Project: 6050 (Run 0, Clone 119, Gen 406)

[bollix47]

FYI

Code: Select all

[10:21:42] Connecting to http://171.64.65.54:8080/
[10:21:42] Posted data.
[10:21:42] Initial: 0000; - Receiving payload (expected size: 42776)
[10:21:43] - Downloaded at ~41 kB/s
[10:21:43] - Averaged speed for that direction ~446 kB/s
[10:21:43] + Received work.
[10:21:43] Trying to send all finished work units
[10:21:43] + No unsent completed units remaining.
[10:21:43] + Closed connections
[10:21:43]
[10:21:43] + Processing work unit
[10:21:43] Core required: FahCore_a3.exe
[10:21:43] Core found.
[10:21:43] Working on queue slot 03 [June 4 10:21:43 UTC]
[10:21:43] + Working ...
[10:21:43] - Calling '.\FahCore_a3.exe -dir work/ -nice 19 -suffix 03 -np
iority 96 -checkpoint 30 -verbose -lifeline 2812 -version 634'

[10:21:43]
[10:21:43] *------------------------------*
[10:21:43] Folding@Home Gromacs SMP Core
[10:21:43] Version 2.27 (Dec. 15, 2010)
[10:21:43]
[10:21:43] Preparing to commence simulation
[10:21:43] - Looking at optimizations...
[10:21:43] - Created dyn
[10:21:43] - Files status OK
[10:21:43] - Expanded 42264 -> 115421 (decompressed 273.0 percent)
[10:21:43] Called DecompressByteArray: compressed_data_size=42264 data_siz
21, decompressed_data_size=115421 diff=0
[10:21:43] - Digital signature verified
[10:21:43]
[10:21:43] Project: 6050 (Run 0, Clone 119, Gen 406)
[10:21:43]
[10:21:43] Assembly optimizations on if available.
[10:21:43] Entering M.D.
[10:21:49] Mapping NT from 15 to 15
[10:41:29] CoreStatus = C0000417 (-1073740777)
[10:41:29] Client-core communications error: ERROR 0xc0000417
[10:41:29] Deleting current work unit & continuing...

Windows 7 Pro 64-bit pop-up ... FahCore_a3.exe has stopped running.

Code: Select all

Problem Event Name:	BEX
  Application Name:	FahCore_a3.exe
  Application Version:	0.0.0.0
  Application Timestamp:	4d4720af
  Fault Module Name:	FahCore_a3.exe
  Fault Module Version:	0.0.0.0
  Fault Module Timestamp:	4d4720af
  Exception Offset:	0008816d
  Exception Code:	c0000417
  Exception Data:	00000000
  OS Version:	6.1.7601.2.1.0.256.48
  Locale ID:	4105
  Additional Information 1:	88f7
  Additional Information 2:	88f70b5904b84d8cc95e82b3c6f7647f
  Additional Information 3:	fa46
  Additional Information 4:	fa46fc87ddfc5983a29bd91072f459f4

[Sleepee]

Running into the same problem here.

i5-760 @ Stock
4GB DDR3-1600

Deleting the WU from queue and/or deleting the work folder has no effect; the exact same unit redownloads. It's currently stuck on that rig, and till then, to prevent further WU failures, I've turned it off.

[PantherX]

The WU has been reported as a bad one:

The WU (P6050,R0,C119,G406) has been reported as a bad WU.

Thanks for the report.

Sleepee -> Welcome to the F@H Forum Sleepee,
Please read this post to resolve your issue (viewtopic.php?f=19&t=16526#p164322).

[Grandpa_01]

The WU has been reported as a bad one:

The WU (P6050,R0,C119,G406) has been reported as a bad WU.

Thanks for the report.

Sleepee -> Welcome to the F@H Forum Sleepee,
Please read this post to resolve your issue (viewtopic.php?f=19&t=16526#p164322).

I do not think it is a bad WU bollix47 is running -smp 15 and it is failing instantly bollix47 try running -smp 16 or 14.

[10:21:43] Project: 6050 (Run 0, Clone 119, Gen 406)
[10:21:43]
[10:21:43] Assembly optimizations on if available.
[10:21:43] Entering M.D.
[10:21:49] Mapping NT from 15 to 15
[10:41:29] CoreStatus = C0000417 (-1073740777)
[10:41:29] Client-core communications error: ERROR 0xc0000417
[10:41:29] Deleting current work unit & continuing...

[bollix47]

Can't rerun that WU as the client deleted it and moved on to something else. This client has completed well over 500 WUs (both smp2 regular and bigadv) with the -15 setting without a problem. Using the other core to feed a gtx 480 and I have tried both -14 and -16 but -15 works best.

[PantherX]

Thanks for the catch Grandpa_01, I have asked around and let's see what happens.

[bollix47]

Why would using 15 be a problem? I thought it was only prime numbers that caused a problem with the exceptions of 3 and maybe 5:

viewtopic.php?f=66&t=18549&p=186053#p186053

[Grandpa_01]

Why would using 15 be a problem? I thought it was only prime numbers that caused a problem with the exceptions of 3 and maybe 5:

viewtopic.php?f=66&t=18549&p=186053#p186053

In the link you provided there is a post from Kasson that explaines what the problem might be. 15 would be 15x1

by kasson » Mon May 09, 2011 3:48 pm

-smp 3 should be fine, -smp 5 probably ok. The problem is that Gromacs does a 2D decomposition based on factoring the number of threads you give it. So if you give it a number like 7 or 13, the best it can do is 7x1 or 13x1, whereas 12 can yield 4x3 and 8 can yield 4x2. The more thinly the system gets broken up, the more likely it is to fail. Hence easily factorable numbers are better...

[bruce]

[quote="Grandpa_01"]In the link you provided there is a post from Kasson that explaines what the problem might be. 15 would be 15x1

Not necessarily. Why not 15=5x3.

[Sleepee]

Thanks Bruce. I'm crunching on a different WU now.

To others: I don't think -smp x would be the problem. I was failing units with -smp 4, my maximum amount.

[Grandpa_01]

In the link you provided there is a post from Kasson that explaines what the problem might be. 15 would be 15x1

Not necessarily. Why not 15=5x3.

Must have been the drugs. I just jot out of surgery when I posted that.