**US** BitFury Setup Guide

[chamber32]

On a 3.0 M-Card, which way is the RPi oriented, board up or hanging down?

EDIT:  Never mind, I found the pin 1 markings. (board up, BTW)

thx,

c32

[BTClobsta]

NOTE

Due to my personal epic screw-up the SD image does not contain a working binary. And on top of this some orders were shipped without instructions also We have whipped ourselves vigorously here over the weekend and will try to improve our process so these bloopers don't happen in the future.

Here's how to get your miner working:
1. Check your router for bitfury IP
2. SSH into the raspberry pi (username pi password raspberry)
3. Run following commands:

Code:

cd /opt/bitfury/chainminer
sudo make clean
sudo make
sudo killall screen
sudo /etc/rc.local

You should now be mining. You can check the web interface of the miner with a normal browser and setup your pools there (use the IP that you used for SSH).

after followed the instructions here and for some reason my miner would only load up to the login and password for the rpi and then turn off. the strange thing is that it didn't do this before i followed the instructions. did i do something wrong?

[BTClobsta]

i've also notice that my hboards are getting really hot even when i'm not mining.

miner: no process found

what does that mean. is something wrong with my sd?

[klondike_bar]

ive got an issue: all my chips are reading "speed->0" or "turn off" within a few minutes of zero hashing.

i have tried several restarts, stop/start miner commands, and verified the proxy id is correct. The issue happened on 2 h-boards at the same time, and suddenly after days of stable hashing

edit: modified the second worker to use port 3334 (ghash.io) and left the first at port 3333 - looks like its ramping back up slowly

edit 2: great -after a few minutes the cards read 51GH (quite low), and the pool topped at 2.5GH. Its now dropped back to zero again, whats going on? (I have a usb asic on the pool just fine)

edit 3: ive decided to spread my effort
worker 1- ghash.io port 3333    - nothing seen at pool
worker 2 - ghash.io port 3334 - nothing seen at pool
worker 3 - mint.bitminter.com - briefly 9GH, now nothing?

the reported hashrate at the bitfury webUI started at 70Gh and over 5min has slowly dropped to 25Gh. wtf is going on here?

[tom99]

ive got an issue: all my chips are reading "speed->0" or "turn off" within a few minutes of zero hashing.

i have tried several restarts, stop/start miner commands, and verified the proxy id is correct. The issue happened on 2 h-boards at the same time, and suddenly after days of stable hashing

edit: modified the second worker to use port 3334 (ghash.io) and left the first at port 3333 - looks like its ramping back up slowly

edit 2: great -after a few minutes the cards read 51GH (quite low), and the pool topped at 2.5GH. Its now dropped back to zero again, whats going on? (I have a usb asic on the pool just fine)

edit 3: ive decided to spread my effort
worker 1- ghash.io port 3333    - nothing seen at pool
worker 2 - ghash.io port 3334 - nothing seen at pool
worker 3 - mint.bitminter.com - briefly 9GH, now nothing?

the reported hashrate at the bitfury webUI started at 70Gh and over 5min has slowly dropped to 25Gh. wtf is going on here?

   take bad hboard outside for few minutes then put back and run it.

ps: I got 2 hboards did same to me.

[klondike_bar]

ive got an issue: all my chips are reading "speed->0" or "turn off" within a few minutes of zero hashing.

i have tried several restarts, stop/start miner commands, and verified the proxy id is correct. The issue happened on 2 h-boards at the same time, and suddenly after days of stable hashing

edit: modified the second worker to use port 3334 (ghash.io) and left the first at port 3333 - looks like its ramping back up slowly

edit 2: great -after a few minutes the cards read 51GH (quite low), and the pool topped at 2.5GH. Its now dropped back to zero again, whats going on? (I have a usb asic on the pool just fine)

edit 3: ive decided to spread my effort
worker 1- ghash.io port 3333    - nothing seen at pool
worker 2 - ghash.io port 3334 - nothing seen at pool
worker 3 - mint.bitminter.com - briefly 9GH, now nothing?

the reported hashrate at the bitfury webUI started at 70Gh and over 5min has slowly dropped to 25Gh. wtf is going on here?

  take bad hboard outside for few minutes then put back and run it.

ps: I got 2 hboards did same to me.

I dont think thats the issue - i already pulled the boards and downvolted them a little to 0.81V o see if they were getting voltage shutdowns/issues - no luck

Im at wits end and i hate to go to bed knowing my machine is doing nothing  I have restarted my router, my modem, my router again, the bitfury system several times, and the issue is the same:
- the bitfury webUI page is accessible
- Ive tried a few different pools (briefly had it work for about 1 minute then it stopped again)
- ive start/stop miner several times and confirmed all proxy info
- router is forwarding ports 3330-3340

EDIT: now its working again (cross fingers). I set ghash as worker 1 and slush as worker 2. both report hashrate. However, the proxy info is the same so i dont understand what possibly changed

[BTClobsta]

is there any live chat like maybe irc or something. i don't have a lot of time and as days goes by i lose out on the chance to mine.

[CoinHoarder]

is there any live chat like maybe irc or something. i don't have a lot of time and as days goes by i lose out on the chance to mine.

on the bottom of the bitfury user guide, it has an irc channel/server listed for support. OI can't think of it off the top of my head though.

Find the user guide under "mining support" sub forum in the OP of the thread: [guide] bitfury support/tuning

[klondike_bar]

i've also notice that my hboards are getting really hot even when i'm not mining.

miner: no process found

what does that mean. is something wrong with my sd?

i think you did the chainminer update wrong. without going back to dissect each change you made, heres the best way to start from scratch:

1) download the newest SD image for your version of m-board
2) connect a screen and keyboard and start up the system. login as pi/raspberry
3) on first boot it should ask you if you want to change the ip. check your router (usually 192.168.X.1) and set the bitfury to 192.168.X.249   (x being a number, likely 1 or 0)
4) set up pool info via ssh or webUI
5) you should be mining already. to update the chainminer to improve the error rates:

ChainMiner for M Board Version 2.X boards
DO not use this guide for Version 1x m-boards.
It will fry your chips.  This is only meant for Version 2 of the M-board.
I apologize for grammar as I took 2 sleeping pills and I can barely feel my fingers.

1.login as pi or root

2.  type "nano /run/shm/.stat.log"   if you wish to see the performance of all the chips on your board before upgrading chain miner.
(optional step)
3.(backup your chainminer version)  (this creates a copy of the folder)
a. sudo cp -a /opt/bitfury /usr/bitfury.backup
b. cd /opt/bitfury/
c. pwd  (make sure you are in /opt/bitfury)
d. ls -al (or type dir)
e. rm -rf chainminer  (this removes everything in the chainminer folder and the chainminer folder itself)
4.  make sure you are in the /opt/bitfury directory and not in /opt/bitfury/chainminer (it should be deleted anyway
5.  type "git clone https://github.com/bfsb/chainminer.git"
6.  it should now start getting the chainminer git.
7.  now type "cd chainminer"  you should now be in the chainminer directory
8.  type "make"
9.  Its going to take 5-10 minutes to create the chainminer...maybe longer.
 I got some warning but everything was fine after it all finished.
 it will go back to a  bash prompt when done.
9a.  Just wait...this is the longest part of doing this change..
10.  sudo reboot

[klondike_bar]

is there any live chat like maybe irc or something. i don't have a lot of time and as days goes by i lose out on the chance to mine.

on the bottom of the bitfury user guide, it has an irc channel/server listed for support. OI can't think of it off the top of my head though.

Find the user guide under "mining support" sub forum in the OP of the thread: [guide] bitfury support/tuning

bfsbsupport on freenode IRC

[shade88]

Since chainminer is not working for me, I got the new M-Board works with bfgminer and minepeon. The problem is, it need "sudo" to make SPI available for bfgminer. Which mean I need to type in password every time bfgminer starts. I've never works with linux before, any idea how I can skip the password entering every time bfgminer start?

