Project: 6806 (Run 3987, Clone 2, Gen 10) - Sudden errors.

[HendricksSA]

ChrisM101, the usual advice on how to get rid of a problem work unit should apply. Stop the client if it isn't processing anything. Then delete the work directory and the queue.dat file. Restart client and see if you get new work. If not, change your machine id and repeat previous step and you should get a new work unit. Good luck.

[bruce]

Please take this work unit off the server until its fixed. It has totally ruined the Folding on the GPU it was running on.
6806 (Run 3987, Clone 2, Gen 10) FAILS everytime

That may be true, but it may also be because of something Tracker is doing. Note that the first log says
[05:33:18] Gpu type=3 species=30.
and your log says
[18:29:11] Gpu species not recognized. several times. It looks like your client is not configured correctly.

Tracker is a 3rd party application and you may have to get support from the developer.

[HendricksSA]

ChrisM101, I'm not sure how you replace a client using Tracker v2, but your client is an older version, 6.30r1. The latest client is 6.41r2. Please note that the console version download is the bare essentials executable and you will need the supporting cuda files. Some folks who like the console version install the system tray version to get those supporting files and just copy them into their console directory. I'm not sure how this will impact your multi-GPU install. What nVidia cards are you running in this computer?

[ChrisM101]

SLI GTX 480s win 7 64
Ive totally deleted and reinstalled and whats the first unit i get. 6806 ... FAIL
I guess I will just suspend all GPU work until I can get a concrete answer as to whether the UNIT is Bad or if i need to further update. GPU Tracker has update features and Im not sure that the log i posted was after i updated or after a reinstall. I will check it if i go home for lunch to see if its 6.41 now.

But Really do you have any results on this unit? Surely by now someone has verified the WU? it takes what 1.5hrs on a GTx480-580?

Like i said Ive folded over a million points just this month, the only problem is this WU.

[HendricksSA]

You got the 6806 in question? You are incredibly lucky ... time to buy a lottery ticket! Usually the machine id change is a guaranteed success. The GPU not recognized is a real problem as Bruce said. Looking forward to your next log post.

[ChrisM101]

Code: Select all

--- Opening Log file [March 24 18:02:42 UTC] 


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

                       Folding@Home Client Version 6.30r1

                          http://folding.stanford.edu

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

Launch directory: D:\Downloads\FAH_GPU_Tracker_V2\FAH GPU Tracker V2\GPU0
Executable: D:\Downloads\FAH_GPU_Tracker_V2\FAH GPU Tracker V2\FAH_GPU3.exe
Arguments: -oneunit -forcegpu nvidia_fermi -verbosity 9 -gpu 0 

[18:02:42] - Ask before connecting: No
[18:02:42] - User name: ChrisM101 (Team 111065)
[18:02:42] - User ID: 21BE43D7336DE836
[18:02:42] - Machine ID: 3
[18:02:42] 
[18:02:42] Gpu species not recognized.
[18:02:42] Work directory not found. Creating...
[18:02:42] Could not open work queue, generating new queue...
[18:02:42] - Preparing to get new work unit...
[18:02:42] Cleaning up work directory
[18:02:42] - Autosending finished units... [18:02:42]
[18:02:42] Trying to send all finished work units
[18:02:42] + Attempting to get work packet
[18:02:42] + No unsent completed units remaining.
[18:02:42] Passkey found
[18:02:42] - Autosend completed
[18:02:42] - Will indicate memory of 6135 MB
[18:02:42] Gpu species not recognized.
[18:02:42] - Detect CPU. Vendor: GenuineIntel, Family: 6, Model: 10, Stepping: 5
[18:02:42] - Connecting to assignment server
[18:02:42] Connecting to http://assign-GPU.stanford.edu:8080/
[18:02:42] Posted data.
[18:02:42] Initial: 40AB; - Successful: assigned to (171.64.65.64).
[18:02:42] + News From Folding@Home: Welcome to Folding@Home
[18:02:42] Loaded queue successfully.
[18:02:42] Gpu species not recognized.
[18:02:42] Sent data
[18:02:42] Connecting to http://171.64.65.64:8080/
[18:02:43] Posted data.
[18:02:43] Initial: 0000; - Receiving payload (expected size: 44211)
[18:02:43] Conversation time very short, giving reduced weight in bandwidth avg
[18:02:43] - Downloaded at ~86 kB/s
[18:02:43] - Averaged speed for that direction ~86 kB/s
[18:02:43] + Received work.
[18:02:43] + Closed connections
[18:02:43] 
[18:02:43] + Processing work unit
[18:02:43] Core required: FahCore_15.exe
[18:02:43] Core found.
[18:02:43] Working on queue slot 01 [March 24 18:02:43 UTC]
[18:02:43] + Working ...
[18:02:43] - Calling '.\FahCore_15.exe -dir work/ -suffix 01 -nice 19 -priority 96 -nocpulock -checkpoint 3 -verbose -lifeline 5500 -version 630'

