Is anyone else having issues with 6.22 -smp ?

S

smileyscout

Gawd
Joined
Mar 29, 2005
Messages
631
I've been running 5.91 stable for quite a while now but 6.22 -smp is not playing nice at all.

Code: 
--- Opening Log file [August 2 19:57:48 UTC] 


# Windows SMP Console Edition #################################################
###############################################################################

                       Folding@Home Client Version 6.22 SMP Beta2

                          http://folding.stanford.edu

###############################################################################
###############################################################################

Launch directory: C:\Program Files (x86)\Folding@Home Windows SMP Client V1.01
Executable: C:\Program Files (x86)\Folding@Home Windows SMP Client V1.01\[email protected]
Arguments: -smp 

[19:57:48] - Ask before connecting: No
[19:57:48] - User name: smileyscout (Team 33)
[19:57:48] - User ID: 6A1ECBA01EBAB75C
[19:57:48] - Machine ID: 1
[19:57:48] 
[19:57:48] Loaded queue successfully.
[19:57:48] 
[19:57:48] + Processing work unit
[19:57:48] Work type 82 not eligible for variable processors
[19:57:48] Core required: FahCore_82.exe
[19:57:48] Core found.
[19:57:48] Using generic mpiexec calls
[19:57:48] Working on queue slot 01 [August 2 19:57:48 UTC]
[19:57:48] + Working ...
[19:57:48] 
[19:57:48] *------------------------------*
[19:57:48] Folding@Home PMD Core
[19:57:48] Version 1.03 (September 7, 2005)
[19:57:48] 
[19:57:48] Preparing to commence simulation
[19:57:48] - Ensuring status. Please wait.
[19:58:05] - Looking at optimizations...
[19:58:05] - Working with standard loops on this execution.
[19:58:05] - Previous termination of core was improper.
[19:58:05] - Going to use standard loops.
[19:58:05] - Files status OK
[19:58:05] - Expanded 12588 -> 77217 (decompressed 613.4 percent)
[19:58:05] 
[19:58:05] Project: 4593 (Run 45, Clone 8, Gen 22)
[19:58:05] 
[19:58:05] Error: Could not write local file.  Exiting.
[19:58:11] - Shutting down core
[19:58:11] 
[19:58:11] Folding@home Core Shutdown: FILE_IO_ERROR

Folding@Home Client Shutdown at user request.

Folding@Home Client Shutdown.


--- Opening Log file [August 2 23:32:08 UTC] 


# Windows SMP Console Edition #################################################
###############################################################################

                       Folding@Home Client Version 6.22 SMP Beta2

                          http://folding.stanford.edu

###############################################################################
###############################################################################

Launch directory: C:\Program Files (x86)\Folding@Home Windows SMP Client V1.01
Executable: C:\Program Files (x86)\Folding@Home Windows SMP Client V1.01\[email protected]
Arguments: -smp 

[23:32:08] - Ask before connecting: No
[23:32:08] - User name: smileyscout (Team 33)
[23:32:08] - User ID: 6A1ECBA01EBAB75C
[23:32:08] - Machine ID: 1
[23:32:08] 
[23:32:08] Loaded queue successfully.
[23:32:08] 
[23:32:08] + Processing work unit
[23:32:08] Work type a1 not eligible for variable processors
[23:32:08] Core required: FahCore_a1.exe
[23:32:08] Core found.
[23:32:08] Using generic mpiexec calls
[23:32:08] Working on queue slot 02 [August 2 23:32:08 UTC]
[23:32:08] + Working ...
[23:32:09] 
[23:32:09] *------------------------------*
[23:32:09] Folding@Home Gromacs SMP Core
[23:32:09] Version 1.74 (March 10, 2007)
[23:32:09] 
[23:32:09] Preparing to commence simulation
[23:32:09] - Ensuring status. Please wait.
[23:32:26] - Looking at optimizations...
[23:32:26] - Working with standard loops on this execution.
[23:32:26] - Previous termination of core was improper.
[23:32:26] - Going to use standard loops.
[23:32:26] - Files status OK
[23:34:26] 
[23:34:26] Folding@home Core Shutdown: MISSING_WORK_FILES
[23:34:26] Finalizing output
[23:34:29] CoreStatus = 1 (1)
[23:34:29] Client-core communications error: ERROR 0x1
[23:34:29] This is a sign of more serious problems, shutting down.


--- Opening Log file [August 2 23:56:44 UTC] 


# Windows SMP Console Edition #################################################
###############################################################################

                       Folding@Home Client Version 6.22 SMP Beta2

                          http://folding.stanford.edu

###############################################################################
###############################################################################

Launch directory: C:\Program Files (x86)\Folding@Home Windows SMP Client V1.01
Executable: C:\Program Files (x86)\Folding@Home Windows SMP Client V1.01\[email protected]
Arguments: -smp 

[23:56:44] - Ask before connecting: No
[23:56:44] - User name: smileyscout (Team 33)
[23:56:44] - User ID: 6A1ECBA01EBAB75C
[23:56:44] - Machine ID: 1
[23:56:44] 
[23:56:44] Loaded queue successfully.
[23:56:44] 
[23:56:44] + Processing work unit
[23:56:44] Work type a1 not eligible for variable processors
[23:56:44] Core required: FahCore_a1.exe
[23:56:44] Core found.
[23:56:44] Using generic mpiexec calls
[23:56:44] Working on queue slot 02 [August 2 23:56:44 UTC]
[23:56:44] + Working ...
[23:56:44] 
[23:56:44] *------------------------------*
[23:56:44] Folding@Home Gromacs SMP Core
[23:56:44] Version 1.74 (March 10, 2007)
[23:56:44] 
[23:56:44] Preparing to commence simulation
[23:56:44] - Ensuring status. Please wait.
[23:57:01] - Looking at optimizations...
[23:57:01] - Working with standard loops on this execution.
[23:57:01] - Previous termination of core was improper.
[23:57:01] - Going to use standard loops.
[23:57:01] - Files status OK
[23:57:01] 
[23:57:01] Folding@home Core Shutdown: MISSING_WORK_FILES
[23:57:01] Finalizing output
[23:59:04] CoreStatus = 1 (1)
[23:59:04] Client-core communications error: ERROR 0x1
[23:59:04] This is a sign of more serious problems, shutting down.


