:1896.1/m
ST: 3 SS: 0 NB: 7 LW: 0 GF: 8600 RF: 0
Connected to 192.168.1.99 diff 511M without LP as user user
Block: 00000ecf0282783a... Diff:511M Started: [06:36:46] Best share: 54.9K
--------------------------------------------------------------------------------
[P]ool management ettings [D]isplay options [Q]uit
KnC 0: | 211.1G/135.2Gh/s | A:0 R:0 HW:1223 WU:1898.9/m
--------------------------------------------------------------------------------

[2013-11-12 06:43:16] Pool 0 not providing work fast enough
[2013-11-12 06:43:16] Pool 0 not providing work fast enough
[2013-11-12 06:43:16] Pool 0 not providing work fast enough
[2013-11-12 06:43:21] Pool 0 not providing work fast enough
[2013-11-12 06:43:21] Pool 0 not providing work fast enough
[2013-11-12 06:43:21] Pool 0 not providing work fast enough


What should I do to fix it or it's normal for Solo mine with zero difficulty accepted?
Thank you

PaulK

[os2sam]

In CGMiner you can increase your ST (Staged Work) by increasing your queue by going to settings and then queue and increase it from the default of 1 to 90 or so.  That will help.  Also setup some backup pools so that when Bitcoin-QT doesn't supply work fast enough you miner can leak some shares to standard pools.  That way it's not a complete loss.

Also the only share you will see submitted to the solo will be block solves.  So not seeing any accepted shares is normal operation for solo mining.
Sam

[28nmASIC]

Hi Sam,

I increase cgminer queue to the maximum 90, then the message is gone. Is it necessary to 
tuning its work queue as minimum or I can use queue with 90 without the effect?  Do you have any command recommend to add more in bitcoin.conf to increase luck or performance? I plan to do Solo mine just only one month, it looks like lottory (@-@).

 cgminer version 3.6.6 - Started: [2013-11-12 20:36:04]
--------------------------------------------------------------------------------
 (5s):148.2G (avg):140.5Gh/s | A:0  R:0  HW:691  WU:1981.1/m
 ST: 92  SS: 0  NB: 5  LW: 0  GF: 3603  RF: 0
 Connected to 192.168.1.99 diff 511M without LP as user user
 Block: 00063def6bf32304...  Diff:511M  Started: [20:55:22]  Best share: 15.1K
--------------------------------------------------------------------------------
 [P]ool management ettings [D]isplay options [Q]uit
 KnC 0:                | 212.0G/141.0Gh/s | A:0 R:0 HW:692 WU:1986.6/m
--------------------------------------------------------------------------------

 [2013-11-12 20:54:13] New block detected on network
 [2013-11-12 20:54:13] KnC running flushwork
 [2013-11-12 20:54:13] KnC: accepted by FPGA 3 works, but only 0 submitted
 [2013-11-12 20:55:22] New block detected on network
 [2013-11-12 20:55:22] KnC running flushwork
 [2013-11-12 20:55:22] KnC: accepted by FPGA 4 works, but only 0 submitted

Thank you very much
PaulK

[os2sam]

Hi Sam,

I increase cgminer queue to the maximum 90, then the message is gone. Is it necessary to 
tuning its work queue as minimum or I can use queue with 90 without the effect?  Do you have any command recommend to add more in bitcoin.conf to increase luck or performance? I plan to do Solo mine just only one month, it looks like lottory (@-@).

I don't use the .conf file myself.  I use "-Q 10" command line switch then manually increase it when I start the miner.  The .conf file may allow larger values then the command line.

As for tuning, I don't know how much difference other values will make.  I use a queue of 60 to 120 as seems to vary how busy it keeps my BE's.  You have allot more hashrate than I do with my meager 20GHs or so.

Just make sure your client has as many connections as possible.

Good Luck,
Sam

[28nmASIC]

Hi Sam,

I got the message again in next 10 minutes but the message occurred quite slower than using its default queue.

That it means my bitcoin-qt reaches its capacities in the current environment, I have to set up backup pool follow your recommended.

I don't know the other persons face this problem like me that using bitcoin client on Windows7 Virtual machine. Is it not generate the issue incase not using Virtual machine or using other OS? Anyone can confirm that solo mining is working normally on Windows7 when having hashing load at 140GH/s up?

Note: Currently, I only have Macbook Pro with small hard disk so it needs to use VMware in case.

Thanks
PaulK

[os2sam]

What many people do who solo mine with high hash rates it so setup their own pool software.  That's more complex than I want to get.  So I just mine with the occasional  "Pool 0 not providing work fast enough" message and not worry too much about it.  And setup failover pools to take up the slack.  I'm happy with that.  Of course I would be happier with finding a block, but hey, can't have everything I guess .
Sam

[wmcorless]

listen=1
gen=0
rpcallowip=192.168.*.*
rpcuser=user
rpcpassword=password
rpcport=9999
server=1
daemon=1

What's the difference between server=1 and daemon=1 if you are running the QT version?