[18:02:43] 
[18:02:43] *------------------------------*
[18:02:43] Folding@Home GPU Core
[18:02:43] Version 2.15 (Tue Nov 16 09:05:18 PST 2010)
[18:02:43] 
[18:02:43] Build host: SimbiosNvdWin7
[18:02:43] Board Type: NVIDIA/CUDA
[18:02:43] Core      : x=15
[18:02:43]  Window's signal control handler registered.
[18:02:43] Preparing to commence simulation
[18:02:43] - Looking at optimizations...
[18:02:43] DeleteFrameFiles: successfully deleted file=work/wudata_01.ckp
[18:02:43] - Created dyn
[18:02:43] - Files status OK
[18:02:43] sizeof(CORE_PACKET_HDR) = 512 file=<>
[18:02:43] - Expanded 43699 -> 172159 (decompressed 393.9 percent)
[18:02:43] Called DecompressByteArray: compressed_data_size=43699 data_size=172159, decompressed_data_size=172159 diff=0
[18:02:43] - Digital signature verified
[18:02:43] 
[18:02:43] Project: 6806 (Run 3987, Clone 2, Gen 10)
[18:02:43] 
[18:02:43] Assembly optimizations on if available.
[18:02:43] Entering M.D.
[18:02:45] Tpr hash work/wudata_01.tpr:  1399213409 2679800626 620498429 1228651168 3715002604
[18:02:45] Working on 2 PEPTIDE (1-42)
[18:02:45] Client config found, loading data.
[18:02:46] Starting GUI Server
[18:02:46] Setting checkpoint frequency: 500000
[18:02:46] Setting checkpoint frequency: 500000
[18:03:59] Completed    500000 out of 50000000 steps (1%).
[18:04:00] mdrun_gpu returned 52
[18:04:00] NANs detected on GPU
[18:04:00] 
[18:04:00] Folding@home Core Shutdown: UNSTABLE_MACHINE
[18:04:03] CoreStatus = 7A (122)
[18:04:03] Sending work to server
[18:04:03] Project: 6806 (Run 3987, Clone 2, Gen 10)
[18:04:03] - Read packet limit of 540015616... Set to 524286976.
[18:04:03] - Error: Could not get length of results file work/wuresults_01.dat
[18:04:03] - Error: Could not read unit 01 file. Removing from queue.
[18:04:03] Trying to send all finished work units
[18:04:03] + No unsent completed units remaining.
[18:04:03] + -oneunit flag given and have now finished a unit. Exiting.***** Got a SIGTERM signal (2)
[18:04:03] Killing all core threads

Folding@Home Client Shutdown.

[HendricksSA]

ChrisM101, your last log still shows the same client version, machine ID and launch directory. Since the log indicates the client is creating a new work directory I'm assuming you deleted the gpu0 directory. You may need to change the machine ID to another number after deleting the work directory and queue file. That should get you another work unit, hopefully. There is still the issue of the GPU not being recognized. You should be a type 3 species xx. I'm sure GTX480s are in the whitelist. Perhaps PantherX can shed some light on this. As for the work unit, I have no explanation for why it is still wondering around 4 days later. A mod can shed light on that. Let us know how it is going.

[ChrisM101]

Well any modification to the client file fudges it up. it no longer has my name or passkey or any details in it, all i changed was the machine id from 3 to 6 and hit save

[ChrisM101]

Ok even better if I delete it all and mess with my client and it erases my name and passkey ill get a 6801 and run it and get no credit (yeah) or if i reinstall and put my info back in, I get the same 6806 again... (FAIL)

[bruce]

Well any modification to the client file fudges it up. it no longer has my name or passkey or any details in it, all i changed was the machine id from 3 to 6 and hit save

You cannot modify the client file. You must use the -configonly option (Console client) or the context menu (systray client) to make changes to it. [The EULA does prohibit changes to FAH's files. You should let FAH make those changes for you.]

How did you configure the client initially?

[bruce]

SLI GTX 480s win 7 64
Ive totally deleted and reinstalled and whats the first unit i get. 6806 ... FAIL
I guess I will just suspend all GPU work until I can get a concrete answer as to whether the UNIT is Bad or if i need to further update.

I still have no information about that specific WU except what you've provided. The database only shows completed WUs, and it should not be reassigned until the deadline expires anyway. How rapidly you process it is not a factor.

Until you fix the "Gpu species not recognized" problem there's nothing we can do.

[ChrisM101]

Updated to the latest greatest r2 client and im still failing this, and its the same Bleeping Unit.

Code: Select all

--- Opening Log file [March 25 01:22:31 UTC] 


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

                       Folding@Home Client Version 6.41r2

                          http://folding.stanford.edu

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

Launch directory: D:\Downloads\FAH_GPU_Tracker_V2\FAH GPU Tracker V2\GPU0
Executable: D:\Downloads\FAH_GPU_Tracker_V2\FAH GPU Tracker V2\FAH_GPU3.exe
Arguments: -oneunit -forcegpu nvidia_fermi -verbosity 9 -gpu 0 

[01:22:31] - Ask before connecting: No
[01:22:31] - User name: ChrisM101 (Team 111065)
[01:22:31] - User ID: 21BE43D7336DE836
[01:22:31] - Machine ID: 3
[01:22:31] 
[01:22:31] Gpu type=3 species=20.
[01:22:31] Could not open work queue, generating new queue...
[01:22:31] - Preparing to get new work unit...
[01:22:31] - Autosending finished units... [March 25 01:22:31 UTC]
[01:22:31] Cleaning up work directory
[01:22:31] Trying to send all finished work units
[01:22:31] + No unsent completed units remaining.
[01:22:31] - Autosend completed
[01:22:31] + Attempting to get work packet
[01:22:31] Passkey found
[01:22:31] - Will indicate memory of 6135 MB
[01:22:31] Gpu type=3 species=20.
[01:22:31] - Detect CPU. Vendor: GenuineIntel, Family: 6, Model: 10, Stepping: 5
[01:22:31] - Connecting to assignment server
[01:22:31] Connecting to http://assign-GPU.stanford.edu:8080/
[01:22:31] Posted data.
[01:22:31] Initial: 40AB; - Successful: assigned to (171.64.65.64).
[01:22:31] + News From Folding@Home: Welcome to Folding@Home
[01:22:31] Loaded queue successfully.
[01:22:31] Gpu type=3 species=20.
[01:22:31] Sent data
[01:22:31] Connecting to http://171.64.65.64:8080/
[01:22:31] Posted data.
[01:22:31] Initial: 0000; - Receiving payload (expected size: 44211)
[01:22:32] - Downloaded at ~43 kB/s
[01:22:32] - Averaged speed for that direction ~43 kB/s
[01:22:32] + Received work.
[01:22:32] + Closed connections
[01:22:32] 
[01:22:32] + Processing work unit
[01:22:32] Core required: FahCore_15.exe
[01:22:32] Core found.
[01:22:32] Working on queue slot 01 [March 25 01:22:32 UTC]
[01:22:32] + Working ...
[01:22:32] - Calling '.\FahCore_15.exe -dir work/ -suffix 01 -nice 19 -priority 96 -nocpulock -checkpoint 3 -verbose -lifeline 1128 -version 641'

[01:22:32] 
[01:22:32] *------------------------------*
[01:22:32] Folding@Home GPU Core
[01:22:32] Version 2.15 (Tue Nov 16 09:05:18 PST 2010)
[01:22:32] 
[01:22:32] Build host: SimbiosNvdWin7
[01:22:32] Board Type: NVIDIA/CUDA
[01:22:32] Core      : x=15
[01:22:32]  Window's signal control handler registered.
[01:22:32] Preparing to commence simulation
[01:22:32] - Looking at optimizations...
[01:22:32] DeleteFrameFiles: successfully deleted file=work/wudata_01.ckp
[01:22:32] - Created dyn
[01:22:32] - Files status OK
[01:22:32] sizeof(CORE_PACKET_HDR) = 512 file=<>
[01:22:32] - Expanded 43699 -> 172159 (decompressed 393.9 percent)
[01:22:32] Called DecompressByteArray: compressed_data_size=43699 data_size=172159, decompressed_data_size=172159 diff=0
[01:22:32] - Digital signature verified
[01:22:32] 
[01:22:32] Project: 6806 (Run 3987, Clone 2, Gen 10)
[01:22:32] 
[01:22:32] Assembly optimizations on if available.
[01:22:32] Entering M.D.
[01:22:34] Tpr hash work/wudata_01.tpr:  1399213409 2679800626 620498429 1228651168 3715002604
[01:22:34] Working on 2 PEPTIDE (1-42)
[01:22:34] Client config found, loading data.
[01:22:34] Starting GUI Server
[01:22:35] Setting checkpoint frequency: 500000
[01:22:35] Setting checkpoint frequency: 500000
[01:23:48] Completed    500000 out of 50000000 steps (1%).
[01:23:48] mdrun_gpu returned 52
[01:23:48] NANs detected on GPU
[01:23:48] 
[01:23:48] Folding@home Core Shutdown: UNSTABLE_MACHINE
[01:23:52] CoreStatus = 7A (122)
[01:23:52] Sending work to server
[01:23:52] Project: 6806 (Run 3987, Clone 2, Gen 10)
[01:23:52] - Read packet limit of 540015616... Set to 524286976.
[01:23:52] - Error: Could not get length of results file work/wuresults_01.dat
[01:23:52] - Error: Could not read unit 01 file. Removing from queue.
[01:23:52] Trying to send all finished work units
[01:23:52] + No unsent completed units remaining.
[01:23:52] + -oneunit flag given and have now finished a unit. Exiting.***** Got a SIGTERM signal (2)
[01:23:52] Killing all core threads

Folding@Home Client Shutdown.

Other people obviously have gotten this run also and its the same result. CAN I Block this Unit?

[7im]

I was just going to recommend the upgrade. You can remove the -forcegpu nvidia_fermi now.

NaNs is usually a hardware issue. Too much OC? Try standard clocks, and work up from there.

No, you can't block it, but you can move on to the next one.

Stop client. Delete the Work folder, unitinfo.txt and queue.dat file. Reconfigure (remove said flag) and change the Machine ID up or down a number. Restart client, should get different WU.

[bruce]

Well, you did get rid of the "Gpu species not recognized" problem.

I've reported the WU as a bad WU. You'll still need to follow the instructions given (twice) on the previous page of this topic and again by 7im (above) so it isn't reassigned to you or simply wait until tomorrow.

[ChrisM101]

just took the force flag off, ill post back in a min
for reference Im clocked at 720 and at 1800 memory and 1.050 volts

Code: Select all

--- Opening Log file [March 25 01:39:22 UTC] 


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

                       Folding@Home Client Version 6.41r2

                          http://folding.stanford.edu

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

Launch directory: D:\Downloads\FAH_GPU_Tracker_V2\FAH GPU Tracker V2\GPU0
Executable: D:\Downloads\FAH_GPU_Tracker_V2\FAH GPU Tracker V2\FAH_GPU3.exe
Arguments: -oneunit -verbosity 9 -gpu 0 

[01:39:22] - Ask before connecting: No
[01:39:22] - User name: ChrisM101 (Team 111065)
[01:39:22] - User ID: 21BE43D7336DE836
[01:39:22] - Machine ID: 3
[01:39:22] 
[01:39:22] Gpu type=3 species=20.
[01:39:22] Work directory not found. Creating...
[01:39:22] Could not open work queue, generating new queue...
[01:39:22] - Preparing to get new work unit...
[01:39:22] Cleaning up work directory
[01:39:22] + Attempting to get work packet
[01:39:22] Passkey found
[01:39:22] - Will indicate memory of 6135 MB
[01:39:22] Gpu type=3 species=20.
[01:39:22] - Detect CPU. Vendor: GenuineIntel, Family: 6, Model: 10, Stepping: 5
[01:39:22] - Connecting to assignment server
[01:39:22] Connecting to http://assign-GPU.stanford.edu:8080/
[01:39:22] - Autosending finished units... [March 25 01:39:22 UTC]
[01:39:22] Trying to send all finished work units
[01:39:22] + No unsent completed units remaining.
[01:39:22] - Autosend completed
[01:39:22] Posted data.
[01:39:22] Initial: 40AB; - Successful: assigned to (171.64.65.64).
[01:39:22] + News From Folding@Home: Welcome to Folding@Home
[01:39:22] Loaded queue successfully.
[01:39:22] Gpu type=3 species=20.
[01:39:22] Sent data
[01:39:22] Connecting to http://171.64.65.64:8080/
[01:39:22] Posted data.
[01:39:22] Initial: 0000; - Receiving payload (expected size: 44211)
[01:39:23] - Downloaded at ~43 kB/s
[01:39:23] - Averaged speed for that direction ~43 kB/s
[01:39:23] + Received work.
[01:39:23] + Closed connections
[01:39:23] 
[01:39:23] + Processing work unit
[01:39:23] Core required: FahCore_15.exe
[01:39:23] Core found.
[01:39:23] Working on queue slot 01 [March 25 01:39:23 UTC]
[01:39:23] + Working ...
[01:39:23] - Calling '.\FahCore_15.exe -dir work/ -suffix 01 -nice 19 -priority 96 -nocpulock -checkpoint 3 -verbose -lifeline 5268 -version 641'

[01:39:23] 
[01:39:23] *------------------------------*
[01:39:23] Folding@Home GPU Core
[01:39:23] Version 2.15 (Tue Nov 16 09:05:18 PST 2010)
[01:39:23] 
[01:39:23] Build host: SimbiosNvdWin7
[01:39:23] Board Type: NVIDIA/CUDA
[01:39:23] Core      : x=15
[01:39:23]  Window's signal control handler registered.
[01:39:23] Preparing to commence simulation
[01:39:23] - Looking at optimizations...
[01:39:23] DeleteFrameFiles: successfully deleted file=work/wudata_01.ckp
[01:39:23] - Created dyn
[01:39:23] - Files status OK
[01:39:23] sizeof(CORE_PACKET_HDR) = 512 file=<>
[01:39:23] - Expanded 43699 -> 172159 (decompressed 393.9 percent)
[01:39:23] Called DecompressByteArray: compressed_data_size=43699 data_size=172159, decompressed_data_size=172159 diff=0
[01:39:23] - Digital signature verified
[01:39:23] 
[01:39:23] Project: 6806 (Run 3987, Clone 2, Gen 10)
[01:39:23] 
[01:39:23] Assembly optimizations on if available.
[01:39:23] Entering M.D.
[01:39:25] Tpr hash work/wudata_01.tpr:  1399213409 2679800626 620498429 1228651168 3715002604
[01:39:25] Working on 2 PEPTIDE (1-42)
[01:39:25] Client config found, loading data.
[01:39:26] Starting GUI Server
[01:39:26] Setting checkpoint frequency: 500000
[01:39:26] Setting checkpoint frequency: 500000
[01:39:26] ***** Got a SIGTERM signal (2)
[01:39:26] Killing all core threads

Folding@Home Client Shutdown.


--- Opening Log file [March 25 01:39:28 UTC] 


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

                       Folding@Home Client Version 6.41r2

                          http://folding.stanford.edu

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

Launch directory: D:\Downloads\FAH_GPU_Tracker_V2\FAH GPU Tracker V2\GPU0
Executable: D:\Downloads\FAH_GPU_Tracker_V2\FAH GPU Tracker V2\FAH_GPU3.exe
Arguments: -oneunit -verbosity 9 -gpu 0 

[01:39:28] - Ask before connecting: No
[01:39:28] - User name: ChrisM101 (Team 111065)
[01:39:28] - User ID: 21BE43D7336DE836
[01:39:28] - Machine ID: 3
[01:39:28] 
[01:39:28] Gpu type=3 species=20.
[01:39:28] Loaded queue successfully.
[01:39:28] 
[01:39:28] + Processing work unit
[01:39:28] - Autosending finished units... [01:39:28]
[01:39:28] Core required: FahCore_15.exe
[01:39:28] Trying to send all finished work units
[01:39:28] + No unsent completed units remaining.
[01:39:28] - Autosend completed
[01:39:28] Core found.
[01:39:28] Working on queue slot 01 [March 25 01:39:28 UTC]
[01:39:28] + Working ...
[01:39:28] - Calling '.\FahCore_15.exe -dir work/ -suffix 01 -nice 19 -priority 96 -nocpulock -checkpoint 3 -verbose -lifeline 2924 -version 641'

[01:39:28] 
[01:39:28] *------------------------------*
[01:39:28] Folding@Home GPU Core
[01:39:28] Version 2.15 (Tue Nov 16 09:05:18 PST 2010)
[01:39:28] 
[01:39:28] Build host: SimbiosNvdWin7
[01:39:28] Board Type: NVIDIA/CUDA
[01:39:28] Core      : x=15
[01:39:28]  Window's signal control handler registered.
[01:39:28] Preparing to commence simulation
[01:39:28] - Ensuring status. Please wait.
[01:39:34] 
[01:39:34] Folding@home Core Shutdown: CLIENT_DIED
[01:40:55] INE
[01:40:58] CoreStatus = 7A (122)
[01:40:58] Sending work to server
[01:40:58] Project: 6806 (Run 3987, Clone 2, Gen 10)
[01:40:58] - Read packet limit of 540015616... Set to 524286976.
[01:40:58] - Error: Could not get length of results file work/wuresults_01.dat
[01:40:58] - Error: Could not read unit 01 file. Removing from queue.
[01:40:58] Trying to send all finished work units
[01:40:58] + No unsent completed units remaining.
[01:40:58] + -oneunit flag given and have now finished a unit. Exiting.***** Got a SIGTERM signal (2)
[01:40:58] Killing all core threads

Folding@Home Client Shutdown.


--- Opening Log file [March 25 01:40:58 UTC] 


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

                       Folding@Home Client Version 6.41r2

                          http://folding.stanford.edu

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

Launch directory: D:\Downloads\FAH_GPU_Tracker_V2\FAH GPU Tracker V2\GPU0
Executable: D:\Downloads\FAH_GPU_Tracker_V2\FAH GPU Tracker V2\FAH_GPU3.exe
Arguments: -oneunit -verbosity 9 -gpu 0 

