python OpenCL bitcoin miner

[TheMoneyStorm]

It's something with the new connect method. If I revert to the old longPollThread it works fine.

They'll get it ironed out soon enough, hopefully.

[RaTTuS]

My backup pools with @ in them work now ... so I put it down to a temp glitch

[TheMoneyStorm]

My backup pools with @ in them work now ... so I put it down to a temp glitch

A temp glitch with some software, say it aint so . Glad it's working for you now.

[Luke-Jr]

m0mchil: Why'd you remove the retry-on-network-error code when you merged my branch? This greatly improves yield on many pools. Also, why remove GW/Efficiency? Useful data, that! Finally, when will phatk be optional?

Why would you need to retry submission? Because you either a) have connectivity problems or b) the pool is overloaded. Anyway, probability of result being valid is going down with time.

Every pool I've tried lately (BTCGuild, Bitcoins.lc, and of course Eligius) has had an occasional network error submitting blocks, even when it's not overloaded. Overload is pretty common during longpolls no matter how powerful the pool system, since it has to get new work for every single miner. What's the rationale behind giving up after one try, rather than giving it at least 5 minutes?

As for GW/Efficiency - this will soon be irrelevant because of some new protocols being developed. Even now, it assumes difficulty of 1 and will show (@ 400 Mh/s) efficiency of 50% at non-'time rolling' pools and anything above 100% (even 700%) with 'time rolling' ones (Eligius). Feel free to explain to users what/why is this. Not that it is wrong, it just doesn't make my life exactly easier.

Perhaps it should be changed from a % to a simple decimal number? Or a hidden option to enable it? --I-know-what-efficiency-is

Finally, what exactly is wrong with phatk? As far as I know, it's better on everything AMD 5xxx and up (majority of users). Nvidia users should have their own optimized miner anyway.

It uses too much system memory. In particular, I give my mining VM 256 MB RAM, and phatk just gets OOM-killed unless I add swap to it. If I go ahead and waste more RAM on it, I get a total of 1 MH/s improvement-- not worth it to me.

[Mobius]

On the latest version (yesterday) I keep getting this error on my 2nd rig Ubuntu 10.10 Cat11.6 SDK 2.4  (2nd card 5830), identical to the 1st rig and the miner just hangs there, can't ctrl-c out and if I close the terminal - the whole machine locks up. The first card (5870 - Still working)

pit.deepbit.net:8332 06/07/2011 15:19:50, long poll exception:                  
Traceback (most recent call last):
  File "/home/aa/poclbm/BitcoinMiner.py", line 338, in longPollThread
    (connection, result) = self.request(connection, url, self.headers)
  File "/home/aa/poclbm/BitcoinMiner.py", line 296, in request
    response = connection.getresponse()
  File "/usr/lib/python2.6/httplib.py", line 990, in getresponse
    response.begin()
  File "/usr/lib/python2.6/httplib.py", line 391, in begin
    version, status, reason = self._read_status()
  File "/usr/lib/python2.6/httplib.py", line 349, in _read_status
    line = self.fp.readline()
  File "/usr/lib/python2.6/socket.py", line 427, in readline
    data = recv(1)
error: [Errno 104] Connection reset by peer
pit.deepbit.net:8332 06/07/2011 15:20:40, long poll: new block XXXXXXXXXXXXXXXc
pit.deepbit.net:8332 06/07/2011 15:22:19, long poll: new block XXXXXXXXXXXXXXX8
pit.deepbit.net:8332 06/07/2011 15:27:26, long poll exception:                  
Traceback (most recent call last):
  File "/home/aa/poclbm/BitcoinMiner.py", line 338, in longPollThread
    (connection, result) = self.request(connection, url, self.headers)
  File "/home/aa/poclbm/BitcoinMiner.py", line 296, in request
    response = connection.getresponse()
  File "/usr/lib/python2.6/httplib.py", line 990, in getresponse
    response.begin()
  File "/usr/lib/python2.6/httplib.py", line 391, in begin
    version, status, reason = self._read_status()
  File "/usr/lib/python2.6/httplib.py", line 349, in _read_status
    line = self.fp.readline()
  File "/usr/lib/python2.6/socket.py", line 427, in readline
    data = recv(1)
error: [Errno 104] Connection reset by peer


Any ideas or advice?

