New F@H problem

[Markfw]

OK, I know you see this one in a while, and always, right after an unstable machine error. This video card is only one of 22 that was one a week getting an unstable machine @STOCK, but yesterday, right after finishing a work unit (with no problems) it got unable to get work unit errors. Then I created a new directory, and re-setup the card. Got one work unit, then went back to the problematic state. Then I created a third directory, to no avail. Its not been idle almost 24 hours, and this is one of my fastest cards, 9800 GTX+ @ 756 mhz stock !. So I am wasting 5000 ppd.

Also note, the other card on this same machine is working fine

WTF ???

Help please.


Note: Please read the license agreement (Folding@home-Win32-GPU.exe -license). F
urther
use of this software requires that you have read and accepted this agreement.



--- Opening Log file [April 7 13:52:00 UTC]


# Windows GPU Console Edition #################################################
###############################################################################

Folding@Home Client Version 6.20

http://folding.stanford.edu

###############################################################################
###############################################################################

Launch directory: C:\folding_gpu2
Executable: C:\folding_gpu2\Folding@home-Win32-GPU.exe
Arguments: -gpu 1 -forcegpu nvidia_g92 -verbosity 9

[13:52:00] - Ask before connecting: No
[13:52:00] - User name: Mark_F_Williams (Team 198)
[13:52:00] - User ID: 3F5744BF5275AE21
[13:52:00] - Machine ID: 6
[13:52:00]
[13:52:00] Loaded queue successfully.
[13:52:00] - Preparing to get new work unit...
[13:52:00] + Attempting to get work packet
[13:52:00] - Autosending finished units... [April 7 13:52:00 UTC]
[13:52:00] - Will indicate memory of 2046 MB
[13:52:00] - Detect CPU. Vendor: GenuineIntel, Family: 6, Model: 15, Stepping: 6

[13:52:00] Trying to send all finished work units
[13:52:00] - Connecting to assignment server
[13:52:00] Connecting to http://assign-GPU.stanford.edu:8080/
[13:52:00] + No unsent completed units remaining.
[13:52:00] - Autosend completed
[13:52:00] Posted data.
[13:52:00] Initial: 0000; + No appropriate work server was available; will try a
gain in a bit.
[13:52:00] + Couldn't get work instructions.
[13:52:00] - Attempt #1 to get work failed, and no other work to do.
Waiting before retry.
[13:52:07] + Attempting to get work packet
[13:52:07] - Will indicate memory of 2046 MB
[13:52:07] - Connecting to assignment server
[13:52:07] Connecting to http://assign-GPU.stanford.edu:8080/
[13:52:07] Posted data.
[13:52:07] Initial: 0000; + No appropriate work server was available; will try a
gain in a bit.
[13:52:07] + Couldn't get work instructions.
[13:52:07] - Attempt #2 to get work failed, and no other work to do.
Waiting before retry.
[13:52:27] + Attempting to get work packet
[13:52:27] - Will indicate memory of 2046 MB
[13:52:27] - Connecting to assignment server
[13:52:27] Connecting to http://assign-GPU.stanford.edu:8080/
[13:52:27] Posted data.
[13:52:27] Initial: 0000; + No appropriate work server was available; will try a
gain in a bit.
[13:52:27] + Couldn't get work instructions.
[13:52:27] - Attempt #3 to get work failed, and no other work to do.
Waiting before retry.
[13:52:53] + Attempting to get work packet
[13:52:53] - Will indicate memory of 2046 MB
[13:52:53] - Connecting to assignment server
[13:52:53] Connecting to http://assign-GPU.stanford.edu:8080/
[13:52:53] Posted data.
[13:52:53] Initial: 0000; + No appropriate work server was available; will try a
gain in a bit.
[13:52:53] + Couldn't get work instructions.
[13:52:53] - Attempt #4 to get work failed, and no other work to do.
Waiting before retry.
[13:53:43] + Attempting to get work packet
[13:53:43] - Will indicate memory of 2046 MB
[13:53:43] - Connecting to assignment server
[13:53:43] Connecting to http://assign-GPU.stanford.edu:8080/
[13:53:44] Posted data.
[13:53:44] Initial: 0000; + No appropriate work server was available; will try a
gain in a bit.
[13:53:44] + Couldn't get work instructions.
[13:53:44] - Attempt #5 to get work failed, and no other work to do.
Waiting before retry.
[13:55:14] + Attempting to get work packet
[13:55:14] - Will indicate memory of 2046 MB
[13:55:14] - Connecting to assignment server
[13:55:14] Connecting to http://assign-GPU.stanford.edu:8080/
[13:55:14] Posted data.
[13:55:14] Initial: 0000; + No appropriate work server was available; will try a
gain in a bit.
[13:55:14] + Couldn't get work instructions.
[13:55:14] - Attempt #6 to get work failed, and no other work to do.
Waiting before retry.

Also note, the other card on this same machine is working fine

 

[theAnimal]

I've had this happen twice in the past week. I stopped and restarted the client. If it didn't get a WU on the 1st attempt, I repeated until it did. I think it took less than a dozen tries to work.

 

[Markfw]

I have tried 20-30 times, to no avail. I mean quit all together. It tried 20 times overnight.

 

[Insidious]

I think you may need to completely uninstall (use add/remove programs, boot, delete remnants of ALL your folding folders and then set them up again. Apparently, just making a new directory left what ever was messed up behind.

-Sid

 

[Insidious]

The part that seems weird to me is that it says it is detecting your CPU but this is supposed to be a GPU client. I only see that line in my SMP clients ???

I wouldn't have noticed, but it says no appropriate server available. I don't think it knows this is a GPU client.

That client is SERIOUSLY confused. I suspect that accidentally folder contents have been jumbled up.

-Sid

 

[Markfw]

OK< I figured it out, but I don't believe it. This box has been running 24/7 for weeks, without touching it. But the problem was, that my shortcut had "-forcegpu nvidia_g92" and I changed it to 80, since I know thats invalid, and bam, working.

So how could it have been working before ?????

 

[Insidious]

The mind is the.... second thing to go! 😛

-Sid

(I've never used that switch.... do you really even need it?)

 

[Markfw]

Yes, its a must on >1 GPU in a system, at least in XP. I have no Vista systems.

And how could it been working for months, and now ???

 

[Insidious]

Originally posted by: Markfw900
Yes, its a must on >1 GPU in a system, at least in XP. I have no Vista systems.

And how could it been working for months, and now ???

Actually, that was my question. Both of my (WinXP) PCs are two GPU and two SMP clients. I haven't ever used that flag. So I not only wonder how it saved you, but I am wondering why my GPUs are running fine without it.

-Sid

 

[Markfw]

Probably the same reason mine was fine for so long ?

 

[Duvie]

Originally posted by: Markfw900
Probably the same reason mine was fine for so long ?

I can concur with mark here...When I fogot that line in my shortcut my 2 cards would not run correctly together. weird that Sid doesn't have it and he works fine....

Should be noted mine are 9800GTs...

 

[Insidious]

Originally posted by: Duvie

Originally posted by: Markfw900
Probably the same reason mine was fine for so long ?

I can concur with mark here...When I fogot that line in my shortcut my 2 cards would not run correctly together. weird that Sid doesn't have it and he works fine....

Should be noted mine are 9800GTs...

I have been wondering if it is something specific in hardware. Maybe different BIOS of GPU or Motherboard or type of card. ???

In each of my (WXP sp3) machines, I mix one ASUS 8800GTS (G92) with one ASUS 9800GTX+

I've been too lazy to see if the f@h forums could shed light. Besides, I haven't been in the mood to slap the trolls so much lately. :evil: (always a pleasant diversion when I'm in there actually trying to learn something..... NOT)

-Sid