F@H Bad Work Unit

[iwajabitw]

Seeing several of these when I have a failed process. Log says BAD_WORK_UNIT. My question , is this common like MW's computation error,and Seti's Inconclusive, or is it something on my end? I have it happen on either gpu several times a day. If it's common that's fine, I just wish the client would flush it out, download a new unit and proceed on. I don't like having to baby sit it all day, to clear a failed task out.

 

[Markfw]

are your video cards overclocked ? or running hot ? When that happens sometomes it will say that, so it MIGHT be your setup. I have not had that in years. I now don;t OC my cards at all, but buy the fastest factory OC'ed parts I can find. Has worked well for me.

Edit: I now have 11 active video cards, and have not hard that in years, so I really thing it may be something in your setup, sorry.

 

[TennesseeTony]

He's not overclocked.

....Nothings over clocked, just Edge, HWC monitor, and afterburner open.

Now that you mention it, Here's my error log, not really sure of the severity except that one bad WU though:

*********************** Log Started 2016-10-24T14:36:41Z ***********************
18:58:24:WARNING:WU02:FS00:FahCore returned an unknown error code which probably indicates that it crashed
18:58:24:WARNING:WU02:FS00:FahCore returned: UNKNOWN_ENUM (127 = 0x7f)
******************************* Date: 2016-10-24 *******************************
21:58:20:WU00:FS01:0x21:WARNING:Console control signal 1 on PID 10012
21:58:20:WU01:FS00:0x21:WARNING:Console control signal 1 on PID 11976
21:58:34:WARNING:WU00:FS01:FahCore returned: BAD_WORK_UNIT (114 = 0x72)
21:58:35:WARNING:WU00:FS01:Server did not like results, dumping
******************************* Date: 2016-10-25 *******************************
******************************* Date: 2016-10-25 *******************************
******************************* Date: 2016-10-25 *******************************
******************************* Date: 2016-10-25 *******************************
******************************* Date: 2016-10-26 *******************************
03:50:09:WU02:FS01:0x21:WARNING:Console control signal 1 on PID 14052
03:50:09:WU00:FS00:0x21:WARNING:Console control signal 1 on PID 7968
******************************* Date: 2016-10-26 *******************************
09:53:41:WU02:FS00:0x21:WARNING:Console control signal 1 on PID 6484
******************************* Date: 2016-10-26 *******************************
20:10:58:ERROR:WU00:FS00:Exception: Server did not assign work unit
******************************* Date: 2016-10-26 *******************************

 

[Markfw]

Tony, you may have the same issue, even if not overclocked. The only way to know, is to check your observed ppd, and the credited ppd. The 2 are quite often different. If a lot different, your units could be getting rejected. I have seen this, but not often.

edit: OOPS, I have seen it;;; in the last 3 days on my 1080:

22:11:50:WU01:FS02:0x21:Completed 2750000 out of 5000000 steps (55%)
22:11:59:WU00:FS01:0xa4:Completed 57500 out of 250000 steps (23%)
22:13:18:WU02:FS00:0x21:Completed 75000 out of 2500000 steps (3%)
22:13:25:WU01:FS02:0x21:Completed 2800000 out of 5000000 steps (56%)
22:13:38:WU00:FS01:0xa4:Completed 60000 out of 250000 steps (24%)
22:14:59:WU01:FS02:0x21:Completed 2850000 out of 5000000 steps (57%)
22:15:17:WU00:FS01:0xa4:Completed 62500 out of 250000 steps (25%)
22:16:34:WU01:FS02:0x21:Completed 2900000 out of 5000000 steps (58%)
22:16:56:WU00:FS01:0xa4:Completed 65000 out of 250000 steps (26%)
22:16:59:WU02:FS00:0x21:Completed 100000 out of 2500000 steps (4%)
22:17:58:WU01:FS02:0x21:ERROR:exception: Error downloading array energyBuffer: clEnqueueReadBuffer (-5)
22:17:58:WU01:FS02:0x21:Saving result file logfile_01.txt
22:17:58:WU01:FS02:0x21:Saving result file log.txt
22:17:58:WU01:FS02:0x21:Folding@home Core Shutdown: BAD_WORK_UNIT
22:18:04:WARNING:WU01:FS02:FahCore returned: BAD_WORK_UNIT (114 = 0x72)
22:18:04:WU01:FS02:Sending unit results: id:01 state:SEND error:FAULTY project:13500 run:3 clone:21 gen:21 core:0x21 unit:0x0000001f8ca304f457a35a2a24240247
22:18:04:WU01:FS02:Uploading 3.13KiB to 140.163.4.244
22:18:04:WU01:FS02:Connecting to 140.163.4.244:8080
22:18:04:WU01:FS02:Upload complete
22:18:04:WU01:FS02:Server responded WORK_ACK (400)
22:18:04:WU01:FS02:Cleaning up
22:18:04:WU03:FS02:Connecting to assign-GPU.stanford.edu:80
22:18:05:WU03:FS02:News:
22:18:05:WU03:FS02:Assigned to work server 140.163.4.245
22:18:05:WU03:FS02:Requesting new work unit for slot 02: READY gpu:3:\"GP104 [GeForce GTX 1080]\" from 140.163.4.245
22:18:05:WU03:FS02:Connecting to 140.163.4.245:8080
22:18:05:WU03:FS02ownloading 5.13MiB
22:18:08:WU03:FS02ownload complete
22:18:08:WU03:FS02:Received Unit: id:03 stateOWNLOAD error:NO_ERROR project:10495 run:15 clone:47 gen:98 core:0x21 unit:0x000000828ca304f556ba63e2a8fa8404
22:18:08:WU03:FS02:Starting
22:18:08:WU03:FS02:Running FahCore: \"C:\\Program Files (x86)\\FAHClient/FAHCoreWrapper.exe\" C:/ProgramData/FAHClient/cores/web.stanford.edu/~pande/Win32/AMD64/NVIDIA/Fermi/Core_21.fah/FahCore_21.exe -dir 03 -suffix 01 -version 702 -lifeline 3316 -checkpoint 15 -gpu 1
22:18:08:WU03:FS02:Started FahCore on PID 5860
22:18:08:WU03:FS02:Core PID:5940
22:18:08:WU03:FS02:FahCore 0x21 started
22:18:08:WU03:FS02:0x21:*********************** Log Started 2016-10-23T22:18:08Z ***********************
22:18:08:WU03:FS02:0x21roject: 10495 (Run 15, Clone 47, Gen 98)
22:18:08:WU03:FS02:0x21:Unit: 0x000000828ca304f556ba63e2a8fa8404
22:18:08:WU03:FS02:0x21:CPU: 0x00000000000000000000000000000000
22:18:08:WU03:FS02:0x21:Machine: 2
22:18:08:WU03:FS02:0x21:Reading tar file core.xml
22:18:08:WU03:FS02:0x21:Reading tar file system.xml
22:18:08:WU03:FS02:0x21:Reading tar file integrator.xml
22:18:08:WU03:FS02:0x21:Reading tar file state.xml
22:18:09:WU03:FS02:0x21igital signatures verified
22:18:09:WU03:FS02:0x21:Folding@home GPU Core21 Folding@home Core
22:18:09:WU03:FS02:0x21:Version 0.0.17
22:18:17:WU03:FS02:0x21:Completed 0 out of 5000000 steps (0%)
22:18:17:WU03:FS02:0x21:Temperature control disabled. Requirements: single Nvidia GPU, tmax must be < 110 and twait >= 900
22:18:36:WU00:FS01:0xa4:Completed 67500 out of 250000 steps (27%)
22:19:27:WU03:FS02:0x21:Completed 50000 out of 5000000 steps (1%)
22:20:15:WU00:FS01:0xa4:Completed 70000 out of 250000 steps (28%)
22:20:38:WU03:FS02:0x21:Completed 100000 out of 5000000 steps (2%)
22:20:43:WU02:FS00:0x21:Completed 125000 out of 2500000 steps (5%)
22:21:49:WU03:FS02:0x21:Completed 150000 out of 5000000 steps (3%)
22:21:54:WU00:FS01:0xa4:Completed 72500 out of 250000 steps (29%)

 

[iwajabitw]

Thanks, I was thinking about the heat today at work. Top card does get up to 80c, 2nd around 74c. I had ordered some Fractal Venture HP-14 140mm case fans last week that just got here. They move some air too. Have 2 on the intake and 2 exhausting out of the top of the case. Also updated Nvidia drivers to 375.63 just now to see if things smooth out. I'm even running the client on medium instead of full to see if that helps.

 

[TennesseeTony]

https://folding.stanford.edu/home/download-utilities/

Some of these utilities might prove helpful if trouble persists. Maybe give DDU (display driver uninstaller) a try, to completely erase all former vid drivers.

 

[iwajabitw]

Both cards passed the gpu tests with now errors.

 

[iwajabitw]

Nest one that I get, Ill post the log unless I am at work. Too hard to copy and paste from my phone through Teamviewer.

 

[iwajabitw]

4:06:22:WU01:FS02:Starting
04:06:22:WU01:FS02:Running FahCore: "C:\Program Files (x86)\FAHClient/FAHCoreWrapper.exe" C:/Users/witjm/AppData/Roaming/FAHClient/cores/web.stanford.edu/~pande/Win32/AMD64/NVIDIA/Fermi/Core_21.fah/FahCore_21.exe -dir 01 -suffix 01 -version 704 -lifeline 3584 -checkpoint 3 -gpu 1 -gpu-vendor nvidia
04:06:22:WU01:FS02:Started FahCore on PID 4660
04:06:22:WU01:FS02:Core PID:7200
04:06:22:WU01:FS02:FahCore 0x21 started
04:06:25:WU01:FS02:0x21:*********************** Log Started 2016-10-27T04:06:24Z ***********************
04:06:25:WU01:FS02:0x21roject: 9213 (Run 39, Clone 10, Gen 27)
04:06:25:WU01:FS02:0x21:Unit: 0x0000004eab436c685796c11cfd06258a
04:06:25:WU01:FS02:0x21:CPU: 0x00000000000000000000000000000000
04:06:25:WU01:FS02:0x21:Machine: 2
04:06:25:WU01:FS02:0x21:Reading tar file core.xml
04:06:25:WU01:FS02:0x21:Reading tar file integrator.xml
04:06:25:WU01:FS02:0x21:Reading tar file state.xml
04:06:25:WU01:FS02:0x21:Reading tar file system.xml
04:06:25:WU01:FS02:0x21igital signatures verified
04:06:25:WU01:FS02:0x21:Folding@home GPU Core21 Folding@home Core
04:06:25:WU01:FS02:0x21:Version 0.0.17
04:06:57:WU01:FS02:0x21:ERROR:exception: Error downloading array interactionCount: clEnqueueReadBuffer (-5)
04:06:57:WU01:FS02:0x21:Saving result file logfile_01.txt
04:06:57:WU01:FS02:0x21:Saving result file log.txt
04:06:57:WU01:FS02:0x21:Folding@home Core Shutdown: BAD_WORK_UNIT
04:07:00:WARNING:WU01:FS02:FahCore returned: BAD_WORK_UNIT (114 = 0x72)

 

[iwajabitw]

Woke up to both gpu slots with the same error above. After all the searching last night, I'm not convinced yet that it's on my end.

 

[Markfw]

are your 2 cards in SLI ? I read that its OK, but I have never tried it. I have 8 computers with 11 cards running, and they don't have the issue you show above.

 

[iwajabitw]

They are in sli, but in the NVidia control panel, I set up boinc and f@h for single gpu use.

 

[Markfw]

Well, I would recommend that you un-SLI them and try it.

First, remove the GPU client in FAHControl, then shut down, then remove the link, then start up, and if FAHControl did not auto-add GPU client, then add it back in

 

[iwajabitw]

Well I might be dropping f@H if that's the case, I game, as well as boinc runs just fine on other projects.

 

[Markfw]

Well I might be dropping f@H if that's the case, I game, as well as boinc runs just fine on other projects.

Well, you won't even know if thats the problem unless you try. I personally would love to know if thats a problem.

 

[TennesseeTony]

And I'd personally love you see it work for you and have you fold for us during December's race. BOINC's good too though, if we can't get fah going for you.

 

[iwajabitw]

Finally had some time to dig. Apperently the 375.xx drivers have messed up f@h for a bunch of NVidia folks, those like me that let the Geforce Experience update them for game ready. So rolling back to 373.06, which seems to be working good for most.

Forgot the link

https://foldingforum.org/viewtopic.php?f=80&t=29276

 

[iwajabitw]

First morning, no failed tasks! Now that this works, off to setup and config Untangle firewall. The real headache starts.

 

[iwajabitw]

24hrs now with no failed tasks, game on!