Topic: BAMT version 0.5 - Easy USB based mining Linux with farm wide management tools - page 38. (Read 324176 times)

So does the screen stay black or does it go red after a while? If it stays black, do any graphs pop up after a while?

Can you open a root console and type 'gpumon'? what happens then?

Yes,all gpus show in the top right,but just can't get it work

Does it ever get to the GUI desktop? Does it show all the GPUs in the top right?

I tried to run 3 cards ,got the same problem,
keep showing "starting mining processes" and suddenly exit again
 

try ubuntu x64!

No, the only official supported BAMT version 0.5 is x86.

any other version of BAMT has 64bit?

You should ask someone for a 64-bit version of bamt if you want 8 GPUs (lordcrappo didn't make it public but maybe you could ask him to seed it for a day or two)

When building a new rig with multiple GPUs I have found that it's sometimes easier to troubleshoot by starting with one GPU first to verify that everything is running, then start adding in additional. I am thinking that you may have problem with four 5970 being seen as a total of eight GPUs. Maybe just pull one card out and see if it runs then. You should see six GPUs.

I got problem when I runing 4 5970,
I have set the mining profile(bamt.conf and pools),and run all the fixer,restart mining sessions,it shows
"stoping mining processes:mine
starting mining processes:mine, starting mining processes:mine, starting mining processes:mine,

and suddenly exit

I tried to reinstall the os ,still got the same problem


2 BTC bounty!!!!!!!!!!!!!!!!

Please note that lodcrappo is on a road trip.  Not sure how often/frequently he'll be checking in on forum in the meantime.

As has been stated several times I believe in this thread, .5 = .4 with all the fixes already installed.  There's nothing else different. 

Is there a command I can use to push my farm or individual rig to the next pool in my config file via SSH? If not, can this be added in a fix? I would gladly donate again for this to be done.

Just wanted to drop in a "BAMT is good stuff..."
Got my rig finished building today, booted up, running right off the bat. Set config and pool files and bam... mining away no problems (plug and mine as someone else said).

Just have my 6850 in the rig so far, is is set for 1000/300 and is stable at 65 degrees, fan set to 65% (can't be heard in our bedroom at that level). Pulling about 273 Mhash, a 5% improvement over when I had that card in my desktop and about a 15 degree difference to the better. Tomorrow I'll drop in the 5870 to go along with it.

A couple of notes, if you try to set the temperature in the configuration file and use a preceding 0 (as in 070 instead of 70) on the fan speed, BAMT sets the fan speed to the last two digits minus 14. So I had set fan speed for 70 to start and was only getting 56. Just thought I'd mention that odd behavior. Also the following motherboard should probably go on the compatible list since there were no problems whatsoever booting from USB and getting running, didn't even have to connect a monitor and enter bios. I did the entire set-up for the BAMT

system is headless, caseless and consists of the following:

ASRock 970 EXTREME3 AM3+ AMD 970 SATA 6Gb/s USB 3.0 ATX AMD Motherboard
http://www.newegg.com/Product/Product.aspx?Item=N82E16813157280&nm_mc=EMC-IGNEFL050112&cm_mmc=EMC-IGNEFL050112-_-EMC-050112-Index-_-AMDMotherboards-_-13157280-L0J

AMD Athlon II X2 250 Regor 3.0GHz Socket AM3 65W Dual-Core Desktop Processor ADX250OCK23GM http://www.newegg.com/Product/Product.aspx?Item=N82E16819103904&Tpk=AMD%20Athlon%20II%20X2%20250%20Regor%203.0GHz%20Socket%20AM3%2065W

Kingston ValueRAM 1GB 240-Pin DDR3 SDRAM DDR3 1333 (PC3 10600) Desktop Memory Model KVR1333D3N9/1G (2 of these)
http://www.newegg.com/Product/Product.aspx?Item=N82E16820134657

hec Zephyr MX 750 750W Peak Output ATX12V V2.3 / EPS12V V2.91 SLI Ready CrossFire Ready 80 PLUS Certified Active PFC Power Supply
http://www.newegg.com/Product/Product.aspx?Item=N82E16817339002

COOLER MASTER Hyper 212 Plus RR-B10-212P-G1 "Heatpipe Direct Contact" Long Life Sleeve 120mm CPU Cooler Compatible Intel Core i5 & Intel Core i7
http://www.newegg.com/Product/Product.aspx?Item=N82E16835103065


Thanks...
Thralen

I dont know much about using SSH and so I shy away from it.  I use the GUI remotely by installing xrdp, which of course is still no good for fixing network problems... 

FYI, the board I had the problem is ASUS M4A79XTD EVO.  The LAN chipset is Realtek 8112L.  I managed to fix it (I believe) by doing the fix mentioned and setting the ip manually by clicking the icon.  Of course I had to lug the monitor to the rig to do it.   

Hmm.  From experience, if you have a system running bamt and the monitor is connected and then you disconnect it, you have a pretty good chance that it will stop reporting and spit out the magic cookie/Xauthority errors.  I find that doing a fresh bamt w/no monitor connected and then SSHing into the box is the absolute best way.  Thus anything in the GUI is not accessible.  


We got two 7970s mining with bamt on one machine, been running stable for weeks now.  We were able to get CPU usage to be 60-80%, and it also reports to the other machines running mgpumon using the local API...  BUT mining has to be started manually from the /opt/miners/cgminer folder.

Hi.

I finally managed to get my 7970 working on BAMT with cgminer.
Upgraded to SDK 2.6 and installed the 8.921 ATI driver (wich is the only one that doesn't cause HW errors with cgminer at least)
Installed also the last cgminer (2.4.1 now) and I get around 715 Mh/s.

All good except the damn high CPU usage bug, so I have to use GPU_USE_SYNC_OBJECTS=1 to get rid of it. The thing is I don't know (yes, noob) how to correctly integrate this in BAMT scripts.
I wandered desperately through those scripts and finally added in /opt/bamt/common.pl after line 955 (cgminer section):
$ENV{GPU_USE_SYNC_OBJECTS} = "1";

Seems to be working, the CPU usage is back to normal but I have no idea if this is the proper way to to do it.
Can someone point me to the correct way?

Thanks

Of course, it's not lost it's DHCP address yet.  Typical... I wonder if adding the second NIC has something to do with it?

Hey A, I seem to be having this this same issue with my change over to 0.5. All of my MB's are MSI 870-G45's. I changed over 2 of these machines today and they both dropped off the network about the same time - within 2 hours of booting up.

I'm sorry I don't have any more specific information. I'm not on location with the 2 and they're not accessible via network.

I'll be able to get more info in the morning.

Cheers

I don't suppose you have a fully patched/latest image of 0.4 laying around I can download, do you? While I resolve this 0.5 networking issue...?

Or how about an archive of all of the patches that the 0.4 fixer applies - that I may point the fixer to on a network or thumb drive.?

No, drop it in /opt/miners.  It extracts to /opt/miners/cgminer  .  There is a script in there called "update.sh" - run that and it will compile the new version - but I just realized I forgot to add "git pull" as the first line of the script, doh.  I usually do that manually so it wasn't in the script.

There's nothing additional that needs to be installed and it "just works" once it's done compiling.  Creates a binary /opt/miners/cgminer/cgminer ... has bitforce support as well.