Mining rig randomly shuts off?

[clonedone]

Wow, way the go AMD. Way to keep up your reputation for having shitty drivers.

The problem affects a very slim percent of users. If you are running the card as a dedicated opencl accelerator, chances are you aren't the type of guy to hang out on Youtube all day. On a dedicated mining rig, no other app should be running that triggers the bug.
Besides, nVidia drivers aren't any better either. Remember the nforce chipset?

clonedone: run through the windows services, see if there's anything suspicious there (such as Windows Media Center). Also check if there are any scheduled tasks. I run a 6950 and any time it hangs, it's from an application I know I opened, never from something that opened in the background.

i just reformatted the whole system, installed the 11.4 cata, installed guiminer, and started it. it mined for 20 mins, then did the same wierd shutdown, where it dosent fully shut down, but nothing is running except the fans(as far as i can tell). then i restarted it and tried it again, it ran for an hour and shut off. now im trying it again, i never run any other programs or anything else, i updates windows though to service pack 1 for windows 7, but that was after i reformatted, doubt it has anything to do with it.

media center isnt running im sure of it. any other suggestion?

[TurboK]

Screensavers?

[clonedone]

Have you tried it with a single card? If your power supply is actually defective, perhaps it will be able to handle one card.

When you install the drivers, do you choose custom and only install the driver and APP? I have no reason to think the other crap will cause any instability, but I always deselect it anyway.

nope i selected express.

Screensavers?

not at all.

the thing about testing it out, is that sometiems it takes a couple of hours sometimes it takes minutes, its so confusing. i was thinking of setting it up without a moniter and see what happens

[Mugabuga]

i got a 1250 watt power supply, planning on running three 6990s on it, they will not be overclocked however.

i already changed the sleep settings. anyone know how to hook up the third one using windows 7?

You can't run three 6990s, max of 4 GPUs in CF.

[clonedone]

i got a 1250 watt power supply, planning on running three 6990s on it, they will not be overclocked however.

i already changed the sleep settings. anyone know how to hook up the third one using windows 7?

You can't run three 6990s, max of 4 GPUs in CF.

yeah im probably gonna sell it or return it. my third 6990 is useless right now.

but other than that, I am now running phoenix miner instead of guiminer. atfirst it shut down after 20  minutes twice. then i removed VECTORS and also removed WORKSPACE=128 and also removed BFI thingy and only left agression=10.
it has been running for two hours straight, other than the fact that deepbit crashed for 10 minutes, it still ran smoothly, but at 290 per gpu instead of 333.
I am hoping it will run nonstop finally, then i can adjust the script to see which one will let my computer last longer.
ill update tomorrow

[clonedone]

UPDATE
so heres whats going on. I ran it without vectors, workspace=128, an bfi_int. (pheonix miner)
it never shut down on me. so what i did afterwards was to run it with all of them again, and guess what it still didnt shut down on me for a full 24 hours.
so i decided that it must have been fixed. i decided to just restart the computer just to see what would happen if i shut the computer down, and it started messing up again, this time @ 20 minutes it would just die on me. so here I am again removing the three flags and its running fine again. we will see what happens next time.

hope this can help ppl with 6990

[TurboK]

I still think you have some specific application in the background that kicks in only periodically. Perhaps check your scheduled tasks, or leave task manager open and note down the running applications - since the screen freezes when the radeons hang, you can still read the taskmanager.

[clonedone]

I still think you have some specific application in the background that kicks in only periodically. Perhaps check your scheduled tasks, or leave task manager open and note down the running applications - since the screen freezes when the radeons hang, you can still read the taskmanager.

the screen loses connection. i checked sched tasks, there are none that i can find. if i am even looking correctly lol.
well ill try the task manager today, im really desperate to find out whats going on cus i wanna buy another rig.

anyway do u think ubuntu might react differently?'

my two 6990s are still running to this moment with the three flags off.(workspace vectors bfi_int)

[TurboK]

Ubuntu might behave differently since it is a wholly different OS architecture.

See if you have turned off all power saving options. I know that "turning off monitor after x mins" will not crash my 6950, but perhaps there's something else there about sleep state and somesuch. If the monitor loses connection as you are using the machine, then that may be unrelated to the UVD bug, since when that happens, the monitor picture stays frozen and even a hard reset takes a few seconds to be acknowledged.

[clonedone]

Ubuntu might behave differently since it is a wholly different OS architecture.

See if you have turned off all power saving options. I know that "turning off monitor after x mins" will not crash my 6950, but perhaps there's something else there about sleep state and somesuch. If the monitor loses connection as you are using the machine, then that may be unrelated to the UVD bug, since when that happens, the monitor picture stays frozen and even a hard reset takes a few seconds to be acknowledged.

yeah everything is on high performance, no sleep, no moniter sleep, no hibernate no nuttin.

[clonedone]

its very strange. now that i started mining without teh three flags (vectors, workspace, and bfiint)
then i add them afterwards
now it is NOT shutting down...
anyone know why?