Problem: HadTwo SMP Work units stopped early and delete

Scott66

Senior member
Feb 7, 2004
501
0
0
Hi People
I had two 2652 WU's get to 93% and then quit. It states it is going to send the completed work to Stanford but it comes up with a communications error. Here is an excerpt from my log.

[11:02:38] Warning: long 1-4 interactions
[11:02:38] Gromacs cannot continue further.
[11:02:38] Going to send back what have done.
[11:02:38] logfile size: 569184
[11:02:38] - Writing 569720 bytes of core data to disk...
[11:02:38] ... Done.
[11:02:38] - Failed to delete work/wudata_02.bed
[11:02:38] Warning: check for stray files
[11:04:38]
[11:04:38] Folding@home Core Shutdown: EARLY_UNIT_END
[11:04:38]
[11:04:38] Folding@home Core Shutdown: EARLY_UNIT_END
[11:04:40] CoreStatus = 7B (123)
[11:04:40] Client-core communications error: ERROR 0x7b
[11:04:40] Deleting current work unit & continuing...

Anybody else having this problem.
 

GLeeM

Elite Member
Apr 2, 2004
7,199
128
106
Were they the same WU, sometime if a WU gets deleted you get sent the same one again :confused:

I lost a few WUs too, the last went to 100% and deleted :(

I downloaded the client again and it uninstalled the old and reinstalled a new. I haven't had a problem yet, knock on wood :) Make sure you run the install.bat from the folding folder.
 

Scott66

Senior member
Feb 7, 2004
501
0
0
They were both 2652's, how do you tell if they are the same one? The third WU is one is as well.
 

GLeeM

Elite Member
Apr 2, 2004
7,199
128
106
The WU has a project number (Run #, Clone #, Gen #)

If it is the same WU all the numbers match. If the WUs are different at least one number will be different.

Here is an example:

Project: 2651 (Run 0, Clone 5, Gen 23)

You can see it in the log or on the console window.
 

Scott66

Senior member
Feb 7, 2004
501
0
0
Yes it is and it resent it a third time now. Well I guess I will be doing the uninstall/reinstall you talked about above
 

Alyx

Golden Member
Apr 28, 2007
1,181
0
0
From my understanding a higher percent of WUs fail with the SMP, something with how the units are made. Even if it fails Stanford learns some information from it, and if they notice the same work unit failing repeatedly I believe they pull it. I read a thread about this on the FAH forums a while back.