BAMT - Easy persistent USB key based linux for dedicated miners/mining farms

[Transisto]

Question: Is there an easy was to figure out which GPU is locking up on a rig? I can still log into the box and can run the top command and can see that a phoenix thread is defunct. But how can I correlate that to a specific GPU?

I usually just run stop_mining and look for the card still stuck at 99% load with atitweak -s.  Doesn't gpumon and mgpumon show 0mhs with 99% load on a locked card though? I haven't locked one up in awhile.

The one that is hung usually have much lower number of shares.

But often time a bad gpu can hang the PC or all gpu at once, I think it depend on the motherboard.

[1714223223]

1714223223

[1714223223]

1714223223

[lodcrappo]

28" only resolution matters

I'd use a display that rotate,  1080x1920 is almost double.

the resolutions is 1920x1200. They are mounted on the wall but are not able to swivel with the position of the desk at the moment.

Question: Is there an easy was to figure out which GPU is locking up on a rig? I can still log into the box and can run the top command and can see that a phoenix thread is defunct. But how can I correlate that to a specific GPU?

Code:

for p in `ps --no-heading -o pid -C screen`; do pstree -apG $p; done

will show all processes associated with each GPU

[jamesg]

Code:

for p in `ps --no-heading -o pid -C screen`; do pstree -apG $p; done

will show all processes associated with each GPU

Thanks for this.

Has there been any discussion of using cgminer with bamt at all? Has anyone setup cgminer as an alternative miner on bamt? Seems like there have been some interesting features added into cgminer recently that would really help out.

[gnar1ta$]

Has there been any discussion of using cgminer with bamt at all? Has anyone setup cgminer as an alternative miner on bamt? Seems like there have been some interesting features added into cgminer recently that would really help out.

I have one rig running cgminer on a BAMT drive.  Works great, you can even use gpumon and mgpumon to see load, temp, fanspeed, and clocks - just not hash and pools of course.  Throw in a post_oc_cmd to a startup script and forget about it for weeks.  Though it really defeats the purpose of a simple, fast mining os.

[lodcrappo]

Has there been any discussion of using cgminer with bamt at all? Has anyone setup cgminer as an alternative miner on bamt? Seems like there have been some interesting features added into cgminer recently that would really help out.

I have one rig running cgminer on a BAMT drive.  Works great, you can even use gpumon and mgpumon to see load, temp, fanspeed, and clocks - just not hash and pools of course.  Throw in a post_oc_cmd to a startup script and forget about it for weeks.  Though it really defeats the purpose of a simple, fast mining os.

agree, cgminer is simply not in line with the "philosophy" of bamt.  you can of course use any linux miner on bamt but you will be giving up some functionality and the instant setup, so whats the point.  may as well use linuxcoin at that point, or your own install of linux.

the new proxy provides all of the features i find interesting from cgminer, and everything anyone else has said they like.  It allows zero configuration deployment of new rigs.  simply write an image, plug into machine, turn it on.  nothing else required.  this is the future of bamt, not cgminer.

[jamesg]

How can I add -q 2 command line param for when bamt starts up phoenix?

EDIT:

Here is a conversation I had with El at btcguild about phoenix...

