Pages:
Author

Topic: BAMT - Easy persistent USB key based linux for dedicated miners/mining farms - page 43. (Read 167468 times)

donator
Activity: 798
Merit: 500
Can we get mineco.in added to API support next update?  I've been dealing with some latency issues causing frequent disconnects and noticed the pools won't loop back - if primary and secondary pools fail it stops mining.  My workaround was to just put 500 lines of the same pool in the pool file - which is working great - unlike my other rig (not BAMT) which has taken countless hours of work to with no success.  It took less than an hour to test and reconfigure the BAMT rig to compensate and it just keeps mining, still losing time in reconnects but that's the ISP.  Thanks again for the work, another donation on it's way as soon as my earning from mineco's 11.9 million share block clear.  By the way removing the -P 2 from the doOC routine works great on my 6950's, didn't even have to add anything to .conf file.
donator
Activity: 1731
Merit: 1008

I'M using /Example-bamt.conf and I always get Code: YAML_PARSE_ERR_INCONSISTENT_INDENTATION
Line: 56

there is no newline at end of  example conf

linking to plain text version,,,  and having two or 5 gpu block alrealy filled would stop people like me from messing with the YAML systax / indentation.

Thanks

I often get "Starting mining processes...: mineNo protocol specified"  what could that mean ?
newbie
Activity: 73
Merit: 0
Did anybody succeed in setting up wireless usb adapter with BAMT?
full member
Activity: 196
Merit: 100
Oikos.cash | Decentralized Finance on Tron
BAMT is wonderful! Easy to set up, stable to run.

Aaron, I don't have an easy way to send Bitcoin, but if you will PM me a PayPal address I'll send along some US$. Thanks for all your work on BAMT.
full member
Activity: 153
Merit: 100
Is there any place where BAMT stores the logs of the miner ?

I can see in the web interface that some miners have been not working some hours and I would like to search for the reason.

hero member
Activity: 616
Merit: 506
BAMT Fix #15 - See what happens when you 'motivate' me? Smiley

#1 - Added a few more pool stat gathering routines.. current list is:

bitclockers btcguild deepbit slush bithasher mtred ozcoin nofee rfc polmine btcpool triple

To see your current balance, hashrate, estimated reward etc for any of the above in gpumon,
add your API key in bamt.conf, in the settings: section, in the general form of:

apikey_poolname: key

For instance:

apikey_bitclockers: 4Ft$SDFSGhSDFsfSDFSdfSDFSDF%$$

#2 - Added new overclocking settings for controlling the individual GPU profiles:

As some have noticed, the existing settings of mem_clock, core_clock, etc only effect profile 2.
This makes setting low memory speeds difficult, since you cannot set a lower value in profile 2
than is in profile 1.  The old settings will continue to work as they always have to avoid any
trouble with existing configs, but I've added settings for each profile as well.

The new settings take the form:

mem_clock_X: X

For instance, to set the memory clock to 300 in both profile 1 and 2 (handy):

mem_clock_1: 300
mem_clock_2: 300

mem_clock_X, core_clock_X, and core_voltage_X are available per GPU.

#3 - random bugfixes, rearrange some of the BAMT code to be more sensible


As always, use the built in fixer to grab this update: /opt/bamt/fixer
Should be online in a few minutes.

Your the best!  Another donation sent your way!  Any chance you could put up an updated bamt.conf file with all the latest tweaks?  I want to be sure I am using all the current stuff since 0.4b Smiley

https://github.com/aaronwolfe/Big-A-Miner-Thing/wiki/Example-bamt.conf

I'll try to keep that current
newbie
Activity: 47
Merit: 0
BAMT Fix #15 - See what happens when you 'motivate' me? Smiley

#1 - Added a few more pool stat gathering routines.. current list is:

bitclockers btcguild deepbit slush bithasher mtred ozcoin nofee rfc polmine btcpool triple

