Project 6097 Run 0 Clone 36 Gen 192

[HutchinsonJC]

I'm running FAH6.34 win32 SMP and having problems with this work unit. I can't seem to skip it by deleting the queue.dat and the contents in the 'work' folder. Basically it just keeps cycling through giving a Client-core communications error: ERROR 0xc0000005. I looked it up, and it appears it's some kind of memory error, but I have strong doubts about that as I've been able to game, and encode very large video files with no problem. I've also restarted this machine hoping to get beyond this supposed memory error, but this wu always crashes before it even gets to 1 out of 500,000 steps. Google brings me this one from a few days ago, but it appears to be a different error code. http://foldingforum.org/viewtopic.php?f=19&t=21188&start=0 Any ideas what I can do to just skip this work unit or get it moving on again?

 

[VirtualLarry]

Your overclock is unstable. What is stable for gaming, and video encoding is, unsurprisingly, not stable for F@H.

 

[HutchinsonJC]

I'm pretty sure it is stable. I've run IBT, prime, I have gamed while encoding x264 while FAH and never had an issue until this wu. It won't even get passed 1 of 500,000 steps before crashing and it's only the FahCore_a3.exe that's crashing. Not the whole system.

I'm currently torrenting, FAH GPU3 on a gtx 580, running handbrake (x264), and gaming and I have no issues with anything else at all. How can I skip this work unit. Someone else is having an infinite loop crash on the same wu if you clicked the link I found on google just some few days ago.

Edit: If you read that thread I linked, there's actually two different people complaining about the wu. The OP on April 4th, then a mod posts on April 5, then someone else complains of this same WU on the 7th of April. In all 3 cases, the thing that is similar is that we all crash on this wu over and over again.

The first guy gets:
Client-core communications error: ERROR 0x8b

The second guy gets this over and over:
[05:53:53] Completed 0 out of 500000 steps (0%)
[05:53:56] CoreStatus = FF (255)
[05:53:56] Sending work to server
[05:53:56] Project: 6097 (Run 0, Clone 36, Gen 192)
[05:53:56] - Error: Could not get length of results file work/wuresults_01.dat
[05:53:56] - Error: Could not read unit 01 file. Removing from queue.

and I get this over and over:
[18:20:41] Completed 0 out of 500000 steps (0%)
[18:25:05] CoreStatus = C0000005 (-1073741819)
[18:25:05] Client-core communications error: ERROR 0xc0000005 [18:25:05] Deleting current work unit & continuing...
[18:25:43] - Preparing to get new work unit...

Edit Again:
I was able to get a different work unit by
1) deleting everything in "C:\Program Files (x86)\FAH CPU" except client.cfg and FAH6.34-win32-SMP.exe
2) creating a shortcut with the -config flag and running it
3) forcing -smp 2, turning off -bigadv
4) then running the original .exe

Currently crunching a Project 8024 run 204, clone 21, gen 6. I've already interrupted this wu and changed my settings back to what they were, hopefully when this wu finishes I'll be back on my way with -bigadv. Will Update in a few, won't take long to crunch out this 8024.

 

[VirtualLarry]

I apologize, perhaps it is the WU, if it is erroring out without completing any steps. I used to have a few like that occasionally.

 

[biodoc]

Glad you solved the problem HutchinsonJC. :thumbsup:

 

[HutchinsonJC]

Well, it doesn't appear it's actually solved. When I turned off -bigadv and deleted everything in the folder, it started to crunch a wu from project 8024. I halted it midway through, changed my settings back, and then restarted it. It finished that 8024 and another 8024 after that, but then the first -bigadv it tried to take was from project 6097. And not just any wu from 6097, but Run 0, Clone 36, Gen 192. So it's back to infinite loop of crashes.

Turning -bigadv off again, and gonna leave it off for a while. Don't know what else to do.

 

[theAnimal]

6097 is not a bigadv WU, regular SMP. It would probably be best to report the bad WU at foldingforum.org in the Issues with a specific WU sub-forum.

 

[HutchinsonJC]

Done. You can read the response from the mod from the same link in my original post.

 

[biodoc]

Excellent.