[ANN][BLC] Blakecoin Blake-256 for GPU/FPGA With Merged Mined Pools Stable Net

[kramble]

I am still using your python Blakecoin miner for the Lancelot boards and set that to diff 1 works a treat on the pools

That's good to know. I've actually switched to using cgminer. It needs to be compiled specifically for the lancelot via #define LANCELOT_52 in driver-icarus.c ... I was lazy and didn't bother to make it a command line option or rebuild the bitstream for the CM1-style protocol). Anyway its working fine, with the added bonus of automatic failover.

I've still got the problem of the raspberry pi network stack hanging roughly once a day, but that's been solved with a watchdog and automatic reboot, so its been running fine for the last few weeks (I even flashed the lancelot MCS bitsteam, so it should survive a power outage, not that we get many of them here).

Jude: Good luck with X6500, hopefully we can get this working soon.

[Jude Austin]

Ugh, now I can't get mine.bat to work.

Before I just spammed it a couple times but now nothing.

Run Summary:
-------------
Traceback (most recent call last):
  File "mine.py", line 271, in <module>
    logger.printSummary(settings)
  File "C:\Users\block_000\Desktop\x6500-miner-master\ConsoleLogger.py", line 21
0, in printSummary
    self.say('Device: %d' % self.devicenum, True, True)
AttributeError: 'ConsoleLogger' object has no attribute 'devicenum'

[Jude Austin]

Ah ha!

Unplugged the X6500, plugged it back in, now it works, haha.

But nothing but rejected:

2014-02-03 12:30:34 | Device 0 opened (A5VOPER6)
2014-02-03 12:30:34 | Connected to 2 FPGAs
2014-02-03 12:30:35 | FPGA 0 is running at 100MHz
2014-02-03 12:30:35 | FPGA 1 is running at 100MHz
2014-02-03 12:30:35 | Connected to server
2014-02-03 12:30:37 | rejected 4e7dae8eL
2014-02-03 12:30:41 | rejected 24330c9aL
2014-02-03 12:30:43 | rejected 335cff23L
2014-02-03 12:30:43 | rejected 74275a99L
2014-02-03 12:30:48 | rejected 4e7dae8eL
2014-02-03 12:30:51 | rejected 24330c9aL
2014-02-03 12:30:54 | rejected 335cff23L
2014-02-03 12:30:54 | rejected 74275a99L
2014-02-03 12:31:36 | rejected 30ca017fL
2014-02-03 12:31:38 | rejected 7b92ffe3L
2014-02-03 12:31:46 | rejected 272f58c7L
2014-02-03 12:31:47 | rejected 30ca017fL
2014-02-03 12:31:49 | rejected 7b92ffe3L
2014-02-03 12:32:01 | rejected 417a37ceL
0 kH/s | 0/14/0 100.00%/0.00%

[BlueDragon747]

Ah ha!

Unplugged the X6500, plugged it back in, now it works, haha.

But nothing but rejected:

2014-02-03 12:30:34 | Device 0 opened (A5VOPER6)
2014-02-03 12:30:34 | Connected to 2 FPGAs
2014-02-03 12:30:35 | FPGA 0 is running at 100MHz
2014-02-03 12:30:35 | FPGA 1 is running at 100MHz
2014-02-03 12:30:35 | Connected to server
2014-02-03 12:30:37 | rejected 4e7dae8eL
2014-02-03 12:30:41 | rejected 24330c9aL
2014-02-03 12:30:43 | rejected 335cff23L
2014-02-03 12:30:43 | rejected 74275a99L
2014-02-03 12:30:48 | rejected 4e7dae8eL
2014-02-03 12:30:51 | rejected 24330c9aL
2014-02-03 12:30:54 | rejected 335cff23L
2014-02-03 12:30:54 | rejected 74275a99L
2014-02-03 12:31:36 | rejected 30ca017fL
2014-02-03 12:31:38 | rejected 7b92ffe3L
2014-02-03 12:31:46 | rejected 272f58c7L
2014-02-03 12:31:47 | rejected 30ca017fL
2014-02-03 12:31:49 | rejected 7b92ffe3L
2014-02-03 12:32:01 | rejected 417a37ceL
0 kH/s | 0/14/0 100.00%/0.00%

which pool? or is that solo ?

