Author

Topic: bitHopper: Python Pool Hopper Proxy - page 171. (Read 355816 times)

full member
Activity: 176
Merit: 100
July 20, 2011, 02:20:10 PM
Probably this is a known error or I am missing some lib but I get this error everytime I start the hopper then there everything works as it should or at least I think it does.
Quote
2011-07-20 14:16:49-0500 [-] Unhandled Error
        Traceback (most recent call last):
          File "bitHopper.py", line 367, in
            main()
          File "bitHopper.py", line 363, in main
            reactor.run()
          File "/opt/python2.7/lib/python2.7/site-packages/twisted/internet/base.py", line 1162, in run
            self.mainLoop()
          File "/opt/python2.7/lib/python2.7/site-packages/twisted/internet/base.py", line 1171, in mainLoop
            self.runUntilCurrent()
        --- ---
          File "/opt/python2.7/lib/python2.7/site-packages/twisted/internet/base.py", line 793, in runUntilCurrent
            call.func(*call.args, **call.kw)
          File "/root/bitHopper/pool.py", line 249, in update_api_servers
            d = work.get(self.bitHopper.json_agent,info['api_address'])
        exceptions.KeyError: 'api_address'

I am using the main branch of coow's github fully updated.

http://pastebin.com/YK3Kj3QJ
legendary
Activity: 924
Merit: 1004
Firstbits: 1pirata
July 20, 2011, 02:12:12 PM
Hey guys, I'm in a dilemma... if you don't mind merging some code on github we're hopping a little inefficient. I like all of your forks and every one has it's unique features but can't decide over which one to use. Thanks
member
Activity: 84
Merit: 10
July 20, 2011, 02:08:26 PM

Cause you didnt apply the test branch patch i told you to do.

Go here: https://github.com/ryouiki/bitHopper/commit/e8fa81053efa238c062d03603c44eeecf4a87523

You are correct, I didn't apply that.  Looks like it may be working by the first poll done.  Thanks.
hero member
Activity: 504
Merit: 502
July 20, 2011, 02:05:55 PM
Couple of feature requests:

1) Add unhoppable pools such as deepbit or slush as backups - I'd rather failback to deepbit than eligius I think. Seems deepbit has a better payout due to super short (<30 Minute) rounds on average.
2) Add web based control (in ryouiki's fork) to set backup pool - something like a B button to go along with X/O of enable/disable.

Thanks!

1. Just use arsbitcoin as backup, dont really need more than one backup.

I currently have ars and eligius as backups (as is default i believe). But the point is that deepbit seems to be more efficient than the pps from ars or eligius.

the SMPPS by arsbitcoin at 0% fee pays at a minimum 10% more than PPS from deepbit.

Not sure how you could work on 'feel' when in practice there is no chance deepbits PPS @ 10% fee will pay you more.

Regarding bitcoins.lc, I have a feeling they changed their Prop to score based Prop or something, I just got screwed over by them just finishing a 3hr 30mins block, I mind pretty much solidly for 2hrs 30ms of the block however my original estimate earnings for that period would have been 2.8BTC, and I received 0.18BTC for missing 1hr of the period ? sounds like they should be avoided.
full member
Activity: 168
Merit: 100
July 20, 2011, 01:59:20 PM
Couple of feature requests:

1) Add unhoppable pools such as deepbit or slush as backups - I'd rather failback to deepbit than eligius I think. Seems deepbit has a better payout due to super short (<30 Minute) rounds on average.
2) Add web based control (in ryouiki's fork) to set backup pool - something like a B button to go along with X/O of enable/disable.

Thanks!

1. Just use arsbitcoin as backup, dont really need more than one backup.

I currently have ars and eligius as backups (as is default i believe). But the point is that deepbit seems to be more efficient than the pps from ars or eligius.
hero member
Activity: 504
Merit: 502
July 20, 2011, 01:58:40 PM
If you followed this thread, you would find the bitcoins.lc stats fix.

Or you can goto ryo github under test branch.

I'm not seeing what the change is.

My browser says

"round_shares":"1645384"

from the stats.json.

Bithopper says they are 16k then 70k then 66k shares.  It doesn't seem to be working.  I want it to work because they seem to be the only pool solving blocks right now.

Cause you didnt apply the test branch patch i told you to do.

Go here: https://github.com/ryouiki/bitHopper/commit/e8fa81053efa238c062d03603c44eeecf4a87523
full member
Activity: 168
Merit: 100
July 20, 2011, 01:56:43 PM
I concur with muyoso - there are discrepancies between bithopper reported shares and the web reported shares.
member
Activity: 84
Merit: 10
July 20, 2011, 01:35:18 PM
If you followed this thread, you would find the bitcoins.lc stats fix.

Or you can goto ryo github under test branch.

I'm not seeing what the change is.

My browser says

"round_shares":"1645384"

from the stats.json.

Bithopper says they are 16k then 70k then 66k shares.  It doesn't seem to be working.  I want it to work because they seem to be the only pool solving blocks right now.
hero member
Activity: 504
Merit: 502
July 20, 2011, 01:21:09 PM
Couple of feature requests:

1) Add unhoppable pools such as deepbit or slush as backups - I'd rather failback to deepbit than eligius I think. Seems deepbit has a better payout due to super short (<30 Minute) rounds on average.
2) Add web based control (in ryouiki's fork) to set backup pool - something like a B button to go along with X/O of enable/disable.

Thanks!

1. Just use arsbitcoin as backup, dont really need more than one backup.
hero member
Activity: 504
Merit: 502
July 20, 2011, 01:20:31 PM
If you followed this thread, you would find the bitcoins.lc stats fix.

Or you can goto ryo github under test branch.
full member
Activity: 168
Merit: 100
July 20, 2011, 01:20:06 PM
Couple of feature requests:

1) Add unhoppable pools such as deepbit or slush as backups - I'd rather failback to deepbit than eligius I think. Seems deepbit has a better payout due to super short (<30 Minute) rounds on average.
2) Add web based control (in ryouiki's fork) to set backup pool - something like a B button to go along with X/O of enable/disable.

Thanks!
sr. member
Activity: 476
Merit: 250
moOo
July 20, 2011, 12:59:18 PM
yeah screw bitcoin.lc with a chainsaw.. i want nothing to do with them.

They ban hoppers, steal their coins, and screw with the mining stats to make sure we always hop to them.

and when you withdrawn your coin, all you can get out is in increments of 0.01 coins.. the change just annoyingly sits there.
member
Activity: 78
Merit: 10
July 20, 2011, 12:57:36 PM

as for x8s, they added json yesterday.
recent main branch of my fork may work on x8s, but not tested yet..

Works very well. Added to pool.py literally 5 minutes before they found the block Cheesy
full member
Activity: 168
Merit: 100
July 20, 2011, 12:56:44 PM
looks like crazy crap.
sr. member
Activity: 476
Merit: 250
moOo
July 20, 2011, 12:51:04 PM
Quote
i hope u are mining on bitcoins.lc guys ....


WHY?!?!


they have been annoying the crap out of me since i discovered them.

are you actually working on new blocks or are they doing that crazy crap with their stats still.. to make us mine there?
full member
Activity: 196
Merit: 100
July 20, 2011, 12:22:10 PM
1) API polling and ozco getting disabled?
There was a bug in an older version which caused things to get disabled too soon.

2) Once the api is disabled the pool isn't?
Um yeah. I need to call a server_update. Forgot about that. My bad.

3)x8s?
I'll finish it.
newbie
Activity: 33
Merit: 0
July 20, 2011, 11:23:10 AM

Bit unrelated to recent discussions but have you figured out what bitcoinpool is doing to their shares yet, or atleast which shares to work from. json shares looks more realistic but it seems they payout based on the website shares/elapsedtime and those stats are totally out of wack.

Also this is prob more directed to c00w/flower/ryo, any idea yet how to capture the counting current round_shares for x8s ?

http://www.bitcoinpool.com/forum/viewtopic.php?f=1&t=103

they say, they would reduce our earning based on mining time window.
in fact, we can beat this kind of algorithm with forced (scheduled) hopping

as for x8s, they added json yesterday.
recent main branch of my fork may work on x8s, but not tested yet..

Code:
                'x8s':{ 'name': 'btc.x8s.de',
                    'mine_address': 'pit.x8s.de:8337', 'user': x8s_user, 'pass': x8s_pass,
                    'api_address':'http://btc.x8s.de/api/global.json', 'role':'disable'},

Code:
    def x8s_sharesResponse(self, response):
        round_shares = int(json.loads(response)['round_shares'])
        self.UpdateShares('x8s',round_shares)

Code:
            'x8s':self.x8s_sharesResponse,
member
Activity: 66
Merit: 10
July 20, 2011, 11:17:43 AM
Something is obviously off about the math I used before or how I understand the polling refresh rate increments, cause I just had ozco get disabled because of timeouts, it seems. It was significantly less time than 5 1/2 hours after I last started bitHopper.

However, once ozco was disabled, bitHopper didn't immediately switch to another pool, either. It did eventually switch over to a backup pool, but I would think it should immediately switch, right?
newbie
Activity: 33
Merit: 0
July 20, 2011, 11:00:37 AM
i hope u are mining on bitcoins.lc guys ....


@ryouiki i cant disable manually ars ffs

in the latest version (master branch), you can manually remove ars ( 'role':'removefromlist' ) by modifying pool.py
member
Activity: 66
Merit: 10
July 20, 2011, 10:48:44 AM
2) So what have you been doing over the last two days?
Database changes, Stability changes, code reorg, a flat info page, dynamic api polling changes, oh and the difficulty module now will deal with changing difficulty correctly.
So the main code is really solid now.

About the dynamic api polling - if shares remain the same from previous poll to current poll, refresh rate is incremented by 10% of current rate, correct? (If I read the code correctly.) Once refresh rate exceeds 1800 seconds, the pool is disabled. With polling starting at 60 seconds, if a pool's shares became static, it would take 5 1/2 hours before that pool is disabled. I got that from summing the incremental increases from 60 to 1854 over 37 10% steps.

Am I missing something? Is my math wrong? I wouldn't think we would want to wait 5 1/2 hours before changing from a low-ball static round share api poll.
Jump to: