Topic: BAMT - Easy persistent USB key based linux for dedicated miners/mining farms - page 47. (Read 167468 times)

I've liked BAMT for mining but (as stated above) my motherboard wont boot from most linux bootable usb so I want to install it in the HD. I have tried googling for Norton Ghost but there are a lot of versions and I dont really know what I should get. Can you give me a bit more detailed explanation, specially on the software to get? Ideally BAMT would live as another OS bootable from GRUB, but if I have to wipe out the HD and install only BAMT so be it. Thanks.

I can not get BAMT to work, I can not even boot into the desktop.

I have downloaded BAMT twice (in case there was a problem with the download), I have tested it in two computers and from two USB devices. Same result. I get a failure about not being able to mount the file system and it drops me to a terminal.

I have installed it with imagewriter from Ubuntu.

EDIT: Ok, I have no idea why, but made a last attemp on one of the computers, with the same usb that refused to load earlier and now its working. Looks very good, now I need to find how to change the voltage.

EDIT2: It turns out the motherboard of the miners has a bug in the BIOS and does not boot from certain types of USB systems. No BAMT for me.

Here is closure on my problem (2 mining rigs stopping mining)

BAMT is innocent - it works perfectly well. It was my router after all. Not the router attached to the rigs, which I use as a repeater bridge, but my main router. It has an older version of ddwrt on it which is not too compatible with the latest one that's on the remote router. And as such it was occasionally dropping the connection to the repeater...

Upgraded now, and keeping my fingers crossed.

strange stuff.  I would make sure that everything is OK from a hardware perspective.. you wouldn't believe the number of times i forgotten to plug in a power connector or whatever.   assuming thats all fine, check the output of:

aticonfig --list-adapters

everybody showing up?

how about:

atitweak -l

also all good?

if so, try forcing a rebuild of the xorg.conf:

rm /etc/X11/xorg.conf*
coldreboot


see where that gets you

Same cards
OMhs w/o BFI_INT
Worked fine until I added another card (which I removed)

Is it a different type of card from the rest?

Have you tried it without the BFI_INT?

Has this card worked at all with BAMT?

Been having problems with one of my rigs. Getting this error:


The other cards will start fine.  Did this card just go bad??

FYI, the problem with that is that gputempx plugins that rely on gputmp and this last on aticonfig, run under user munin.munin and this doesnt like very much to the X server. So, as an ugly workaround I have disabled X control access and it seems to make the trick:



May be replacing in your gputmp script the call to aticonfig to atitweak is better, but I dont want to break nothing ;-)

Something like this:

atitweak -s | grep temperature | awk '{printf "temp.value %s\n",$2}'

Just on a random note to lodcrappo...

Once I finally got this working (about a week ago) it's been running perfectly ever since.  Loving this OS.

Honestly it sounds like once your room heats up, one of the card flakes out and crashes the rig. If you run your overclock very close to the limit, it can actually be too close once things heat up.

I had a card that worked great at 910 MHz, at night, when it was fairly cool. Around 2-4pm when it was hot outside, the room got hotter, it would lock up and bring down the whole bamt rig. I put it to 905 and it never happened again.

12 hours sounds about the timeframe it happened to me also.

I'd also try a pool file for each gpu, just 1 pool url. Edit out all the backup pool stuff. If it works fine for 12 hours and then dies, it's likely your cards/psu overheating and choking. From what I know BAMT doesn't change anything from hour 1 to hour 20. The backup pool stuff might be something I've never used it so I'd edit that out for now. Once you get it running for 2-3-4+ days then add it back and see if it returns.

Anytime you have issues with bamt, reduce all your engine clocks, and try again. 99% of the crashes due to overclocking lock up bamt and require a hard reboot. Not just a hard reboot, I have to unplug the rig for 15 seconds then fire it back up. Sometimes it's just 1 card. I have a card that if it hits 82C it would lock up. I don't know if its phoenix, X, ati drivers, but when a card locks up it brings everything down with it.

If you say you can't get SSH access that would say to me that it's a hardware failure. Or your loosing the internet connection. Have you checked logs etc ?

I understand, and I'm not blaming BAMT. The way both come down at once is suspect. Nothing else in the household seems affected by any kind of power outages, and I do have equipment that would show it. I will try another circuit but it won't be easy since the next circuit is about 40ft further away.

(edit) hmmm, I just found one of my machines mining on its backup pool for the first time ever. The other one was still on the main pool. i wonder if that has any relation to what is happening? By inspecting my backup pool, I noticed there is more stuff there than should be, even if my second rig had been mining there for the past 2-3h. Too bad I can't get more detailed logs of which client of mine mined where and when. But that still wouldn't explain why both rigs become totally unresponsive after a while... would it?

That sucks, but I don't have any idea how to help.  BAMT has no 12 hour death mode that I know of, and to have 2 machines going down together rules out nearly anything I can think of.   Never heard of anything like it at least.

"Bad" power can cause some really weird stuff..  low voltage or spikes, etc can cause lock ups without making them turn off.  can you try plugging them into completely different circuits?  that wouldn't help if its the whole house though... dunno man.

Damn it happened again. Itls like clockwork every 12h or so my boxes drop off the face of the earth. And because today my pool went through several quick blocks I lost about .3 BTC.

Can't say for sure they go down at the exact same time, but anyway close together - for example yesterday they were both on at the beginning of a block, and down at the end of the next block which was 20 minutes later.
Both machines are in an air-conditioned basement with no vibrations whatsoever. No PSU sharing, each has its own. They're not on an UPS however, but both are set to not wake up after power failure - and they were on when I went to reboot them. That kinda rules out power problems, otherwise I'd have found them turned off.

Today so far so good, coming on the 7h since the last reboot was needed... We'll see how the day goes.

Nope, absolutely nothing. Just checking the status pages remotely from my phone.

thats a new one.  both machines become unresponsive at the same time?  I wonder if your power is involved somehow..  would have to be some common factor so not likely temperature or even GPU related, you're probably fine there.   especially with nothing in the logs, I would be looking to any common external factor... though there really aren't many besides power i guess... strong vibration shaking the usb keys?  weird stuff man..  

one thought-- sharing any psu between both rigs?

You say since the difficulty increase... but did you do anything to the configuration at all?  Even running the auto updater for BAMT... anything at all you did while in the machine.

I have a problem. Don't know if it's BAMT related, but I have moved both my mining machines to BAMT since last week, and I have the same issue on both of them.

Since yesterday, actually just after the difficulty increase, every 10-12h all my miners just stop mining. I couldn't find an explanation why this happens, but the pool doesn't register shares anymore, the http status pages aren't accessible anymore, and I can't get ssh access to the boxes either. I can't get them back without a hard reset (which is difficult to do because of location). I looked through the system logs and couldn't find anything to explain what happens. It has happened 3 times so far, and when it happens both machines come down at once. It's not the router, I checked. The video cards in both are lightly overclocked but temperatures never go over 68-70 degrees on either. This exact same configuration has worked really nicely for over 1 week until this started happening on Monday.

LODCRAPPO,

My pleasure thank you for providing us with such a great OS! It still blows me away how great this is!
That way of making it run on a Hard Drive is so much less stress than the way LinuxCoin demo says, so much simpler to just create a USB drive, Image that sector by sector and 10 Minutes later you have a ready to go Hard Disk or SSD Drive version

Thanks