Introducing CherryPicking - new Windows & Linux Pool Hopper

[marks1976]

I have still been unable to get mtred, rfcpool and bitp to connect using cherry picker.  I am still getting the security messages even after installing the ssl files in the java security folder.

rfcpool is closed, so you don't need to worry about that one.  Bitp.it is SMPPS so you would only use that as a backup.  MtRed is good though, so hopefully you can iron out the issues with that.

Did you try downloading Bloodred's pre-generated jssecacerts file?  Did you put it in C:\Program Files (x86)\Java\jre6\lib\security?

Also, what version of java do you have installed?  32-bit jre6 update 26 works for me.

Good luck

Thank you for your help.  I apparently have an x86 version and 64 bit so I put the certs in the x64 and we are good now.  Thank you for the help.  The community and program are great.

[1714928964]

1714928964

[1714928964]

1714928964

[1714928964]

1714928964

[joker060]

When i try to use the start.bat file, it comes up with error 'java' is not recognized as an internal or external command.

Help please???

[joker060]

Got it working but now it is ignoring ArsBitcoin.

One or both of the regular expression strings needed to parse round shares are empty or missing.

[Folax]

When i try to use the start.bat file, it comes up with error 'java' is not recognized as an internal or external command.

Help please???

You need to add Java to your path.
Please do a search in this thread because it's been asked 2 or 3 pages back.

Googling for the error will give you the same solution: add Java to your Windows Path (also explains how). I solved it with google too.

[scatterbrain]

after about a week of CP my average efficiency from 4 proportional pools is 440%, 190%, 310%, 110%. not too bad, but i'm wondering what these will average out to in a few more weeks. does anyone know if bitclockers, ozco or bithasher displays data anywhere to calculate your efficiency?

[Bloodred]

If you have .csv export enabled you can add up the submitted shares for each pool and see how much you've been paid vs. how much it should have been.

[scatterbrain]

i tried that, but it doesn't import nicely into google docs. all the data from a hop goes into one cell which makes for a lot of cut and paste work. this means i can calculate my efficiency per hop, but i want to know per block. finding the time it took a pool to find a block and then adding up every hop's data in that time span seems like a lot of extra work, especially if there are several .csv files to go through. without a way for CP to keep a running total going for each pool, it seems far less time consuming to just go to each pool's website and look at my shares and payout per block on their stats page. am i missing something here or is there an easier way to calculate my efficiency in each pool using CP?

[Bloodred]

I tried it with Excel and it opened fine, haven't tried Google Docs. You could open the .csv file in a regular text editor and replace all occurences of ';' with ',' and see if Google Docs likes it that way. I could also add a file that keeps track of the total shares submitted on each pool, wouldn't be too difficult. Do you mean the total for that session?

Right now I don't think there's an easier way besides adding up the submitted shares for each pool.

[scatterbrain]

commas did the trick for google docs, thank you. adding a file to keep track of total shares submitted to each pool per session would be better, but for finding efficiency wouldn't i still need to go the each pools page and look up data for how many BTC i mined there for the time period of each session in CP? i don't think the log file tracks BTC mined.

more specifically, i'm trying to analyze my efficiency on each block a pool finds to see how well CP is doing its job. i'm running multiple rigs with different algorithms in an attempt to compare their results.  CP gives data for a session length, which may or may not include several blocks found. is there a way you can split the data to show shares submitted per block rather than per hop? maybe a bit more work on your end so i understand if you don't want to implement it.

there's probably a way to simulate this but real world data is not always so predictable, and it's beyond my math skills.

[Bloodred]

Indeed the per-block efficiency wouldn't be so easy to find, but average efficiency should be pretty easy to calculate. Just see how much you earn on all pools and divide that by the total number of submitted shares (running total for a session is in the .csv). This way you'll get the average value of a share which you can then compared to the expected value which should be 50/diff if I am not mistaken. You can also do this for each pool, but finding the blocks would be difficult with any algorithm besides NORMAL (NORMAL jumps when a new round starts or when 43% is reached).

[scatterbrain]

i'll have to settle for average efficiency for now. looks like i've got a lot of .csv files to start importing

[Folax]

CP on Linux:
I kept trying and found that just installing poclbm as a package completely kinda fixed the 127.0.0.1 issue.
I was now back at the "error=13 permission denied" track.
I solved that by giving poclbm.exe execution rights:

Code:

chmod +x /poclbm.exe

Now when starting CP all is fine, CP is picking the right pool and then: nothing.

Code:

* Cherry Picking!
=====================================
 
 
 
 

When I type info I get the stats: 0 for all.

After some time, CP starts to do a check for most efficient pool again, like it should, and then nothing again, dead silence...

[MrWizard]

CP on Linux:
I kept trying and found that just installing poclbm as a package completely kinda fixed the 127.0.0.1 issue.
I was now back at the "error=13 permission denied" track.
I solved that by giving poclbm.exe execution rights:

Code:

chmod +x /poclbm.exe

Now when starting CP all is fine, CP is picking the right pool and then: nothing.

Code:

* Cherry Picking!
=====================================
 
 
 
 

When I type info I get the stats: 0 for all.

After some time, CP starts to do a check for most efficient pool again, like it should, and then nothing again, dead silence...

Are shares being submitted to the pools?  Maybe poclbm is just not showing its output.  Can you check GPU utilization?

[Bloodred]

CP on Linux:
I kept trying and found that just installing poclbm as a package completely kinda fixed the 127.0.0.1 issue.
I was now back at the "error=13 permission denied" track.
I solved that by giving poclbm.exe execution rights:

Code:

chmod +x /poclbm.exe

Now when starting CP all is fine, CP is picking the right pool and then: nothing.

Code:

* Cherry Picking!
=====================================
 
 
 
 

When I type info I get the stats: 0 for all.

After some time, CP starts to do a check for most efficient pool again, like it should, and then nothing again, dead silence...

poclbm probably isn't working. If you can check GPU loads and see if shares are actually being submitted, but my guess is there aren't.

[critical]

downloaded this software but lots of configuration to do, any pre-configured to download?

[Bloodred]

There's an archive with pool .cfg files available in the 2nd post in this topic. All you need to do is add Miner= entries for your GPU miners to each pool cfg file (your miner accounts on that pool).

Then edit poclbm.cfg to add the path to poclbm.exe and any arguments you may want (-v -w -f, etc)

[critical]

thanks, for the java what to do, they can't seem to find java, though i already downloaded it for sure

[Folax]

Java error: Java needs to be in your Windows Path, please do a search for it in this topic or google the error message, its a very common error and is easily fixed. It's almost asked once every 2 pages here I think, shouldn't be hard to find.

Linux: no shares are being committed. I don't think poclbm is set to do anything.

[Bloodred]

Are any poclbm processes actually running and staying idle or are they simply not starting in the first place?

[tal0n4]

Hey Bloodred would it possible to add an option to set the CPU affinity for the poclbm processes?