Page 1 of 1

Project: 4754 (Run 8, Clone 171, Gen 12)

Posted: Tue Feb 02, 2010 8:45 pm
by emagiul
I have an ATI 5870 and I have problems with the new Wu's P4754. I deleted and restarted the client twice with the same error.

Code: Select all

20:26:28] Folding@Home GPU Core - Beta
[20:26:28] Version 1.24 (Mon Feb 9 11:00:12 PST 2009)
[20:26:28] 
[20:26:28] Compiler  : Microsoft (R) 32-bit C/C++ Optimizing Compiler Version 14.00.50727.762 for 80x86 
[20:26:28] Build host: amoeba
[20:26:28] Board Type: AMD
[20:26:28] Core      : 
[20:26:28] Preparing to commence simulation
[20:26:28] - Looking at optimizations...
[20:26:28] - Created dyn
[20:26:28] - Files status OK
[20:26:28] - Expanded 85663 -> 444252 (decompressed 518.6 percent)
[20:26:28] Called DecompressByteArray: compressed_data_size=85663 data_size=444252, decompressed_data_size=444252 diff=0
[20:26:28] - Digital signature verified
[20:26:28] 
[20:26:28] Project: 4754 (Run 8, Clone 171, Gen 12)
[20:26:28] 
[20:26:28] Assembly optimizations on if available.
[20:26:28] Entering M.D.
[20:26:35] Tpr hash work/wudata_06.tpr:  3875212657 3885216456 1697443020 59358383 4071811839
[20:26:35] Working on 1254 p4754_lam5w_300K_g91
[20:26:35] Client config found, loading data.
[20:26:35] Starting GUI Server
[20:26:37] mdrun_gpu returned 
[20:26:37] NANs detected on GPU
[20:26:37] 
[20:26:37] Folding@home Core Shutdown: UNSTABLE_MACHINE
[20:26:39] CoreStatus = 7A (122)
[20:26:39] Sending work to server
[20:26:39] Project: 4754 (Run 8, Clone 171, Gen 12)
[20:26:39] - Read packet limit of 540015616... Set to 524286976.
[20:26:39] - Error: Could not get length of results file work/wuresults_06.dat
[20:26:39] - Error: Could not read unit 06 file. Removing from queue.
[20:26:39] - Preparing to get new work unit...
[20:26:39] + Attempting to get work packet
[20:26:39] - Connecting to assignment server
[20:26:40] - Successful: assigned to (171.64.65.103).
[20:26:40] + News From Folding@Home: Welcome to Folding@Home
[20:26:40] Loaded queue successfully.
[20:26:41] - Attempt #1  to get work failed, and no other work to do.
Waiting before retry.
[20:26:49] + Attempting to get work packet
[20:26:49] - Connecting to assignment server
[20:26:50] - Successful: assigned to (171.64.65.103).
[20:26:50] + News From Folding@Home: Welcome to Folding@Home
[20:26:50] Loaded queue successfully.
[20:26:52] + Closed connections
[20:26:57] 
[20:26:57] + Processing work unit
[20:26:57] Core required: FahCore_11.exe
[20:26:57] Core found.
[20:26:57] Working on queue slot 07 [February 2 20:26:57 UTC]
[20:26:57] + Working ...
[20:26:58] 
[20:26:58] *------------------------------*
[20:26:58] Folding@Home GPU Core - Beta
[20:26:58] Version 1.24 (Mon Feb 9 11:00:12 PST 2009)
[20:26:58] 
[20:26:58] Compiler  : Microsoft (R) 32-bit C/C++ Optimizing Compiler Version 14.00.50727.762 for 80x86 
[20:26:58] Build host: amoeba
[20:26:58] Board Type: AMD
[20:26:58] Core      : 
[20:26:58] Preparing to commence simulation
[20:26:58] - Looking at optimizations...
[20:26:58] - Created dyn
[20:26:58] - Files status OK
[20:26:58] - Expanded 85621 -> 444252 (decompressed 518.8 percent)
[20:26:58] Called DecompressByteArray: compressed_data_size=85621 data_size=444252, decompressed_data_size=444252 diff=0
[20:26:58] - Digital signature verified
[20:26:58] 
[20:26:58] Project: 4754 (Run 1, Clone 258, Gen 11)
[20:26:58] 
[20:26:58] Assembly optimizations on if available.
[20:26:58] Entering M.D.
[20:27:04] Tpr hash work/wudata_07.tpr:  682606005 1486692238 187868442 2060881817 710036406
[20:27:04] Working on 1254 p4754_lam5w_300K_g91
[20:27:04] Client config found, loading data.
[20:27:04] Starting GUI Server
[20:27:05] mdrun_gpu returned 
[20:27:05] NANs detected on GPU
[20:27:05] 
[20:27:05] Folding@home Core Shutdown: UNSTABLE_MACHINE
[20:27:08] CoreStatus = 7A (122)
[20:27:08] Sending work to server
[20:27:08] Project: 4754 (Run 1, Clone 258, Gen 11)
[20:27:08] - Read packet limit of 540015616... Set to 524286976.
Any advice ?

Re: Project: 4754 (Run 8, Clone 171, Gen 12)

Posted: Wed Feb 03, 2010 3:52 am
by bruce
It is difficult to tell whether an error like that is due to a hardware problem or if it's what we call a "bad WU" In any case, it will be reissued and if someone else completes is successfully, then the your hardware has a problem (but you won't know if someone else completes it). If others have the same problem, it's a bad WU.

All you can really do is file this report and move on to another WU. If you get a series of DIFFERENT WUs with problems, then you should look for a hardware problem, but if it's only a single error in a series of good WUs, don't worry about it.