Pages:
Author

Topic: BAMT version 0.5 - Easy USB based mining Linux with farm wide management tools - page 51. (Read 324176 times)

full member
Activity: 134
Merit: 100
Where did the mgpumon webpage go in .5c? The command line app works but the webpage doesn't exist.
hero member
Activity: 837
Merit: 1000
Can i set in bamt.conf for the rig to stop mining if temp reach 90C for one core ?
vip
Activity: 756
Merit: 503
Is it possible to offer an .iso version? I have a problem with my motherboard that I explain here: https://bitcointalksearch.org/topic/new-motherboard-wont-boot-from-usb-76294

I can install and boot any Linux image via UnetBootin. I can use LinuxCoin but I would prefer to use BAMT. The problem with DiskImager is that the MBR get erased and UnetBootin does not remove it.

I will send you my first 6 mined BTC if you can provide me with an iso version that install via Unetbootin.

Thank's!
full member
Activity: 134
Merit: 100
So I have a couple machines mining with bamt, but on machine will not report mining stats to gpumon. The api ports, address and passwords match. Where should I look to see why its not reporting? Also is there a way in bamt.conf to set intensity for cgminer?

(trimmed config files)

So this seems to have been resolved by a reboot. But why did it happen?
hero member
Activity: 616
Merit: 506
I have been testing BAMT for a while on my rigs and I am getting some odd behavior from some of them.

All the rigs are identical in that they all use the same motherboard, processor, power supplies, ram, flash drives and cards.

Problem #1
1 out of 9 of the rigs will not overclock any of the cards.  All 4 Sapphire 5830's are at 246Mhps.  There is nothing in the BAMT/live/control/active directory to clear.
Problem #2
1 out of the 9 rigs will not overclock any cards except for card 0. There is nothing in the BAMT/live/control/active directory to clear.
Problem #3
This has only happened twice and a quick reboot fixed it I just thought it was odd.  Two different rigs began mining with 1 less card than was shown in the upper right corner of the desktop.  Ie if 4 cards were detected by the OS it would only mine with 3 and didn't show the other in GPUmon etc.

Thanks for your help!

in order of probability..

you have a typo in one of your config files

one or more of your cards is flaky or faulty

the image is corrupt on one of your keys


suggest carefully checking config and then swapping cards between machines where o/c is normal and where it is not normal


member
Activity: 73
Merit: 10
how can i manualy switch between primary and backup pool in gpumon?
newbie
Activity: 40
Merit: 0
I have been testing BAMT for a while on my rigs and I am getting some odd behavior from some of them.

All the rigs are identical in that they all use the same motherboard, processor, power supplies, ram, flash drives and cards.

Problem #1
1 out of 9 of the rigs will not overclock any of the cards.  All 4 Sapphire 5830's are at 246Mhps.  There is nothing in the BAMT/live/control/active directory to clear.
Problem #2
1 out of the 9 rigs will not overclock any cards except for card 0. There is nothing in the BAMT/live/control/active directory to clear.
Problem #3
This has only happened twice and a quick reboot fixed it I just thought it was odd.  Two different rigs began mining with 1 less card than was shown in the upper right corner of the desktop.  Ie if 4 cards were detected by the OS it would only mine with 3 and didn't show the other in GPUmon etc.

Thanks for your help!
donator
Activity: 1218
Merit: 1079
Gerald Davis
No a 7990 @ 1 Ghz will do 2x what a 7970 @ 1 Ghz will do however I don't think a 7990 will do 1 Ghz (based on track record of 5970 and 6990).

Given stock clock on a 7970 is 925 Mhz and it achieves 1 GHz using 4 phase power it is unlikely that with the limits of 3 phase power and double GPU cooling the 7970 will be able to achieve a 1 Ghz clock (mining stable 24/7 for weeks).

I have no interest in betting.  We will find out soon enough.  To avoid this thread going off topic I will leave it at that.  The likely performance of 7990 can be debated in another thread (new or existing).

hero member
Activity: 497
Merit: 500
Crap... wish i would have researched better... 7990. 1.3Ghash per card... $849.99.... thats pretty sweet.  Of course, i wouldn't have had 6 7970's mining for a couple months, so I still came out alright by buying them earlier.... I'll wait for official specs/hashrate/power/temp of the 7990's, but I might very well upgrade as well here pretty soon.

Not sure why people think a 7990 will get 1.3 GH/s?

A 5970 doesn't get 2x the hashrate of a 5870   
A 6990 doesn't get 2x the hashrate of a 6970

An overclocked 5870 can push 450 MH/s.  I wish my 4x5970 rigs pushed 3.6 GH/s.  They push about 3GH/s or ~85% of what a 8x5870 would push.

The 7990 should get >1GH/s.  Maybe 1.1GH/s. I doubt it goes much higher on air.

Well I have 6 7970's all of witch get over 700MH/s so yes 2 x 700 = 1400 (1.4 GH/s)
 So you are saying that a 7990 will not get 1.3? Would you care to place a wager? I know that my 7970's at 1GHZ get about 630MH/s-650MH/s. So a stock clocked @ 1GHZ 7990 will not do 1.3 when overclocked? hmmmm... Sounds like a bet.
hero member
Activity: 616
Merit: 506
So you would suggest leaving fans set to whatever speed if autofan / variable is prone to failures ? 60% all the time OR automatic ATI setting ?

Honestly I don't know.  Most of the "data" we have is anecdotal.

Rapid changes in fan speed are likely damaging but GPU temps tend to rise slowly.  So with a temp-hysteresis of 2+ the fan shouldn't change speed that often or that rapidly.  Fans running "too slow" will likely cause unnecessary wear but who knows what too slow is.  One thing I do know is that on 5970 the ATI/AMD fan slope is too "weak".  AMD seems to be optimizing for noise not temps so I wouldn't recommend that.

Given all the variables of fan speed, ambient temp, fan quality, normal variance, air quality (dirt) I honestly don't know.

I set all my fans at 80% fixed (65% in winter) because they are in the garage and I can't hear the noise.  I use auto-gpu (throttle clocks not fan to keep temps below target) because until recently 5970s didn't play nice with "auto-fan".  Of 24 GPUs I have had 3 fan failures but they were all on used GPUs (and one was less 1 month after purchase) so who knows how they were abused before I got them.  I see no reason to change my setup since it works.

agreed, all we really have is anecdotal evidence.  can't hurt to share it though. patterns may emerge if enough experiences are known.

It's not just 5970s.. the built in fan slope on every GPU I've tested is far too low to be useful in mining.  However, you can easily edit this slope with a bios editor.  If someone wants auto fan speed, I think this would be the best approach.


donator
Activity: 1218
Merit: 1079
Gerald Davis
So you would suggest leaving fans set to whatever speed if autofan / variable is prone to failures ? 60% all the time OR automatic ATI setting ?

Honestly I don't know.  Most of the "data" we have is anecdotal.

Rapid changes in fan speed are likely damaging but GPU temps tend to rise slowly.  So with a temp-hysteresis of 2+ the fan shouldn't change speed that often or that rapidly.  Fans running "too slow" will likely cause unnecessary wear but who knows what too slow is.  One thing I do know is that on 5970 the ATI/AMD fan slope is too "weak".  AMD seems to be optimizing for noise not temps so I wouldn't recommend that.

Given all the variables of fan speed, ambient temp, fan quality, normal variance, air quality (dirt) I honestly don't know.

I set all my fans at 80% fixed (65% in winter) because they are in the garage and I can't hear the noise.  I use auto-gpu (throttle clocks not fan to keep temps below target) because until recently 5970s didn't play nice with "auto-fan".  Of 24 GPUs I have had 3 fan failures but they were all on used GPUs (and one was less 1 month after purchase) so who knows how they were abused before I got them.  I see no reason to change my setup since it works.
hero member
Activity: 616
Merit: 506
I don't quite see why running 100% would be better than autofan setting Huh

100% gets the bearings hotter -> oil evaporates at high temps -> bearing damaged -> fan fails / rattles

Autofan can only prolong life of fans I think ...

this is questionable theory.  fan turning faster means more airflow to cool bearings.  its possible that bearing temp is much higher at 30 or 50% than 100.
i don't think anyone has done enough actual research to make authoritative claims.  I can only offer the information found in my own use, which involved 30 some GPUs from mixed vendors and models over the course of nearly one year.  failure rate on fans using variable speed was nearly 50%.  no failures in the fans using fixed (high) speeds.  whether this was coincidental, or due to the slower speeds the variable fans used, or simply the fact that they ran at many different speeds, or some other factor, who knows.

I would like to see D&T's opinion on this.

So you would suggest leaving fans set to whatever speed if autofan / variable is prone to failures ? 60% all the time OR automatic ATI setting ?

Thanks !


I've had great luck simply setting the fans to 100% and not thinking about it again.  Whether this works as well for you, I cannot predict.

For me, using software to control the fan speed has always seemed risky.  When GPUs are on the edge of a lock up, they tend to ignore software control.  This means you lose the ability to increase the fan at exactly the worst time.    If you must have "auto fan" I'd highly recommend using a bios editor to set an appropriate temp/speed curve right on the card and letting it deal with the fan directly.




hero member
Activity: 518
Merit: 500
I don't quite see why running 100% would be better than autofan setting Huh

100% gets the bearings hotter -> oil evaporates at high temps -> bearing damaged -> fan fails / rattles

Autofan can only prolong life of fans I think ...

this is questionable theory.  fan turning faster means more airflow to cool bearings.  its possible that bearing temp is much higher at 30 or 50% than 100.
i don't think anyone has done enough actual research to make authoritative claims.  I can only offer the information found in my own use, which involved 30 some GPUs from mixed vendors and models over the course of nearly one year.  failure rate on fans using variable speed was nearly 50%.  no failures in the fans using fixed (high) speeds.  whether this was coincidental, or due to the slower speeds the variable fans used, or simply the fact that they ran at many different speeds, or some other factor, who knows.

I would like to see D&T's opinion on this.

So you would suggest leaving fans set to whatever speed if autofan / variable is prone to failures ? 60% all the time OR automatic ATI setting ?

Thanks !
hero member
Activity: 616
Merit: 506
I don't quite see why running 100% would be better than autofan setting Huh

100% gets the bearings hotter -> oil evaporates at high temps -> bearing damaged -> fan fails / rattles

Autofan can only prolong life of fans I think ...

this is questionable theory.  fan turning faster means more airflow to cool bearings.  its possible that bearing temp is much higher at 30 or 50% than 100.
i don't think anyone has done enough actual research to make authoritative claims.  I can only offer the information found in my own use, which involved 30 some GPUs from mixed vendors and models over the course of nearly one year.  failure rate on fans using variable speed was nearly 50%.  no failures in the fans using fixed (high) speeds.  whether this was coincidental, or due to the slower speeds the variable fans used, or simply the fact that they ran at many different speeds, or some other factor, who knows.
 
donator
Activity: 1218
Merit: 1079
Gerald Davis
Crap... wish i would have researched better... 7990. 1.3Ghash per card... $849.99.... thats pretty sweet.  Of course, i wouldn't have had 6 7970's mining for a couple months, so I still came out alright by buying them earlier.... I'll wait for official specs/hashrate/power/temp of the 7990's, but I might very well upgrade as well here pretty soon.

Not sure why people think a 7990 will get 1.3 GH/s?

A 5970 doesn't get 2x the hashrate of a 5870   
A 6990 doesn't get 2x the hashrate of a 6970

An overclocked 5870 can push 450 MH/s.  I wish my 4x5970 rigs pushed 3.6 GH/s.  They push about 3GH/s or ~85% of what a 8x5870 would push.

The 7990 should get >1GH/s.  Maybe 1.1GH/s. I doubt it goes much higher on air.
vip
Activity: 756
Merit: 503
I don't quite see why running 100% would be better than autofan setting Huh

100% gets the bearings hotter -> oil evaporates at high temps -> bearing damaged -> fan fails / rattles

Autofan can only prolong life of fans I think ...
Maybe it's like a car. It is harder for mechanic if you live in a city when driving at irregular speed compared to rural area where you spend more time on the highway.
hero member
Activity: 497
Merit: 500

No don't wait a week for the 7990's to come out.. much more worth the money. 1.3Ghash per card cost 849.99

Thats why I am selling of a lot of my 5 series and 7970's

I assume you mean "wait a week" instead of "don't wait a week"   Wink

Crap... wish i would have researched better... 7990. 1.3Ghash per card... $849.99.... thats pretty sweet.  Of course, i wouldn't have had 6 7970's mining for a couple months, so I still came out alright by buying them earlier.... I'll wait for official specs/hashrate/power/temp of the 7990's, but I might very well upgrade as well here pretty soon.

Yea meant wait a week. I have 6 7970's also funny. I will be first in line for the 7990s!
hero member
Activity: 518
Merit: 500
I don't quite see why running 100% would be better than autofan setting Huh

100% gets the bearings hotter -> oil evaporates at high temps -> bearing damaged -> fan fails / rattles

Autofan can only prolong life of fans I think ...
sr. member
Activity: 309
Merit: 250

No don't wait a week for the 7990's to come out.. much more worth the money. 1.3Ghash per card cost 849.99

Thats why I am selling of a lot of my 5 series and 7970's

I assume you mean "wait a week" instead of "don't wait a week"   Wink

Crap... wish i would have researched better... 7990. 1.3Ghash per card... $849.99.... thats pretty sweet.  Of course, i wouldn't have had 6 7970's mining for a couple months, so I still came out alright by buying them earlier.... I'll wait for official specs/hashrate/power/temp of the 7990's, but I might very well upgrade as well here pretty soon.
full member
Activity: 134
Merit: 100
So I have a couple machines mining with bamt, but on machine will not report mining stats to gpumon. The api ports, address and passwords match. Where should I look to see why its not reporting? Also is there a way in bamt.conf to set intensity for cgminer?

bamt.conf
Code:
settings:
  miner_id: oldhda
  miner_loc: ...
  do_monitor: 1
  do_bcast_status: 1

  smtp_host: ...
  smtp_to: ...
  smtp_auth_user: ...
  smtp_auth_pass: ....
  smtp_tls: 1
  smtp_ssl: 1
  do_mgpumon: 0
  do_autoconf_client: 1
  detect_defunct: 1

  phoenix2: 1
  phoenix2_config: /etc/bamt/phoenix2.conf
  phoenix2_port: 7789
  phoenix2_pass: bamt


gpu0:
  disabled: 0
  phoenix2: 1

  # core_speed_0: 300
  # core_speed_1: 800
  core_speed_2: 900

  mem_speed_0: 175
  mem_speed_1: 175
  mem_speed_2: 175

  fan_speed: 60

  kernel: phatk2
  kernel_params: BFI_INT VECTORS FASTLOOP=false AGGRESSION=9

  pool_file: /etc/bamt/pools

  pool_timeout: 180
  monitor_temp_lo: 45
  monitor_temp_hi: 80
  monitor_load_lo: 80
  monitor_hash_lo: 350
  monitor_fan_lo: 2000
  monitor_reject_hi: 2


phoenix.conf
Code:
[general]
    autodetect = +cl -cpu #Use autodetect for all OpenCL devices, except those which are CPUs.
    verbose = True #Enable verbose logging?
    backend = http://[email protected]:8332 #The primary backend.
    # backups = http://user2:[email protected]:8332 http://bitcoin:bitcoin@localhost:8332 #A space seperated list of backup servers.
    failback = 600 #Seconds between attempts to reconnect to primary backend when using backups. (0 to disable)
    queuesize = 3 #Target/maximum size of the queue
    queuedelay = 5 #Seconds before work expires to request more work (WARNING: don't change this unless you know what you are doing!)
    statusinterval = 1 #Seconds between statusbar updates
    ratesamples = 10 #Number of samples to average for hashrate reporting
    # logfile =  #Set this option to log to a file.
[web]
    disabled = False #Disable the RPC server?
    bind = 127.0.0.1 #IP to bind the RPC server to
    port = 7789 #RPC port MUST MATCH SETTING IN BAMT.CONF
    password = bamt #RPC password  MUST MATCH SETTING IN BAMT.CONF
    root = /var/www/phoenix #Root directory for the web server
    logbuffer = 1000 #How many logs to remember in the getlogs() RPC call

# Hey, you.  yeah, you. You must fix the stuff below to match your GPUs!

[cl:0:0]
    autoconfigure = False #Automatically configure this device?
    kernel = phatk2 #The kernel to use for the device
    name = GPU 0 #The name to display the device as (default is device ID if not specified)
    start_undetected = False #Do not start the kernel if the device is not present
    disabled = False #Disable this device?
    worksize = 128 #Work group size, tweaking this option may improve performance
    vectors = True #Use unit2 vectors? (enable this or vectors4, not both)
    vectors4 = False #Use uint4 vectors? (enable this or vectors, not both)
    bfi_int = True #Use BFI_INT instruction on ATI VLIW GPUs?
    goffset = True #Use OpenCL 1.1 global offset?
    fastloop = False #Use fast internal loop? (ideal for low aggression)
    aggression = 9 #Number of nonces to test per kernel execution (lower value = less desktop lag, higher value = higher hashrate)
Pages:
Jump to: