I keep getting the same errors the last few days from the GPU client and they have in common this project: 5765. I've tried deleting the work folder and I keep getting work units from the same project. Here's the log: Code: --- Opening Log file [January 1 13:06:08 UTC] # Windows GPU Console Edition ################################################# ############################################################################### Folding@Home Client Version 6.23 http://folding.stanford.edu ############################################################################### ############################################################################### Launch directory: D:\stuff\fah-gpu2 Executable: D:\stuff\fah-gpu2\fah.exe [13:06:08] - Ask before connecting: No [13:06:08] - User name: Chicken76 (Team 35947) [13:06:08] - User ID: 6A375AA7736A796A [13:06:08] - Machine ID: 2 [13:06:08] [13:06:08] Work directory not found. Creating... [13:06:08] Loaded queue successfully. [13:06:08] [13:06:08] + Processing work unit [13:06:08] Core required: FahCore_11.exe [13:06:08] Core found. [13:06:08] Working on queue slot 07 [January 1 13:06:08 UTC] [13:06:08] + Working ... [13:06:08] [13:06:08] *------------------------------* [13:06:08] Folding@Home GPU Core [13:06:08] Version 1.31 (Tue Sep 15 10:57:42 PDT 2009) [13:06:08] [13:06:08] Compiler : Microsoft (R) 32-bit C/C++ Optimizing Compiler Version 14.00.50727.762 for 80x86 [13:06:08] Build host: amoeba [13:06:08] Board Type: Nvidia [13:06:08] Core : [13:06:08] Preparing to commence simulation [13:06:08] - Looking at optimizations... [13:06:08] DeleteFrameFiles: successfully deleted file=work/wudata_07.ckp [13:06:08] - Created dyn [13:06:08] - Files status OK [13:06:08] Error: Missing work file=<> [13:06:08] [13:06:08] Folding@home Core Shutdown: MISSING_WORK_FILES [13:06:12] CoreStatus = 74 (116) [13:06:12] The core could not find the work files specified. Removing from queue [13:06:12] Deleting current work unit & continuing... [13:06:16] - Preparing to get new work unit... [13:06:16] + Attempting to get work packet [13:06:16] - Connecting to assignment server [13:06:18] - Successful: assigned to (171.67.108.11). [13:06:18] + News From Folding@Home: Welcome to Folding@Home [13:06:18] Loaded queue successfully. [13:06:19] + Could not connect to Work Server [13:06:19] - Attempt #1 to get work failed, and no other work to do. Waiting before retry. [13:06:38] + Attempting to get work packet [13:06:38] - Connecting to assignment server [13:06:39] - Successful: assigned to (171.67.108.11). [13:06:39] + News From Folding@Home: Welcome to Folding@Home [13:06:39] Loaded queue successfully. [13:06:41] + Closed connections [13:06:46] [13:06:46] + Processing work unit [13:06:46] Core required: FahCore_11.exe [13:06:46] Core found. [13:06:46] Working on queue slot 08 [January 1 13:06:46 UTC] [13:06:46] + Working ... [13:06:46] [13:06:46] *------------------------------* [13:06:46] Folding@Home GPU Core [13:06:46] Version 1.31 (Tue Sep 15 10:57:42 PDT 2009) [13:06:46] [13:06:46] Compiler : Microsoft (R) 32-bit C/C++ Optimizing Compiler Version 14.00.50727.762 for 80x86 [13:06:46] Build host: amoeba [13:06:46] Board Type: Nvidia [13:06:46] Core : [13:06:46] Preparing to commence simulation [13:06:46] - Looking at optimizations... [13:06:46] DeleteFrameFiles: successfully deleted file=work/wudata_08.ckp [13:06:46] - Created dyn [13:06:46] - Files status OK [13:06:46] - Expanded 46621 -> 252912 (decompressed 542.4 percent) [13:06:46] Called DecompressByteArray: compressed_data_size=46621 data_size=252912, decompressed_data_size=252912 diff=0 [13:06:46] - Digital signature verified [13:06:46] [13:06:46] Project: 5765 (Run 3, Clone 36, Gen 537) [13:06:46] [13:06:46] Assembly optimizations on if available. [13:06:46] Entering M.D. [13:06:52] Tpr hash work/wudata_08.tpr: 2605098839 2683697980 2823700340 680068446 1310104847 [13:06:52] [13:06:52] Calling fah_main args: 14 usage=100 [13:06:52] [13:06:52] Working on Protein [13:06:53] Run: exception thrown during GuardedRun [13:06:53] Run: exception thrown in GuardedRun -- Gromacs cannot continue further. [13:06:53] Going to send back what have done -- stepsTotalG=0 [13:06:53] Work fraction=0.0000 steps=0. [13:06:57] logfile size=0 infoLength=0 edr=0 trr=23 [13:06:57] + Opened results file [13:06:57] - Writing 635 bytes of core data to disk... [13:06:57] Done: 123 -> 124 (compressed to 100.8 percent) [13:06:57] ... Done. [13:06:57] DeleteFrameFiles: successfully deleted file=work/wudata_08.ckp [13:06:57] [13:06:57] Folding@home Core Shutdown: UNSTABLE_MACHINE Folding@Home Client Shutdown. I'm pretty sure my computer is OK. Nothing changed in hardware or software over the past few days. Video card can run games fine, temperatures are around 44 Celsius while writing this in Firefox with hardware acceleration and Aero on (disable them and they drop below 40), and while folding on it worked, they never got above 70 Celsius. So I don't think it cooked itself. Have you seen anything like this before?
It's a factory overclocked Asus TOP card. Never had any problems with temperatures. After more than an hour of playing World of Tanks just now, it reached a maximum of 68 Celsius.
No, it's a single GPU machine. For now! Will do that in about a week or so when I plan on opening my PC up and giving it a good clean. There could be some dust buildup on the card as there have been about 4 months since I last did it, and the temps used to be even better. Question: how do I force it to get another project, and not the same P5765 I get every time?
1. Stop the client. 2. Delete the relevant files; queue.dat, unitinfo.txt, FAHlog(s).txt and the work folder. 3. Go into the configuration and change the machine ID number. 4. Re-run the client, and hey presto ....
I've had this on my GTX 295 When I had the issue I remember reading that ANY overclock can fail, factory or user defined. Factory ones are not tested on folding 24/7 for a prolonged period. I wasn't 100% sure if mine was due to overclocking the card, OR some sort of driver issue as reinstalling my system resolved the issue. I had a batch fail for a while (Sorry, but it was 12+months ago now, and I didn't keep a log of which projects they were). I found that I kept getting problems so I lowered the overclock which got me through the problem WU's. After a while I spotted another issue with the card not folding on both cores - there had been a dearth of driver releases, which windows update had found and installed. I rolled back the driver to one that I had on another machine and bingo. I have since cranked the overclock back up to 666 on the core and not had a co-occurrence of the problem. No problems in 12+ months now. Either I've not had the same WU's in that time which caused the overclock to fail, or re-installing the drivers fixed the issue. As for being able to select which wu's you process, I'm afraid you only have limited control over the WU's. For CPU processing If you select the flag -BIGBETA you'll get those WU's, if you select -ADVMETHODS you'll get those, if you use -BIGADV then you'll get that range, but in the eventuality that there are none of those, then you'll get SMP units. There are only GPU WU's available for the GPU client afaik, so you cannot force it to select from a different "pool" type. Let us know how you get on....
Just letting you guys know that the problem seems to have disappeared once I erased everything but the F@H executable that you download. Recreated the client.cfg file and now it works fine, even the dreaded P5765. Thanks for the support!