To see your current balance, hashrate, estimated reward etc for any of the above in gpumon,
add your API key in bamt.conf, in the settings: section, in the general form of:

apikey_poolname: key

For instance:

apikey_bitclockers: 4Ft$SDFSGhSDFsfSDFSdfSDFSDF%$$

#2 - Added new overclocking settings for controlling the individual GPU profiles:

As some have noticed, the existing settings of mem_clock, core_clock, etc only effect profile 2.
This makes setting low memory speeds difficult, since you cannot set a lower value in profile 2
than is in profile 1.  The old settings will continue to work as they always have to avoid any
trouble with existing configs, but I've added settings for each profile as well.

The new settings take the form:

mem_clock_X: X

For instance, to set the memory clock to 300 in both profile 1 and 2 (handy):

mem_clock_1: 300
mem_clock_2: 300

mem_clock_X, core_clock_X, and core_voltage_X are available per GPU.

#3 - random bugfixes, rearrange some of the BAMT code to be more sensible


As always, use the built in fixer to grab this update: /opt/bamt/fixer
Should be online in a few minutes.

Your the best!  Another donation sent your way!  Any chance you could put up an updated bamt.conf file with all the latest tweaks?  I want to be sure I am using all the current stuff since 0.4b Smiley
hero member
Activity: 616
Merit: 506
BAMT Fix #15 - See what happens when you 'motivate' me? Smiley

#1 - Added a few more pool stat gathering routines.. current list is:

bitclockers btcguild deepbit slush bithasher mtred ozcoin nofee rfc polmine btcpool triple

To see your current balance, hashrate, estimated reward etc for any of the above in gpumon,
add your API key in bamt.conf, in the settings: section, in the general form of:

apikey_poolname: key

For instance:

apikey_bitclockers: 4Ft$SDFSGhSDFsfSDFSdfSDFSDF%$$

#2 - Added new overclocking settings for controlling the individual GPU profiles:

As some have noticed, the existing settings of mem_clock, core_clock, etc only effect profile 2.
This makes setting low memory speeds difficult, since you cannot set a lower value in profile 2
than is in profile 1.  The old settings will continue to work as they always have to avoid any
trouble with existing configs, but I've added settings for each profile as well.

The new settings take the form:

mem_clock_X: X

For instance, to set the memory clock to 300 in both profile 1 and 2 (handy):

mem_clock_1: 300
mem_clock_2: 300

mem_clock_X, core_clock_X, and core_voltage_X are available per GPU.

#3 - random bugfixes, rearrange some of the BAMT code to be more sensible


As always, use the built in fixer to grab this update: /opt/bamt/fixer
Should be online in a few minutes.
hero member
Activity: 616
Merit: 506
Lodcrappo - hats off to you !!!

BAMT is the best thing that has ever happened to us miners.
I can't believe it's so simple to install and configure, yet so effective and reliable.

Just sent you 5 BTC.

Keep up the good work !

thank you sir, quite generous.  is there any feature or change you'd like to see first?  the man who puts the quarter in the jukebox gets to pick the song...

full member
Activity: 254
Merit: 100
Lodcrappo - hats off to you !!!

BAMT is the best thing that has ever happened to us miners.
I can't believe it's so simple to install and configure, yet so effective and reliable.

Just sent you 5 BTC.

Keep up the good work !
newbie
Activity: 47
Merit: 0
Also, there are some new per GPU config options:

  core_voltage:  
  pre_oc_cmd:
  post_oc_cmd:



Is there a way to string multiple commands for pre_oc_cmd?  For example:

Code:
 pre_oc_cmd: /etc/init.d/munin-node restart & atitweak -e 965 -m 845 -f 100 -A 0

Do we do two lines of pre_oc_cmd or one with something in between?

I have to use the the above atitweak as the default mem_speed fails to set the clock appropriately



the limitation is there on purpose, because atitweak makes it very easy to crash your machine by default.  i was getting tired of all the "bamt issues" that had nothing to do with bamt and were just people locking up their machines with insane overclocking.

to remove this safeguard, you can remove the "-P 2" from the line:

my $cmd = 'DISPLAY=:0.0 /usr/local/bin/atitweak -P 2 -A ' . $gpu;

in the doOC routine of /opt/bamt/common.pl

I will not respond to questions about why your machine starts locking up after you made this change Smiley  You can also edit how the entire OC routine works right there, it is very simple code and you won't hurt anything (much) by experimenting.  

Another alternative is to use the standard unix command seperator, the semi colon, in between multiple commands in the pre_oc_cmd setting, or to put lots of commands into a script and simply call that from the pre_oc_cmd.  I'm not sure what would happen if you specified multiple pre_oc_cmd values, but I doubt it would be something you'd like.  

Also, note that if you are doing something that has nothing to do with overclocking, and doesn't need to happen every time mining is (re)started, you probably should not be doing it here anyway.  /etc/rc.local is a good place to add small things that need to happen at bootup, or you could take a look at /etc/init.d/mine to hook into the mining startup process at any point that you need.


Thanks for all the help lodcrappo.  I'm going to leave things as is and just run the atitweak command manually for the time being. 
hero member
Activity: 616
Merit: 506
Also, there are some new per GPU config options:

  core_voltage:  
  pre_oc_cmd:
  post_oc_cmd:



Is there a way to string multiple commands for pre_oc_cmd?  For example:

Code:
 pre_oc_cmd: /etc/init.d/munin-node restart & atitweak -e 965 -m 845 -f 100 -A 0

Do we do two lines of pre_oc_cmd or one with something in between?

I have to use the the above atitweak as the default mem_speed fails to set the clock appropriately



the limitation is there on purpose, because atitweak makes it very easy to crash your machine by default.  i was getting tired of all the "bamt issues" that had nothing to do with bamt and were just people locking up their machines with insane overclocking.

to remove this safeguard, you can remove the "-P 2" from the line:

my $cmd = 'DISPLAY=:0.0 /usr/local/bin/atitweak -P 2 -A ' . $gpu;

in the doOC routine of /opt/bamt/common.pl

I will not respond to questions about why your machine starts locking up after you made this change Smiley  You can also edit how the entire OC routine works right there, it is very simple code and you won't hurt anything (much) by experimenting.  

Another alternative is to use the standard unix command seperator, the semi colon, in between multiple commands in the pre_oc_cmd setting, or to put lots of commands into a script and simply call that from the pre_oc_cmd.  I'm not sure what would happen if you specified multiple pre_oc_cmd values, but I doubt it would be something you'd like.  

Also, note that if you are doing something that has nothing to do with overclocking, and doesn't need to happen every time mining is (re)started, you probably should not be doing it here anyway.  /etc/rc.local is a good place to add small things that need to happen at bootup, or you could take a look at /etc/init.d/mine to hook into the mining startup process at any point that you need.
newbie
Activity: 47
Merit: 0
Also, there are some new per GPU config options:

  core_voltage: 
  pre_oc_cmd:
  post_oc_cmd:



Is there a way to string multiple commands for pre_oc_cmd?  For example:

Code:
 pre_oc_cmd: /etc/init.d/munin-node restart & atitweak -e 965 -m 845 -f 100 -A 0

Do we do two lines of pre_oc_cmd or one with something in between?

I have to use the the above atitweak as the default mem_speed fails to set the clock appropriately
full member
Activity: 153
Merit: 100
You are right. Although atitweak doesnt select the adapter based on DISPLAY, it need it set to anything to initialize some stuff.

hero member
Activity: 616
Merit: 506
This is what I see when starting mining:

Quote
GPU 0 OC CMD: DISPLAY=:0.0 /usr/local/bin/atitweak -P 2 -A 0 -f 82 -e 950 -m 300
results:
Setting performance level 2 on adapter 0: engine clock 950MHz, memory clock 300MHz
Setting fan speed to 82% on adapter 0


GPU 1 OC CMD: DISPLAY=:0.0 /usr/local/bin/atitweak -P 2 -A 1 -f 70 -e 950 -m 300
results:
Setting performance level 2 on adapter 1: engine clock 950MHz, memory clock 300MHz
Setting fan speed to 70% on adapter 1


GPU 2 OC CMD: DISPLAY=:0.0 /usr/local/bin/atitweak -P 2 -A 2 -f 70 -e 950 -m 300
results:
Setting performance level 2 on adapter 2: engine clock 950MHz, memory clock 300MHz
Setting fan speed to 70% on adapter 2


First, I want to say that it is working fine, not problem at all but may be you would want to clean the DISPLAY values that should be 0:0, 0:1, 0:2 or also remove them as they are not used by atitweak



#1 - it doesn't matter what sub display you use, so long as its a valid one
#2 - atitweak *does* require display to be set, at least for me

If you'd like to experiment, edit the doOC routine in /opt/bamt/common.pl, its quite straightforward. 
full member
Activity: 226
Merit: 100
Fix #14

Phoenix updated to v1.6.1 (with special BAMT modifications)

Recommend all miners try the new 'phatk2' kernel provided in this version.  Seems to provide a bit better hash rates than any of the kernels included previously on a wide range of cards.

Just edit bamt.conf and set:

  kernel: phatk2

in each GPU you want to try it on.



Nice! Your works is much appreciated! A few days of mining worth of BTC coming your way Smiley
full member
Activity: 153
Merit: 100
This is what I see when starting mining:

Quote
GPU 0 OC CMD: DISPLAY=:0.0 /usr/local/bin/atitweak -P 2 -A 0 -f 82 -e 950 -m 300
results:
Setting performance level 2 on adapter 0: engine clock 950MHz, memory clock 300MHz
Setting fan speed to 82% on adapter 0


GPU 1 OC CMD: DISPLAY=:0.0 /usr/local/bin/atitweak -P 2 -A 1 -f 70 -e 950 -m 300
results:
Setting performance level 2 on adapter 1: engine clock 950MHz, memory clock 300MHz
Setting fan speed to 70% on adapter 1


GPU 2 OC CMD: DISPLAY=:0.0 /usr/local/bin/atitweak -P 2 -A 2 -f 70 -e 950 -m 300
results:
Setting performance level 2 on adapter 2: engine clock 950MHz, memory clock 300MHz
Setting fan speed to 70% on adapter 2


First, I want to say that it is working fine, not problem at all but may be you would want to clean the DISPLAY values that should be 0:0, 0:1, 0:2 or also remove them as they are not used by atitweak

hero member
Activity: 616
Merit: 506
Awsome, I should be using mgpumon anyway with multiple rigs.  What catalyst version is BAMT using?

EDIT: With gpumon if I use do_direct_status: 1.2.3.4 should this broadcast to my local network from remote rigs? If so what ports will need to be forwarded?

it will send directly to the ip address you specify,  port udp 54545
donator
Activity: 798
Merit: 500
Awsome, I should be using mgpumon anyway with multiple rigs.  What catalyst version is BAMT using?

EDIT: With gpumon if I use do_direct_status: 1.2.3.4 should this broadcast to my local network from remote rigs? If so what ports will need to be forwarded?
hero member
Activity: 616
Merit: 506
Thanks, phatk2 +3 on my 6950's.  Would it be a lot of work to add clocks to gpumon?

you can see clocks (and lots of other stuff) using mgpumon.  it works fine even if you have a single machine, just set:

   do_bcast_status: 1

in the settings portion of your bamt.conf (or all your bamt.confs if you have multiple rigs), and run mgpumon on any of them.

Pages:
Jump to: