bitHopper: Python Pool Hopper Proxy

[iopq]

Update.. Downgrading from Python 64bit to 32bit version solved the problem..

I suspect because of OpenSSL

[joulesbeef]

what does mine_friendly do? jump back when they got big blocks?


small request...to make the server names on the website.. links.. seems like we already have most of the links in the cfg though they tend to be commented out.

[Sukrim]

there must be some sort of lag time when connecting to a new server in which your miner is doing no work. This lag IMHO probably eats up the tiny advantage.

There is very little lag and actually one could prefetch getworks to reduce it even further (just stay connected to several pools at once). Currently most of the lost profit was due to shares being submitted to the wrong pool (a few % less) but anyways - as the earnings are enourmously higher than expected if you hop several pools it still paid off.

We'll need the "staying connected to several pools at once" anyways in the future for Long Polling timing

Currently I'm researching autobuild mechanisms... Jenkins seems quite ok. Now to creating a single executable with a python interpreter that runs on Windows under Linux...

[gnaget]

Currently most of the lost profit was due to shares being submitted to the wrong pool (a few % less) but anyways - as the earnings are enourmously higher than expected if you hop several pools it still paid off.

Update to the latest version, I think you'll be quite pleased

[djex]

Any one else getting api_disable with mt.red? I checked the api key and it's correct and api link works in browser.

[joulesbeef]

i dont know how long that api_disabled lasts, I restarted the hopper and it is fine now.. they are nearly at 300% so you arent missing anything right now

[djex]

Yeah just restarted, working fine now.

[Tmoney]

I've been testing out some namecoin mining.

https://i.imgur.com/ARWhJ.png

mine_nmc does what it's supposed to do and stops mining at .40 of namecoin difficulty, however, when bithopper compares a namecoin pool to a bitcoin pool it always prefers namecoin because it uses bitcoins dificulty ratio to compare the two.  Also, the stats page displays namecoins percentages with bitcoins difficulty.  In this picture namebit should be around 28% rather than 1.57%.

[iopq]

should I use mineco.in as a backup pool? it doesn't punish hopping

also, why is it mining slush like crazy? aren't you supposed to hop away from slush earlier than other pools due to the decay factor? it's still mining slush at 37% instead of a backup
it stopped mining slush at 41.48%

[djex]

tmoney: whos version of bitHopper are you using?

[ewibit]

#and for every update run
git pull

now I have done the latest update:
--------------------------------------
xxx@yyy:~/bitHopper$ git pull https://github.com/c00w/bitHopper.git
remote: Counting objects: 5, done.
remote: Compressing objects: 100% (1/1), done.
remote: Total 3 (delta 2), reused 3 (delta 2)
Unpacking objects: 100% (3/3), done
From https://github.com/c00w/bitHopper
 * branch            HEAD       -> FETCH_HEAD
Updating 73e3015..dfa8302
Fast-forward
 index.html |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)
--------------------------------------
and when I start:
--------------------------------------
xxx@yyy:~/bitHopper$ python bitHopper.py
[22:24:18] Updating Difficulty
[22:24:18] 1690895.8030524
[22:24:18] Updating NameCoin Difficulty
[22:24:19] 94037.96
Traceback (most recent call last):
  File "bitHopper.py", line 227, in <module>
    bithopper_global = BitHopper()
  File "bitHopper.py", line 43, in __init__
    self.pool = pool.Pool(self)
  File "/home/xxx/bitHopper/pool.py", line 27, in __init__
    self.servers[pool]['default_role'] = self.servers[pool]['role']
KeyError: 'role'
---------------------------------------
the same as before...

[Tmoney]

Djex, latest from c00w.

[muyoso]

Also had Mt.Red Api disabled.  Restart fixed it.

[iopq]

why is btcworld set to info? what does that even mean? should I try enabling it?

[paraipan]

#and for every update run
git pull

now I have done the latest update:
.................................................
the same as before...

run with --debug like c00w said

xxx@yyy:~/bitHopper$ python bitHopper.py --debug

and post output here or in pastebin

[paraipan]

Also had Mt.Red Api disabled.  Restart fixed it.

seems like mtred slowed down their api update

[Clipse]

Regarding namecoin mining.

I think the approach used to mine up to 40*nmc_difficulty is wrong.

The correct approach to atleast get profitibility out of the time spend mining namecoins should be like this:

- scrape nmc/btc exchange rate, use the daily low here : https://exchange.bitparking.com/main
- scrape required difference required to be 1:1 to BTC(atleast for the purpose of mining time spend on shares) here : http://dot-bit.org/tools/nextDifficulty.php
- follow conventional 0.43*difficulty thus in this case (0.43*nmc_difficulty) * (nmc daily low / nmc required for 1:1 with btc) ie. ( 0.43 * 94037.96 ) * ( 0.02721001 / 0.0556 ) = 19789.0781969645 (we could simplify and use only the whole numbers for difficulty to make it less messy)

What we would get in this case is relatively the same outcome as what we have atm with hopping 0.43*btc_difficulty, it would obviously happen less often but while we apply this we should know that atleast whatever coins we earn we would not end up mining at higher cost than what the exchange offers. The approach works best to sell as soon as you clear any coins, at the available market price for safety.

Im sure c00w or ryo or someone would be able to apply something like this faster than me since I am no python expert other than tweaking what they allready provide "_"

[joulesbeef]

have a question.

Playing with the new phatk 2.0 kernel.. it doesnt work well right now.. doesnt work with linux or the guiminer.

It does appear to work somewhat with AOCLBF 1.75, and does report a higher mhash for me by 2% and this is over the diablo's latest modded kernel.(check it out)

Ok i havent really ever used this miner...but it seems to be reporting a much higher rejection rate than what i see reported on the web stats page and am wondering if this is just due to how the miner works now and AOCLBF is just wrong about the rejects?

[Sukrim]

Currently I'm researching autobuild mechanisms...

Alright, I'm runnning into issues with paths.

Could you please change references to index.html, stats.db and pool.cfg from:

Code:

try:
    read = parser.read(os.path.join(os.path.dirname(os.path.abspath(__file__)), 'pool.cfg'))

to (taken partially from http://en.w3support.net/index.php?db=so&id=404744)

Code:

try:
    # determine if application is a script file or frozen exe
    if hasattr(sys, 'frozen'):
        application_path = os.path.dirname(sys.executable)
    elif __file__:
        application_path = os.path.dirname(__file__)

    read = parser.read(os.path.join(application_path, 'pool.cfg'))

?

This one above goes in pool.py and for website.py there's something very similar:

Code:

try:
    # determine if application is a script file or frozen exe
    if hasattr(sys, 'frozen'):
        application_path = os.path.dirname(sys.executable)
    elif __file__:
        application_path = os.path.dirname(__file__)

    index = parser.read(os.path.join(application_path, 'index.html'))

database.py is written a bit differently - here you need:

Code:

try:
    # determine if application is a script file or frozen exe
    if hasattr(sys, 'frozen'):
        DB_DIR = os.path.dirname(sys.executable)
    elif __file__:
        DB_DIR = os.path.dirname(__file__)
    DB_DIR = os.path.dirname(os.path.abspath(__file__))

and I you also need to import sys there it seems.

Would be a great thing to have and shouldn't really impact anything.

[paraipan]

yo have it tested sukrim ? I'll make a pull req if you know it works for the purpose