Code:

#!/bin/bash
sleep 10
/usr/bin/screen -dmS miner sudo /opt/minepeon/bin/bfgminer -S bfsb:auto -c /opt/minepeon/etc/miner.conf

Thanks

[BTClobsta]

i've also notice that my hboards are getting really hot even when i'm not mining.

miner: no process found

what does that mean. is something wrong with my sd?

i think you did the chainminer update wrong. without going back to dissect each change you made, heres the best way to start from scratch:

1) download the newest SD image for your version of m-board
2) connect a screen and keyboard and start up the system. login as pi/raspberry
3) on first boot it should ask you if you want to change the ip. check your router (usually 192.168.X.1) and set the bitfury to 192.168.X.249   (x being a number, likely 1 or 0)
4) set up pool info via ssh or webUI
5) you should be mining already. to update the chainminer to improve the error rates:

ChainMiner for M Board Version 2.X boards
DO not use this guide for Version 1x m-boards.
It will fry your chips.  This is only meant for Version 2 of the M-board.
I apologize for grammar as I took 2 sleeping pills and I can barely feel my fingers.

1.login as pi or root

2.  type "nano /run/shm/.stat.log"   if you wish to see the performance of all the chips on your board before upgrading chain miner.
(optional step)
3.(backup your chainminer version)  (this creates a copy of the folder)
a. sudo cp -a /opt/bitfury /usr/bitfury.backup
b. cd /opt/bitfury/
c. pwd  (make sure you are in /opt/bitfury)
d. ls -al (or type dir)
e. rm -rf chainminer  (this removes everything in the chainminer folder and the chainminer folder itself)
4.  make sure you are in the /opt/bitfury directory and not in /opt/bitfury/chainminer (it should be deleted anyway
5.  type "git clone https://github.com/bfsb/chainminer.git"
6.  it should now start getting the chainminer git.
7.  now type "cd chainminer"  you should now be in the chainminer directory
8.  type "make"
9.  Its going to take 5-10 minutes to create the chainminer...maybe longer.
 I got some warning but everything was fine after it all finished.
 it will go back to a  bash prompt when done.
9a.  Just wait...this is the longest part of doing this change..
10.  sudo reboot

how do i get the v3.0 mboard image file?

[Keefe]

NOTE

Due to my personal epic screw-up the SD image does not contain a working binary. And on top of this some orders were shipped without instructions also We have whipped ourselves vigorously here over the weekend and will try to improve our process so these bloopers don't happen in the future.

Here's how to get your miner working:
1. Check your router for bitfury IP
2. SSH into the raspberry pi (username pi password raspberry)
3. Run following commands:

Code:

cd /opt/bitfury/chainminer
sudo make clean
sudo make
sudo killall screen
sudo /etc/rc.local

You should now be mining. You can check the web interface of the miner with a normal browser and setup your pools there (use the IP that you used for SSH).

after followed the instructions here and for some reason my miner would only load up to the login and password for the rpi and then turn off. the strange thing is that it didn't do this before i followed the instructions. did i do something wrong?

Sounds like your power supply is shutting off due to overload as soon as the cards begin hashing? Are you sure it is capable of providing enough 12v power to the particular connectors you plugged into the rig?

[BTClobsta]

NOTE

Due to my personal epic screw-up the SD image does not contain a working binary. And on top of this some orders were shipped without instructions also We have whipped ourselves vigorously here over the weekend and will try to improve our process so these bloopers don't happen in the future.

Here's how to get your miner working:
1. Check your router for bitfury IP
2. SSH into the raspberry pi (username pi password raspberry)
3. Run following commands:

Code:

cd /opt/bitfury/chainminer
sudo make clean
sudo make
sudo killall screen
sudo /etc/rc.local

You should now be mining. You can check the web interface of the miner with a normal browser and setup your pools there (use the IP that you used for SSH).

after followed the instructions here and for some reason my miner would only load up to the login and password for the rpi and then turn off. the strange thing is that it didn't do this before i followed the instructions. did i do something wrong?

Sounds like your power supply is shutting off due to overload as soon as the cards begin hashing? Are you sure it is capable of providing enough 12v power to the particular connectors you plugged into the rig?

never got the chance to even hash yet. so i don't know why it is using so much power.
i first used a 500w psu. it only had one pci-e cable. so i added one more by connecting 2 IDE 4pin to 1 pci-e cable. (please keep in mind that i have the v3.0 mboard) it was only able to power on 12 hboards. it turned on for about 5min (not mining/hashing). then the psu sparked up and turned off. was never able to get it running again with that psu. so i went ahead and bought 700w psu (12v is at 570w max) with 2 pci-e cables to replace the 500w psu. surly that should be enough power. it would only power on 13 of the hboards. hey i'm able to power on 1 more now.
but i really don't know what i might be doing wrong. is it the hardware? is it he software? i really don't know.

[BenTuras]

but i really don't know what i might be doing wrong. is it the hardware? is it he software? i really don't know.

Sounds to me your power supply is too weak. Go for at least 1000W and try again, or lower the voltage on all boards and try with your current power supply.

[BTClobsta]

but i really don't know what i might be doing wrong. is it the hardware? is it he software? i really don't know.

Sounds to me your power supply is too weak. Go for at least 1000W and try again, or lower the voltage on all boards and try with your current power supply.

how much power does this thing take? i went from 500w to 700w. i guess i'll try the 1000w this time

[Swimmer63]

I have 2 h-cards, v2.2 available.  I can't use them.
https://bitcointalk.org/index.php?topic=333237.0

[klondike_bar]

but i really don't know what i might be doing wrong. is it the hardware? is it he software? i really don't know.

Sounds to me your power supply is too weak. Go for at least 1000W and try again, or lower the voltage on all boards and try with your current power supply.

how much power does this thing take? i went from 500w to 700w. i guess i'll try the 1000w this time

at 400GH for the system - roughly 400W
at 500GH - roughly 550W
at 550GH - likely 600-650W
at 600GH - could be as high as 700W
(*values are very broad approximations based on some of the dispersed info and kill-a-watt tested rigs I've seen so far in the forums: +/-10% is likely*)

remember this is power across the 12V rail(s). A 700W PSU may only provide 500W on the 12V PCI-E rail(s). If your cards are the new higher stock speeds (32-37GH/card), then its quite likely you are drawing at least 500W on the PCI-e cables. A 700W PSU is likely insufficient

[BTClobsta]

but i really don't know what i might be doing wrong. is it the hardware? is it he software? i really don't know.

Sounds to me your power supply is too weak. Go for at least 1000W and try again, or lower the voltage on all boards and try with your current power supply.

how much power does this thing take? i went from 500w to 700w. i guess i'll try the 1000w this time

at 400GH for the system - roughly 400W
at 500GH - roughly 550W
at 550GH - likely 600-650W
at 600GH - could be as high as 700W
(*values are very broad approximations based on some of the dispersed info and kill-a-watt tested rigs I've seen so far in the forums: +/-10% is likely*)

remember this is power across the 12V rail(s). A 700W PSU may only provide 500W on the 12V PCI-E rail(s). If your cards are the new higher stock speeds (32-37GH/card), then its quite likely you are drawing at least 500W on the PCI-e cables. A 700W PSU is likely insufficient

i just picked up a 1200w psu. with a total of 1050w to the 12v pci-e. that should be enough now. the only problem i'm having now it the fact that i can't get my last 3 hboards to hash. so far i got 13 to hash. but the last 3 just don't want to work. i'm thinking the two pci-e cables connected to the v3.0 mboards might not be enough power. i think i need to add additional power other then just from the two pci-e cables. i know the v3.0 boards can be powered through the 2 pci-e plugs, but is it possible that it isn't enough power. do you think i might need to also add some 12v and ground to the metal screws? can i use both pci-e plugs along with adding additional power to the metal screws?

[ericdc30]

For power supplies stick to Corsair. When they say 750 or 850 Watts they mean 750 or 850 Watts output on one 12v rail (you should still use both PCIE 6 pin connectors).