[01:40:58] - Ask before connecting: No
[01:40:58] - User name: ChrisM101 (Team 111065)
[01:40:58] - User ID: 21BE43D7336DE836
[01:40:58] - Machine ID: 3
[01:40:58] 
[01:40:58] Gpu type=3 species=20.
[01:40:58] Loaded queue successfully.
[01:40:58] - Preparing to get new work unit...
[01:40:58] Cleaning up work directory
[01:40:58] - Autosending finished units... [March 25 01:40:58 UTC]
[01:40:58] Trying to send all finished work units
[01:40:58] + Attempting to get work packet
[01:40:58] + No unsent completed units remaining.
[01:40:58] Passkey found
[01:40:58] - Autosend completed
[01:40:58] - Will indicate memory of 6135 MB
[01:40:58] Gpu type=3 species=20.
[01:40:58] - Detect CPU. Vendor: GenuineIntel, Family: 6, Model: 10, Stepping: 5
[01:40:58] - Connecting to assignment server
[01:40:58] Connecting to http://assign-GPU.stanford.edu:8080/
[01:40:59] Posted data.
[01:40:59] Initial: 40AB; - Successful: assigned to (171.64.65.64).
[01:40:59] + News From Folding@Home: Welcome to Folding@Home
[01:40:59] Loaded queue successfully.
[01:40:59] Gpu type=3 species=20.
[01:40:59] Sent data
[01:40:59] Connecting to http://171.64.65.64:8080/
[01:40:59] Posted data.
[01:40:59] Initial: 0000; - Receiving payload (expected size: 44211)
[01:41:00] - Downloaded at ~43 kB/s
[01:41:00] - Averaged speed for that direction ~43 kB/s
[01:41:00] + Received work.
[01:41:00] + Closed connections
[01:41:00] 
[01:41:00] + Processing work unit
[01:41:00] Core required: FahCore_15.exe
[01:41:00] Core found.
[01:41:00] Working on queue slot 02 [March 25 01:41:00 UTC]
[01:41:00] + Working ...
[01:41:00] - Calling '.\FahCore_15.exe -dir work/ -suffix 02 -nice 19 -priority 96 -nocpulock -checkpoint 3 -verbose -lifeline 4464 -version 641'

[01:41:00] 
[01:41:00] *------------------------------*
[01:41:00] Folding@Home GPU Core
[01:41:00] Version 2.15 (Tue Nov 16 09:05:18 PST 2010)
[01:41:00] 
[01:41:00] Build host: SimbiosNvdWin7
[01:41:00] Board Type: NVIDIA/CUDA
[01:41:00] Core      : x=15
[01:41:00]  Window's signal control handler registered.
[01:41:00] Preparing to commence simulation
[01:41:00] - Looking at optimizations...
[01:41:00] DeleteFrameFiles: successfully deleted file=work/wudata_02.ckp
[01:41:00] - Created dyn
[01:41:00] - Files status OK
[01:41:00] sizeof(CORE_PACKET_HDR) = 512 file=<>
[01:41:00] - Expanded 43699 -> 172159 (decompressed 393.9 percent)
[01:41:00] Called DecompressByteArray: compressed_data_size=43699 data_size=172159, decompressed_data_size=172159 diff=0
[01:41:00] - Digital signature verified
[01:41:00] 
[01:41:00] Project: 6806 (Run 3987, Clone 2, Gen 10)
[01:41:00] 
[01:41:00] Assembly optimizations on if available.
[01:41:00] Entering M.D.
[01:41:02] Tpr hash work/wudata_02.tpr:  1399213409 2679800626 620498429 1228651168 3715002604
[01:41:02] Working on 2 PEPTIDE (1-42)
[01:41:02] Client config found, loading data.
[01:41:02] Setting checkpoint frequency: 500000
[01:41:02] Setting checkpoint frequency: 500000
[01:41:02] Starting GUI Server
[01:42:17] Completed    500000 out of 50000000 steps (1%).
[01:42:17] mdrun_gpu returned 52
[01:42:17] NANs detected on GPU
[01:42:17] 
[01:42:17] Folding@home Core Shutdown: UNSTABLE_MACHINE
[01:42:20] CoreStatus = 7A (122)
[01:42:20] Sending work to server
[01:42:20] Project: 6806 (Run 3987, Clone 2, Gen 10)
[01:42:20] - Read packet limit of 540015616... Set to 524286976.
[01:42:20] - Error: Could not get length of results file work/wuresults_02.dat
[01:42:20] - Error: Could not read unit 02 file. Removing from queue.
[01:42:20] Trying to send all finished work units
[01:42:20] + No unsent completed units remaining.
[01:42:20] + -oneunit flag given and have now finished a unit. Exiting.***** Got a SIGTERM signal (2)
[01:42:20] Killing all core threads

Folding@Home Client Shutdown.

LAST fail... Ill try deleting work and que again and the other file. I havent seen anyone recommend it yet so maybe.
Hmm deleted the WU and files, but i get it back. Not sure if i can manually change the MACHINE ID.