WU 8020 (3, 31, 88) update core cannot find core 15.fah

[low_dough_tb_1g]

For the last 2.5 days, by FAH Control has been trying to update core 15.fah

In the FAH Control client on the machine in question, the Status tab shows "UpdateCore" and it is showing in the Work Unit pane "Waiting on core update" and keeps trying to get that core.

The logs say it is trying to get the file from here:

"04:42:42:WU00:FS01:Downloading core from http://www.stanford.edu/~pande/Win32/AMD64/Core_15.fah
04:42:42:WU00:FS01:Connecting to http://www.stanford.edu:80
04:42:43:ERROR:WU00:FS01:Exception: Failed to access core package."

http://www.stanford.edu/~pande/Win32/AMD64/Core_15.fah returns error 404 in a browser.
I found core 16 here : http://www.stanford.edu/~pande/Win32/AMD64/Core_16.fah
and put it in the (what i assume is the right place) :
C:\Program Files (x86)\FAHClient\cores\www.stanford.edu\~pande\Win32\AMD64\NVIDIA\Fermi
where the previous Core_15.fah was, but that has not worked.

I have tried deleting the Work folder for that slot. It did not work either.

I downloaded and installed the newest version of the FAH client and installed over the top of the previous version, but no change.

Not a newbie here, and really want my Fermi card to churn out some work:
http://folding.extremeoverclocking.com/ ... =&u=379149

The GPU in question has performed without notable errors since I installed it in March of this year.
It is an EVGA GTX 550Ti 2GB card.
The other card that is installed in the same machine : EVGA 880GTS 640 is still working away.
3rd slot is my Core2Duo e6750 @ 3.21GHZ, and it is meeting the deadlines nicely.
Nvidia driver version : 8.17.12.9551
Windows 7 Ultimate x64

Anything else you can recommend I can try?

Edit: updated with system info

[P5-133XL]

Try Prefer IPv4 over IPv6. There have been several issues similar to this with IPv6 being locally implemented.

[uncle_fungus]

I've passed this info on up the chain for confirmation but I think this is a WU issue telling the client to download a non-existent core.

[bruce]

Searching backwards in the log to a point just before it tried to download the core, you should find a segment where the client contacts the Assignment Server and then downloads a WU from another server. Please post that segment of your log, including the first time it tried to download the core and failed.

[low_dough_tb_1g]

Bruce:
Unfortunately I have already deleted that work folder, and do not have those logs anymore.

P5-133XL:
Preferring ipv4 or ipv6 will not matter if the file is not in the folder the client looks in for the file.
In this folder :
http://www.stanford.edu/~pande/Win32/AMD64/
neither I , nor the client, can find it.
But thanks for the tip.

Uncle Fungus:
Thanks for passing this up the chain, I believe that is exactly what is going on, and wish i could help them figure out what happened.
I would really like it if they could help me start folding on that slot again, too.

I guess the next step would be to uninstall the entire client, and all settings, then start again from scratch, but it is only 1 of 3 slots that is having problems.

[P5-133XL]

I would really like it if they could help me start folding on that slot again, too.

I guess the next step would be to uninstall the entire client, and all settings, then start again from scratch, but it is only 1 of 3 slots that is having problems.

If the problem is a WU that requested a non-existent core then you do not need uninstall. You just need different WU's and I can help you with that one.

Pause the entire client. Record the work queue ID numbers for the offending slots. Then go to the work folder (Start->All Programs->FAHClient->Data Directory->Work) and delete the folders with the same numbers as the work queue ID numbers you previously recorded. Now un-pause FAHControl and it will clean everything up and download brand new WU's.

[P5-133XL]

Bruce:
Unfortunately I have already deleted that work folder, and do not have those logs anymore.

The logs are stored in the work folder but also in other places, so they still exist even if you deleted the work folder. The current log is stored in the data Directory (Start->All Programs->FAHClient->Data Directory->Log.txt) and if you need an older log they are stored in the log folder.

[low_dough_tb_1g]

Bruce:
Unfortunately I have already deleted that work folder, and do not have those logs anymore.

The logs are stored in the work folder but also in other places, so they still exist even if you deleted the work folder. The current log is stored in the data Directory (Start->All Programs->FAHClient->Data Directory->Log.txt) and if you need an older log they are stored in the log folder.

Thanks for the assist, here is the first time this is mentioned in the logs :

21:36:15:WU00:FS01:Connecting to assign-GPU.stanford.edu:80
21:36:15:WU00:FS01:News: Welcome to Folding@Home
21:36:15:WU00:FS01:Assigned to work server 171.67.108.143
21:36:15:WU00:FS01:Requesting new work unit for slot 01: READY gpu:1:"GF116 [GeForce GTX 550 Ti]" from 171.67.108.143
21:36:15:WU00:FS01:Connecting to 171.67.108.143:8080
21:36:16:WU00:FS01:Downloading 142.73KiB
21:36:17:WU00:FS01:Download complete
21:36:17:WU00:FS01:Received Unit: id:00 state:DOWNLOAD error:OK project:8020 run:3 clone:31 gen:88 core:0x15 unit:0x000000676953ee2f4f96798ffdef8fa4
21:36:17:WU00:FS01:Downloading core from http://www.stanford.edu/~pande/Win32/AMD64/Core_15.fah
21:36:17:WU00:FS01:Connecting to http://www.stanford.edu:80
21:36:17:ERROR:WU00:FS01:Exception: Failed to access core package.
21:37:17:WU00:FS01:Downloading core from http://www.stanford.edu/~pande/Win32/AMD64/Core_15.fah
21:37:17:WU00:FS01:Connecting to http://www.stanford.edu:80
21:37:17:ERROR:WU00:FS01:Exception: Failed to access core package.

Edit, and now i can see the RCG is actually (3,31,88) not the (2,31,88) i thought it was, so i edited the subject of my first post to correct it there.

[bollix47]

Can you please copy/paste the System section of your log including the config part which should be below the system info. Basically, press the Refresh button above the log and scroll up to the top of the log.

Thanks

btw my Core_15.fah appears to have come from http://www.stanford.edu\~pande\Win32\x86\NVIDIA\Fermi\ but this might be a new core and it may be stored elsewhere.

Looking at the system info part of the log may shed some light on the problem.

p.s. I've escalated this to try and determine why the core isn't where the client is looking for it.

[bollix47]

The core file has been temporarily copied to the location that your client is trying to find it while the source of the problem is being investigated. Please try again and let us know the outcome.

Thanks

[low_dough_tb_1g]

Can you please copy/paste the System section of your log including the config part which should be below the system info. Basically, press the Refresh button above the log and scroll up to the top of the log.

Thanks

btw my Core_15.fah appears to have come from http://www.stanford.edu\~pande\Win32\x86\NVIDIA\Fermi\ but this might be a new core and it may be stored elsewhere.

Looking at the system info part of the log may shed some light on the problem.

p.s. I've escalated this to try and determine why the core isn't where the client is looking for it.

big paste :

*********************** Log Started 2012-08-01T11:29:10Z ***********************
11:29:10:************************* Folding@home Client *************************
11:29:10: Website: http://folding.stanford.edu/
11:29:10: Copyright: (c) 2009-2012 Stanford University
11:29:10: Author: Joseph Coffland <joseph@cauldrondevelopment.com>
11:29:10: Args: --lifeline 4620 --command-port=36330
11:29:10: Config: C:/ProgramData/FAHClient/config.xml
11:29:10:******************************** Build ********************************
11:29:10: Version: 7.1.48
11:29:10: Date: Feb 15 2012
11:29:10: Time: 16:22:58
11:29:10: SVN Rev: 3271
11:29:10: Branch: fah/trunk/client
11:29:10: Compiler: Intel(R) C++ MSVC 1500 mode 1200
11:29:10: Options: /TP /nologo /EHa /Qdiag-disable:4297,4103,1786,279 /Ox -arch:SSE
11:29:10: /QaxSSE2,SSE3,SSSE3,SSE4.1,SSE4.2 /Qopenmp /Qrestrict /MT
11:29:10: Platform: win32 XP
11:29:10: Bits: 32
11:29:10: Mode: Release
11:29:10:******************************* System ********************************
11:29:10: CPU: Intel(R) Core(TM)2 Duo CPU E6750 @ 2.66GHz
11:29:10: CPU ID: GenuineIntel Family 6 Model 15 Stepping 11
11:29:10: CPUs: 2
11:29:10: Memory: 6.00GiB
11:29:10: Free Memory: 4.30GiB
11:29:10: Threads: WINDOWS_THREADS
11:29:10: On Battery: false
11:29:10: UTC offset: -4
11:29:10: PID: 4668
11:29:10: CWD: C:/ProgramData/FAHClient
11:29:10: OS: Windows 7 Ultimate
11:29:10: OS Arch: AMD64
11:29:10: GPUs: 2
11:29:10: GPU 0: NVIDIA:1 G80 [GeForce 8800 GTS]
11:29:10: GPU 1: NVIDIA:1 GF116 [GeForce GTX 550 Ti]
11:29:10: CUDA: 1.0
11:29:10: CUDA Driver: 4010
11:29:10:Win32 Service: false
11:29:10:***********************************************************************
11:29:10:<config>
11:29:10: <!-- FahCore Control -->
11:29:10: <cpu-usage v='90'/>
11:29:10:
11:29:10: <!-- Folding Slot Configuration -->
11:29:10: <gpu v='true'/>
11:29:10:
11:29:10: <!-- Network -->
11:29:10: <proxy v=':8080'/>
11:29:10:
11:29:10: <!-- User Information -->
11:29:10: <passkey v='********************************'/>
11:29:10: <team v='101093'/>
11:29:10: <user v='low_dough_tb_1g'/>
11:29:10:
11:29:10: <!-- Folding Slots -->
11:29:10: <slot id='0' type='GPU'/>
11:29:10: <slot id='1' type='GPU'/>
11:29:10: <slot id='2' type='SMP'>
11:29:10: <cpus v='-1'/>
11:29:10: </slot>
11:29:10:</config>
11:29:11:Trying to access database...
11:29:11:Successfully acquired database lock
11:29:14:Enabled folding slot 00: READY gpu:0:"G80 [GeForce 8800 GTS]"
11:29:14:Enabled folding slot 01: READY gpu:1:"GF116 [GeForce GTX 550 Ti]"
11:29:14:Enabled folding slot 02: READY smp:2
11:29:14:WU03:FS01:Starting
11:29:14:WU03:FS01:Running FahCore: "C:\Program Files (x86)\FAHClient/FAHCoreWrapper.exe" C:/ProgramData/FAHClient/cores/www.stanford.edu/~pande/Win32/AMD64/NVI ... ore_15.exe -dir 03 -suffix 01 -version 701 -lifeline 4668 -checkpoint 15 -cpu 90 -gpu 1
11:29:14:WU03:FS01:Started FahCore on PID 3432
11:29:14:WU03:FS01:Core PID:2688
11:29:14:WU03:FS01:FahCore 0x15 started
11:29:14:Server connection id=1 on 0.0.0.0:36330 from 127.0.0.1
11:29:14:WU00:FS00:Starting
11:29:14:WU00:FS00:Running FahCore: "C:\Program Files (x86)\FAHClient/FAHCoreWrapper.exe" C:/ProgramData/FAHClient/cores/www.stanford.edu/~pande/Win32/AMD64/NVI ... ore_11.exe -dir 00 -suffix 01 -version 701 -lifeline 4668 -checkpoint 15 -cpu 90 -gpu 0
11:29:14:WU00:FS00:Started FahCore on PID 3524
11:29:14:WU00:FS00:Core PID:3520
11:29:14:WU00:FS00:FahCore 0x11 started
11:29:14:WU59:FS02:Starting
11:29:14:WU59:FS02:Running FahCore: "C:\Program Files (x86)\FAHClient/FAHCoreWrapper.exe" C:/ProgramData/FAHClient/cores/www.stanford.edu/~pande/Win32/AMD64/Cor ... ore_a4.exe -dir 59 -suffix 01 -version 701 -lifeline 4668 -checkpoint 15 -cpu 90 -np 2
11:29:14:WU59:FS02:Started FahCore on PID 3512
11:29:15:WU59:FS02:Core PID:4720
11:29:15:WU59:FS02:FahCore 0xa4 started

[low_dough_tb_1g]

The core file has been temporarily copied to the location that your client is trying to find it while the source of the problem is being investigated. Please try again and let us know the outcome.

Thanks

I came home and found the status of the slot had changed.
It now said "Running"
hooray!

Of course, in all my troubleshooting, i had changed the Client slots in the configuration back to both gpu slots using cuda index:-1 and this slot was not doing any work.
I changed them back to cuda index:0 and cuda index:1, and both slots are now maxing out my GPUS.
Temp holding steady at 73C ( a little hot , but not out of the ordinary) and 8min 47 sec/ frame production , results will be returned in just under 15 hours

Let me know what else i can provide to help you troubleshoot.

[bollix47]

C:/ProgramData/FAHClient/cores/www.stanford.edu/~pande/Win32/AMD64/NVI ... ore_15.exe

It appears the core was downloaded from the correct location rather than from the location it was trying to download from last night. Looks like the problem has been corrected.

Thanks for the update.

[low_dough_tb_1g]

C:/ProgramData/FAHClient/cores/www.stanford.edu/~pande/Win32/AMD64/NVI ... ore_15.exe

It appears the core was downloaded from the correct location rather than from the location it was trying to download from last night. Looks like the problem has been corrected.

Thanks for the update.

Actually, that was from the old logs.

[bollix47]

Okay, can you find the Core_15.fah latest download in your log and post that section here please.