[Jude Austin]

Solo!

Haha.

[BlueDragon747]

Solo!

Haha.

try getwork is on port 8337
http://eu1.blakecoin.com
http://ny1.blakecoin.com

might be quicker than waiting for a block find?

[Jude Austin]

Solo!

Haha.

try getwork is on port 8337
http://eu1.blakecoin.com
http://ny1.blakecoin.com

might be quicker than waiting for a block find?

Tried that too:

2014-02-03 12:44:16 | Device 0 opened (A5VOPER6)
2014-02-03 12:44:16 | Connected to 2 FPGAs
2014-02-03 12:44:16 | FPGA 0 is running at 100MHz
2014-02-03 12:44:16 | FPGA 1 is running at 100MHz
2014-02-03 12:44:16 | Connected to server
2014-02-03 12:44:16 | 0: Error getting work! Retrying...
2014-02-03 12:44:17 | 1: Error getting work! Retrying...
2014-02-03 12:44:17 | 0: Error getting work! Retrying...
2014-02-03 12:44:17 | 1: Error getting work! Retrying...
2014-02-03 12:44:17 | 0: Error getting work! Retrying...
2014-02-03 12:44:17 | 1: Error getting work! Retrying...
2014-02-03 12:44:17 | 0: Error getting work! Retrying...
2014-02-03 12:44:17 | 1: Error getting work! Retrying...
2014-02-03 12:44:17 | Long-poll: connected to ny1.blakecoin.com:8337
2014-02-03 12:44:17 | 0: Error getting work! Retrying...
2014-02-03 12:44:17 | 1: Error getting work! Retrying...
2014-02-03 12:44:18 | 0: Error getting work! Retrying...
2014-02-03 12:44:18 | 1: Error getting work! Retrying...
2014-02-03 12:44:18 | 0: Error getting work! Retrying...
2014-02-03 12:44:18 | 1: Error getting work! Retrying...
2014-02-03 12:44:18 | 0: Error getting work! Retrying...
2014-02-03 12:44:18 | 1: Error getting work! Retrying...
2014-02-03 12:44:18 | 0: Error getting work! Retrying...
2014-02-03 12:44:18 | 1: Error getting work! Retrying...
2014-02-03 12:44:19 | 0: Error getting work! Retrying...
2014-02-03 12:44:19 | 1: Error getting work! Retrying...
2014-02-03 12:44:19 | 0: Error getting work! Retrying...
2014-02-03 12:44:19 | 1: Error getting work! Retrying...
2014-02-03 12:44:19 | 0: Error getting work! Retrying...
2014-02-03 12:44:19 | 1: Error getting work! Retrying...
2014-02-03 12:44:19 | 0: Error getting work! Retrying...
2014-02-03 12:44:19 | 1: Error getting work! Retrying...
2014-02-03 12:44:20 | 0: Error getting work! Retrying...
2014-02-03 12:44:20 | 1: Error getting work! Retrying...
2014-02-03 12:44:20 | 0: Error getting work! Retrying...
2014-02-03 12:44:20 | 1: Error getting work! Retrying...
2014-02-03 12:44:20 | 0: Error getting work! Retrying...
2014-02-03 12:44:20 | 1: Error getting work! Retrying...
2014-02-03 12:44:20 | 0: Error getting work! Retrying...
2014-02-03 12:44:20 | 1: Error getting work! Retrying...
2014-02-03 12:44:20 | 0: Error getting work! Retrying...
2014-02-03 12:44:20 | 1: Error getting work! Retrying...
2014-02-03 12:44:21 | 0: Error getting work! Retrying...
2014-02-03 12:44:21 | 1: Error getting work! Retrying...
2014-02-03 12:44:21 | 0: Error getting work! Retrying...
2014-02-03 12:44:21 | 1: Error getting work! Retrying...
2014-02-03 12:44:21 | 0: Error getting work! Retrying...
2014-02-03 12:44:21 | 1: Error getting work! Retrying...
2014-02-03 12:44:21 | 0: Error getting work! Retrying...
2014-02-03 12:44:21 | 1: Error getting work! Retrying...
2014-02-03 12:44:21 | 0: Error getting work! Retrying...
2014-02-03 12:44:22 | Exiting...

[BlueDragon747]

Solo!

Haha.

try getwork is on port 8337
http://eu1.blakecoin.com
http://ny1.blakecoin.com

might be quicker than waiting for a block find?

Tried that too:

got a Lancelot mining that pool with kramble's python Blakecoin miner and can getwork somethings not setup right  

are you on a win x86 or x64?

[Jude Austin]

Solo!

Haha.

try getwork is on port 8337
http://eu1.blakecoin.com
http://ny1.blakecoin.com

might be quicker than waiting for a block find?

Tried that too:

got a Lancelot mining that pool with kramble's python Blakecoin miner and can get work somethings not setup right  

are you on a win x86 or x64?

Windows 8.1 x64

I am thinking Windows is the issue.

[BlueDragon747]

Solo!

Haha.

try getwork is on port 8337
http://eu1.blakecoin.com
http://ny1.blakecoin.com

might be quicker than waiting for a block find?

Tried that too:

got a Lancelot mining that pool with kramble's python Blakecoin miner and can get work somethings not setup right  

are you on a win x86 or x64?

Windows 8.1 x64

I am thinking Windows is the issue.

yeah could be issues with the x64 drivers

any old xp x86 laptops or spare machines about you can test on?

or run a vm?

[Jude Austin]

Solo!

Haha.

try getwork is on port 8337
http://eu1.blakecoin.com
http://ny1.blakecoin.com

might be quicker than waiting for a block find?

Tried that too:

got a Lancelot mining that pool with kramble's python Blakecoin miner and can get work somethings not setup right  

are you on a win x86 or x64?

Windows 8.1 x64

I am thinking Windows is the issue.

yeah could be issues with the x64 drivers

any old xp x86 laptops or spare machines about you can test on?

Yep, I will set one up right now.

[kramble]

The lancelot miner uses ServiceProxy for the network stuff, while it looks like the x6500 miner uses httplib (see rpcClient.py).

Unfortunately my python-fu is not strong, so I'm not sure how we'd go about debugging this, other than the obvious initial step of printing out the various parameters/results in getwork() and possibly longpoll_loop() too.

[kramble]

I did a little debugging on an old test version of the miner I had hacked up earlier, and I get errors too using ...

python mine.py -u http://eu1.blakecoin.com:8337 -w user.worker:password

Try this though, it seems to do something (NB This OMITS the http://) ...

python mine.py -u eu1.blakecoin.com:8337 -w user.worker:password

It still says error though, so maybe it's still not right (the pools doesn't seem to care if the username exists either!)

Code:

# python mine.py -u eu1.blakecoin.com:8337 -w user.worker:password
getwork loop
getnewjob
IN getwork
Connecting to server...
IN CONNECT
<httplib.HTTPConnection instance at 0x0167BB98>
2014-02-03 19:36:22 | Connected to server
0 kH/s | 0/0/0 0.00%/0.00%send: 'POST / HTTP/1.1\r\nHost: eu1.blakecoin.com:8337
\r\nAccept-Encoding: identity\r\nContent-Length: 49\r\nContent-Type: application
/json\r\nAuthorization: Basic dXNlci53b3JrZXI6cGFzc3dvcmQ=\r\nUser-Agent: x6500-
miner\r\n\r\n{"params": [], "method": "getwork", "id": "json"}'
getjob
reply: 'HTTP/1.1 200 OK\r\n'
header: X-Source-Code: /src/
header: Content-Length: 549
header: X-Roll-NTime: expire=120
header: X-Long-Polling: /LP
header: Server: Eloipool
header: Date: Mon, 03 Feb 2014 19:34:50 GMT
header: Content-Type: application/json
(<httplib.HTTPConnection instance at 0x0167BB98>, {u'error': None, u'result': {u
'hash1': u'000000000000000000000000000000000000000000000000000000000000000000000
08000000000000000000000000000000000000000000000000000010000', u'data': u'0000000
2e840bb8958e5479add63a8c96367586615fe71cc90b9bfa3000ab97d00000000dc97f2f5c9ab5bb
acdd27b3cdc9314d6b925554a077a3426dbb73731b3cf364252efef5a1b07fd6f523130356b00000
08000000000000000000000000000000000000000000000000000000000000000000000000000000
00080020000', u'target': u'fffffffffffffffffffffffffffffffffffffffffffffffffffff
fff00000000', u'submitold': True}, u'id': u'json'})
2014-02-03 19:36:23 | -1: Error getting work! Retrying...

EDIT: Actually it may be OK, the "Error getting work" was coming from fpga.putJob(work) which not surprisingly fails as I've got no FPGA attached (various other hacks were required to get it to run this far without a device).

[Jude Austin]

I give up!

Haha, setup Python 2.7 with PyUSB, installed FPGAMining's driver.

Had to find libusb.0.DLL and associate it with the driver and program.py still says: Error: FT232R device not opened!

[BlueDragon747]

Code:

# CONFIGURATION - SOLO MINING via blakecoind
host = "ny1.blakecoin.com"	# Set hostname here (usually localhost)
#host = "eu1.blakecoin.com"	# Set hostname here (usually localhost)
user = "UserName.WorkerName"
password = "WorkerPass"
http_port = "8337"	# Getwork port

# CONFIGURATION - CHANGE THIS (eg try COM1, COM2, COM3, COM4 etc)
serial_port = "COM7"
# serial_port = "/dev/ttyUSB0"	# raspberry pi

# CONFIGURATION - how often to refresh work
askrate = 1

this is what I have working for Lancelot in your other miner

you can use use any username.worker but unless it is registered I will get accounted for block find by default, not shares just finds

[kramble]

this is what I have working for Lancelot in your other miner

Yep, though interestingly towards the bottom of the script we have ...

Code:

url = 'http://' + user + ':' + password + '@' + host + ':' + http_port

bitcoin = ServiceProxy(url)

And the rpcClient.py does this ...

ret = connector(host, strict=True, timeout=timeout)

So I think the host needs to be without the http:// prefix.

Could you (Jude) just confirm the command line you are using? And try it without the http:// prefix to ny1.blakecoin.com ?

HMM, then again I see that the longpoll did connect sucessfully in your log, so maybe this is just a red herring here?

[Jude Austin]

this is what I have working for Lancelot in your other miner

Yep, though interestingly towards the bottom of the script we have ...

Code:

url = 'http://' + user + ':' + password + '@' + host + ':' + http_port

bitcoin = ServiceProxy(url)

And the rpcClient.py does this ...

ret = connector(host, strict=True, timeout=timeout)

So I think the host needs to be without the http:// prefix.

Could you (Jude) just confirm the command line you are using? And try it without the http:// prefix to ny1.blakecoin.com ?

HMM, then again I see that the longpoll did connect sucessfully in your log, so maybe this is just a red herring here?

I have tried it with both.

Nothing changes.

Here is what I am using in the batch file: C:\Python27\python.exe program.py X6500-Robust-v04-2core-fmax-100MHz.bit

[kramble]

OK, could you add this at line 135 of rpcClient.py

         print (connection, result)

Just prior to

         return (connection, result['result'])

If you don't know python, then be very careful with the edit. The idiots who designed the language made whitespace part of the syntax, so you need to copy the spacing exactly (and its tabs not spaces in this case), so just copy the following line (return) and use its whitespace prefix as a template.

You can also try uncommenting the connection.set_debuglevel(1) a few lines further up.

[Jude Austin]

OK, could you add this at line 135 of rpcClient.py

         print (connection, result)

Just prior to

         return (connection, result['result'])

If you don't know python, then be very careful with the edit. The idiots who designed the language made whitespace part of the syntax, so you need to copy the spacing exactly (and its tabs not spaces in this case), so just copy the following line (return) and use its whitespace prefix as a template.

You can also try uncommenting the connection.set_debuglevel(1) a few lines further up.

I will try shortly.

My kids are driving me fucking nuts right now, their mother should be home soon.

[kramble]

I will try shortly.

My kids are driving me fucking nuts right now, their mother should be home soon.

No problem, its still early evening here so I've got a while.

I think I may have found the problem (other than the http:// prefix). The x6500 miner is expecting the pool to send midstate, which is missing. I think we can get away with commenting out line 436 of fpga.py vis:
      #job.midstate = work['midstate']

Or if that doesn't work, set it to something benign, eg
      job.midstate = work['target']