[16:37] <gigavps> @Eleuthria still getting miner idle warnings, since you suggested it was somehow on my end i reset all network equipment
[16:37] <gigavps> [05/10/2011 20:34:26] Phoenix v1.6.2 starting... [05/10/2011 20:34:26] Setting auto kill signal for 180 seconds. [05/10/2011 20:34:26] Connected to server [05/10/2011 20:34:26] Currently on block: 148211 [05/10/2011 20:34:30] Result: 050dafcb accepted [05/10/2011 20:34:33] Result: c74016fc accepted [05/10/2011 20:34:34] Result: 099d91dd accepted [05/10/2011 20:34:45] Result: 6c6bc416 accepted [05/10/2011 20:34:46] Result: 2e039bbe accepted [05/10/2011 20:34:50] Result: 0a9de4ce accepted [05/10/2011 20:35:08
[16:37] <gigavps> [05/10/2011 20:35:51] Warning: work queue empty, miner is idle
[16:41] <@Eleuthria> I'd suggest a better miner if you're running that fast on a single client
[16:41] <gigavps> all gpus are 310-317 Mh/s
[16:42] <gigavps> i using phoenix on bamt
[16:42] <gigavps> are you suggesting that phoenix is not a good miner?
[16:42] <@Eleuthria> Yes.
[16:42] <@Eleuthria> Phoenix is terrible and inefficient, it tosses out a getwork after finding a single share instead of exhausting the space
[16:44] <@Eleuthria> cgminer is better in every way
[16:44] <gigavps> i am all for learning and improving, but why do i not have these problems on other pools like arsbitcoin.com, yourbtc.net, mainframe.nl?
[16:44] <@Eleuthria> But if you insist on using phoenix, add -q 2 or -q 3
[16:45] <@Eleuthria> Because we put you in the back of the line after just feeding you work.
[16:45] <@Eleuthria> So you acn't just spam us with getwork requests.
[16:45] <@Eleuthria> Which is whats happening in your log
[16:45] <@Eleuthria> You finished 3 shares in under 3 seconds
[16:45] <Dyaheon> you're the guy with the 20GH/s operation?
[16:46] <@Eleuthria> On phoenix, that means you're issuing a new getwork multiple times per second
[16:46] <@Eleuthria> Even though a single getwork can contain multiple shares
[16:46] <@Eleuthria> poclbm fixed that, cgminer never had that problem, and my understanding is diablo never had that issue either.
[16:47] <gigavps> do you mind if I post this conversation on the bamt forum topic?
[16:47] <@Eleuthria> bamt?
[16:48] <gigavps> https://bitcointalk.org/index.php?topic=28967.0
[16:48] <@Eleuthria> euh...one of those
[16:49] <gigavps> not sure what you mean
[16:50] <@Eleuthria> Never been a fan of these linuxcoin distros
[16:50] <gigavps> how do you decide when I go to the back of the queue? by worker? by ip address?
[16:50] <gigavps> i run 5 gpus per worker
[16:50] <@Eleuthria> Are you able to add command-line arguments to phoenix in that thing
[16:51] <@Eleuthria> -q 3 would fix it

[AniceInovation]

Interesting... i've been getting a few idle  warning on my bamt rigs too... i'll try that -q 3

[lodcrappo]

How can I add -q 2 command line param for when bamt starts up phoenix?

EDIT:

Here is a conversation I had with El at btcguild about phoenix...

[16:37] <gigavps> @Eleuthria still getting miner idle warnings, since you suggested it was somehow on my end i reset all network equipment
[16:37] <gigavps> [05/10/2011 20:34:26] Phoenix v1.6.2 starting... [05/10/2011 20:34:26] Setting auto kill signal for 180 seconds. [05/10/2011 20:34:26] Connected to server [05/10/2011 20:34:26] Currently on block: 148211 [05/10/2011 20:34:30] Result: 050dafcb accepted [05/10/2011 20:34:33] Result: c74016fc accepted [05/10/2011 20:34:34] Result: 099d91dd accepted [05/10/2011 20:34:45] Result: 6c6bc416 accepted [05/10/2011 20:34:46] Result: 2e039bbe accepted [05/10/2011 20:34:50] Result: 0a9de4ce accepted [05/10/2011 20:35:08
[16:37] <gigavps> [05/10/2011 20:35:51] Warning: work queue empty, miner is idle
[16:41] <@Eleuthria> I'd suggest a better miner if you're running that fast on a single client
[16:41] <gigavps> all gpus are 310-317 Mh/s
[16:42] <gigavps> i using phoenix on bamt
[16:42] <gigavps> are you suggesting that phoenix is not a good miner?
[16:42] <@Eleuthria> Yes.
[16:42] <@Eleuthria> Phoenix is terrible and inefficient, it tosses out a getwork after finding a single share instead of exhausting the space
[16:44] <@Eleuthria> cgminer is better in every way
[16:44] <gigavps> i am all for learning and improving, but why do i not have these problems on other pools like arsbitcoin.com, yourbtc.net, mainframe.nl?
[16:44] <@Eleuthria> But if you insist on using phoenix, add -q 2 or -q 3
[16:45] <@Eleuthria> Because we put you in the back of the line after just feeding you work.
[16:45] <@Eleuthria> So you acn't just spam us with getwork requests.
[16:45] <@Eleuthria> Which is whats happening in your log
[16:45] <@Eleuthria> You finished 3 shares in under 3 seconds
[16:45] <Dyaheon> you're the guy with the 20GH/s operation?
[16:46] <@Eleuthria> On phoenix, that means you're issuing a new getwork multiple times per second
[16:46] <@Eleuthria> Even though a single getwork can contain multiple shares
[16:46] <@Eleuthria> poclbm fixed that, cgminer never had that problem, and my understanding is diablo never had that issue either.
[16:47] <gigavps> do you mind if I post this conversation on the bamt forum topic?
[16:47] <@Eleuthria> bamt?
[16:48] <gigavps> https://bitcointalk.org/index.php?topic=28967.0
[16:48] <@Eleuthria> euh...one of those
[16:49] <gigavps> not sure what you mean
[16:50] <@Eleuthria> Never been a fan of these linuxcoin distros
[16:50] <gigavps> how do you decide when I go to the back of the queue? by worker? by ip address?
[16:50] <gigavps> i run 5 gpus per worker
[16:50] <@Eleuthria> Are you able to add command-line arguments to phoenix in that thing
[16:51] <@Eleuthria> -q 3 would fix it

edit line 72 and 76 of /opt/miners/phoenix/wrapper.pl

[jamesg]

Has there been any discussion of using cgminer with bamt at all? Has anyone setup cgminer as an alternative miner on bamt? Seems like there have been some interesting features added into cgminer recently that would really help out.

I have one rig running cgminer on a BAMT drive.  Works great, you can even use gpumon and mgpumon to see load, temp, fanspeed, and clocks - just not hash and pools of course.  Throw in a post_oc_cmd to a startup script and forget about it for weeks.  Though it really defeats the purpose of a simple, fast mining os.

agree, cgminer is simply not in line with the "philosophy" of bamt.  you can of course use any linux miner on bamt but you will be giving up some functionality and the instant setup, so whats the point.  may as well use linuxcoin at that point, or your own install of linux.

the new proxy provides all of the features i find interesting from cgminer, and everything anyone else has said they like.  It allows zero configuration deployment of new rigs.  simply write an image, plug into machine, turn it on.  nothing else required.  this is the future of bamt, not cgminer.

What about the new features in cgminer to set a range for gpu clock, fan and temp? Is this of no value or will it be built into bamt somehow?

[lodcrappo]

Has there been any discussion of using cgminer with bamt at all? Has anyone setup cgminer as an alternative miner on bamt? Seems like there have been some interesting features added into cgminer recently that would really help out.

I have one rig running cgminer on a BAMT drive.  Works great, you can even use gpumon and mgpumon to see load, temp, fanspeed, and clocks - just not hash and pools of course.  Throw in a post_oc_cmd to a startup script and forget about it for weeks.  Though it really defeats the purpose of a simple, fast mining os.

agree, cgminer is simply not in line with the "philosophy" of bamt.  you can of course use any linux miner on bamt but you will be giving up some functionality and the instant setup, so whats the point.  may as well use linuxcoin at that point, or your own install of linux.

the new proxy provides all of the features i find interesting from cgminer, and everything anyone else has said they like.  It allows zero configuration deployment of new rigs.  simply write an image, plug into machine, turn it on.  nothing else required.  this is the future of bamt, not cgminer.

What about the new features in cgminer to set a range for gpu clock, fan and temp? Is this of no value or will it be built into bamt somehow?

it's of no value to me, and given that software control of fans is unreliable, it won't be something you find in bamt.

[AniceInovation]

edit line 72 and 76 of /opt/miners/phoenix/wrapper.pl

can't we just add the -q 3 in the kernel_params on bamt.conf?
I opened that file but it's like chinese to me.

I have a problem. GPU0 starts and works some minutes, maybe even some hours, but then it gets idle for no apparent reason.
I did screen -r gpu0, and the error it gives when that happens is [screen is terminating]
Any idea what it could be?

[lodcrappo]

edit line 72 and 76 of /opt/miners/phoenix/wrapper.pl

can't we just add the -q 3 in the kernel_params on bamt.conf?

no

[jamesg]

lodcrappo - thanks for all of your assistance, BAMT is a great piece of software.

Since phoenix sound like it is going to stay an integral part of bamt, I have started chiming in over on the Phoenix thread regarding what needs to be corrected in phoenix.

thread here - > https://bitcointalk.org/index.php?topic=6458.0

It would seem that slush has identified the problems with phoenix and we are looking for others to help.

There is currently a 30 BTC bounty to get the errors slush has found corrected along with rollNTime and other small fixes. If anyone watching the BAMT thread would like to throw in on the bounty or can do the work necessary, please help out.

Thanks.

[lodcrappo]

lodcrappo - thanks for all of your assistance, BAMT is a great piece of software.

Since phoenix sound like it is going to stay an integral part of bamt, I have started chiming in over on the Phoenix thread regarding what needs to be corrected in phoenix.

thread here - > https://bitcointalk.org/index.php?topic=6458.0

It would seem that slush has identified the problems with phoenix and we are looking for others to help.

There is currently a 30 BTC bounty to get the errors slush has found corrected along with rollNTime and other small fixes. If anyone watching the BAMT thread would like to throw in on the bounty or can do the work necessary, please help out.

Thanks.

good idea.  i have no problems at all with phoenix as it is now, so I'm not sure what the issues are, but if corrections or improvements are made we will of course update the phoenix in bamt to include them.

[DeathAndTaxes]

I can't even get BAMT to boot.

Had 3x 5970 installed.  Dropped it down to single 5970. usb key, sempron, 2GB Ram.

System boots I get:

Code:

 "Kernal panic - not syncing: attempted to kill init!
PID: 1, comm: swapper Tainted: G    D 2.6.32-5-686 #1
Call Trace:
.... 15 lines of call trace

booting halts.

Any ideas?

[lodcrappo]

I can't even get BAMT to boot.

Had 3x 5970 installed.  Dropped it down to single 5970. usb key, sempron, 2GB Ram.

System boots I get:

Code:

 "Kernal panic - not syncing: attempted to kill init!
PID: 1, comm: swapper Tainted: G    D 2.6.32-5-686 #1
Call Trace:
.... 15 lines of call trace

booting halts.

Any ideas?

make sure you have the very latest firmware for your motherboard.  if that doesn't fix it, probably isn't going to work out on that hardware.  linuxcoin is a good alternative.

[DeathAndTaxes]

Thanks I will try that and then check out Linux coin.  Just liked the idea of BAMT remote management of multiple rigs.  Hopefully the firware works.

[m3ta]

lodcrappo - thanks for all of your assistance, BAMT is a great piece of software.

Since phoenix sound like it is going to stay an integral part of bamt, I have started chiming in over on the Phoenix thread regarding what needs to be corrected in phoenix.

thread here - > https://bitcointalk.org/index.php?topic=6458.0

It would seem that slush has identified the problems with phoenix and we are looking for others to help.

There is currently a 30 BTC bounty to get the errors slush has found corrected along with rollNTime and other small fixes. If anyone watching the BAMT thread would like to throw in on the bounty or can do the work necessary, please help out.

Thanks.

good idea.  i have no problems at all with phoenix as it is now, so I'm not sure what the issues are, but if corrections or improvements are made we will of course update the phoenix in bamt to include them.

That's the only possible and logical way, considering I have posted on the cgminer thread for several times about integration with bamt (cg does not write any PIDs, gpumon and the graphs are a no-no) and never got a reply there.

[lodcrappo]

lodcrappo - thanks for all of your assistance, BAMT is a great piece of software.

Since phoenix sound like it is going to stay an integral part of bamt, I have started chiming in over on the Phoenix thread regarding what needs to be corrected in phoenix.

thread here - > https://bitcointalk.org/index.php?topic=6458.0

It would seem that slush has identified the problems with phoenix and we are looking for others to help.

There is currently a 30 BTC bounty to get the errors slush has found corrected along with rollNTime and other small fixes. If anyone watching the BAMT thread would like to throw in on the bounty or can do the work necessary, please help out.

Thanks.

good idea.  i have no problems at all with phoenix as it is now, so I'm not sure what the issues are, but if corrections or improvements are made we will of course update the phoenix in bamt to include them.

That's the only possible and logical way, considering I have posted on the cgminer thread for several times about integration with bamt (cg does not write any PIDs, gpumon and the graphs are a no-no) and never got a reply there.

I've sent PMs to the cgminer guy also, and also no reply.

It seems cgminer wants to be a closed "all in one" system that cannot interoperate with other software.  Like most "all in one" solutions, you make some compromises when you choose it.  One of those is that integration with farm management systems like bamt is impossible.  

[jh1523]

For some reason, one of my 5850 cards drops off and stops mining at random intervals between a couple of hours and a couple of days. It's not due to overclocking since it does it even at stock speeds. Not due to heat, temps are around 53C while mining.

So here's my kludge to get around this and not have to check on the miner all the time:

crontab -e
add line:

Code:

@hourly      /etc/init.d/mine restart

Save. Done.
It restarts the mining process every hour. So far it works like a charm.