[BOARBEAR]

please bring back the old kernel
the new kernel is very bad for my nvidia card

[zimpixa]

Miner cant connect to deepbit, but it also cant connect to backup pool cuz "username or password not correct". They are correct and miner just stuck... Restarting results in going to backup pool or giving same message about login details.

BTW, can u add to circumstances to restart/go to backup

-when message 'verification failed...'
-when wrong username or password (cuz of pool problem or just user mistake)

[IlbiStarz]

Is there not a fallback command for Linux or what? It works on Windows...

[brunoshady]

Is there not a fallback command for Linux or what? It works on Windows...

sure it have

--failback=X

[jgarzik]

Is there not a fallback command for Linux or what? It works on Windows...

Are you sure that you are passing a list of servers to "--servers" ?

[hugolp]

Im getting similar errors:

Code:

useast.btcguild.com:8332 08/07/2011 07:42:54, Setting pool xxxxxx_xxxxxxx @ useast.btcguild.com:8332
useast.btcguild.com:8332 08/07/2011 07:42:56, long poll exception:

Traceback (most recent call last):
  File "/home/ben/poclbm_old/BitcoinMiner.py", line 338, in longPollThread
    (connection, result) = self.request(connection, url, self.headers)
  File "/home/ben/poclbm_old/BitcoinMiner.py", line 299, in request
    response = connection.getresponse()
  File "/usr/lib/python2.7/httplib.py", line 1027, in getresponse
    response.begin()
  File "/usr/lib/python2.7/httplib.py", line 407, in begin
    version, status, reason = self._read_status()
  File "/usr/lib/python2.7/httplib.py", line 371, in _read_status
    raise BadStatusLine(line)
BadStatusLine: ''

api.bitcoin.cz:8332 08/07/2011 07:42:58, Setting pool xxxxxx.xxxxxx @ api.bitcoin.cz:8332
api.bitcoin.cz:8332 08/07/2011 07:42:58, Attempting to fail back to primary pool

EDIT: Also, a couple of miners have gone into a kind of a loop where it fails to connect to firs server, connects to second server and after 1 second tries to connect again to first server, fails, connects to second to second server, after 1 second tries to connect again to first server, fails,... (-failback = 1)

[IlbiStarz]

Is there not a fallback command for Linux or what? It works on Windows...

Are you sure that you are passing a list of servers to "--servers" ?

What do you mean?

[Mobius]

Is there not a fallback command for Linux or what? It works on Windows...

New version today omits the --servers= part of the command line

update via command line
$ cd poclbm
$ git pull

get the options using
$ ./poclbm --help

Update your shell scripts or command line to reflect new option structure

Are you sure that you are passing a list of servers to "--servers" ?

What do you mean?

New version today omits the --servers= part of the command line

update via command line
$ cd poclbm
$ git pull

get the options using
$ ./poclbm --help

Update your shell scripts or command line to reflect new option structure

[zimpixa]

Can enyone explain --estimate function? I dont get it. Also compiling a windows version would be nice.

[Mobius]

Can enyone explain --estimate function? I dont get it. Also compiling a windows version would be nice.

From the read me in the sources, it looks like the window of time that the average rate is calculated 900 secs = 15 min window (the last 15mins of hash rates are averaged) and displayed.

From: https://github.com/m0mchil/poclbm

[m0mchil]

New version is up

Changes:

- support for server provided failback hosts
- most of JSON-RPC fields made optional (to reduce pools bandwidth)
- increased default primary server retry interval to 10 getworks
- moved job processing to main thread
- improvements by gominoa, enolan @ github - options separation, server names, quiet output

[jgarzik]

- most of JSON-RPC fields made optional (to reduce pools bandwidth)

...and how does the client inform the server that it supports this feature?

From the server's perspective, it must support may different clients, and cannot know if 'target' or 'midstate' is optional without additional information.

[m0mchil]

...and how does the client inform the server that it supports this feature?

There is version in user-agent, server can determine miner and act accordingly.

[gominoa]

Please note syntax is updated. Use:

poclbm.py   user1:pass1@host1:port1#serv1   user2:pass2@host2:port2#serv2   user3:pass3@host3:port3#serv3

... instead of ...
poclbm.py --servers=serv1,serv2,serv3
... or ...
poclbm.py --user=x --password=x --host=x --port=x --backup=x

(miner will failover from serv1 to serv2 and serv3 if neccessary)

[jkminkov]

something is wrong with LP, I got rejected shares over 30 seconds after LP message shows up