--- Opening Log file [August 3 04:44:25 UTC] 


# Windows SMP Console Edition #################################################
###############################################################################

                       Folding@Home Client Version 6.22 SMP Beta2

                          http://folding.stanford.edu

###############################################################################
###############################################################################

Launch directory: C:\Program Files (x86)\Folding@Home Windows SMP Client V1.01
Executable: C:\Program Files (x86)\Folding@Home Windows SMP Client V1.01\[email protected]
Arguments: -smp 

[04:44:25] - Ask before connecting: No
[04:44:25] - User name: smileyscout (Team 33)
[04:44:25] - User ID: 6A1ECBA01EBAB75C
[04:44:25] - Machine ID: 1
[04:44:25] 
[04:44:25] Loaded queue successfully.
[04:44:25] 
[04:44:25] + Processing work unit
[04:44:25] Work type a1 not eligible for variable processors
[04:44:25] Core required: FahCore_a1.exe
[04:44:25] Core found.
[04:44:25] Using generic mpiexec calls
[04:44:25] Working on queue slot 02 [August 3 04:44:25 UTC]
[04:44:25] + Working ...
[04:44:25] 
[04:44:25] *------------------------------*
[04:44:25] Folding@Home Gromacs SMP Core
[04:44:25] Version 1.74 (March 10, 2007)
[04:44:25] 
[04:44:25] Preparing to commence simulation
[04:44:25] - Ensuring status. Please wait.
[04:44:42] - Looking at optimizations...
[04:44:42] - Working with standard loops on this execution.
[04:44:42] Examination of work files indicates 8 consecutive improper terminations of core.
[04:44:43] 
[04:44:43] Folding@home Core Shutdown: MISSING_WORK_FILES
[04:44:43] Finalizing output
[04:46:45] CoreStatus = 1 (1)
[04:46:45] Client-core communications error: ERROR 0x1
[04:46:45] This is a sign of more serious problems, shutting down.


--- Opening Log file [August 3 05:16:47 UTC] 


# Windows SMP Console Edition #################################################
###############################################################################

                       Folding@Home Client Version 6.22 SMP Beta2

                          http://folding.stanford.edu

###############################################################################
###############################################################################

Launch directory: C:\Program Files (x86)\Folding@Home Windows SMP Client V1.01
Executable: C:\Program Files (x86)\Folding@Home Windows SMP Client V1.01\[email protected]
Arguments: -smp 

[05:16:47] - Ask before connecting: No
[05:16:47] - User name: smileyscout (Team 33)
[05:16:47] - User ID: 6A1ECBA01EBAB75C
[05:16:47] - Machine ID: 1
[05:16:47] 
[05:16:47] Loaded queue successfully.
[05:16:47] 
[05:16:47] + Processing work unit
[05:16:47] Work type a1 not eligible for variable processors
[05:16:47] Core required: FahCore_a1.exe
[05:16:47] Core not found.
[05:16:47] - Core is not present or corrupted.
[05:16:47] - Attempting to download new core...
[05:16:47] + Downloading new core: FahCore_a1.exe
[05:16:48] + 10240 bytes downloaded
snip...
[05:16:50] + 789667 bytes downloaded
[05:16:50] Verifying core Core_a1.fah...
[05:16:50] Signature is VALID
[05:16:50] 
[05:16:50] Trying to unzip core FahCore_a1.exe
[05:16:50] Decompressed FahCore_a1.exe (2035712 bytes) successfully
[05:16:55] + Core successfully engaged
[05:17:00] 
[05:17:00] + Processing work unit
[05:17:00] Work type a1 not eligible for variable processors
[05:17:00] Core required: FahCore_a1.exe
[05:17:00] Core found.
[05:17:00] Using generic mpiexec calls
[05:17:00] Working on queue slot 02 [August 3 05:17:00 UTC]
[05:17:00] + Working ...
[05:17:00] 
[05:17:00] *------------------------------*
[05:17:00] Folding@Home Gromacs SMP Core
[05:17:00] Version 1.74 (March 10, 2007)
[05:17:00] 
[05:17:00] Preparing to commence simulation
[05:17:00] - Ensuring status. Please wait.
[05:17:17] - Looking at optimizations...
[05:17:17] - Working with standard loops on this execution.
[05:17:17] - Previous termination of core was improper.
[05:17:18] - Going to use standard loops.
[05:17:18] - Files status OK
[05:19:18] 
[05:19:18] Folding@home Core Shutdown: MISSING_WORK_FILES
[05:19:18] Finalizing output
[05:19:20] CoreStatus = 1 (1)
[05:19:20] Client-core communications error: ERROR 0x1
[05:19:20] This is a sign of more serious problems, shutting down.


--- Opening Log file [August 3 06:25:39 UTC] 


# Windows SMP Console Edition #################################################
###############################################################################

                       Folding@Home Client Version 6.22 SMP Beta2

                          http://folding.stanford.edu

###############################################################################
###############################################################################

Launch directory: C:\Program Files (x86)\Folding@Home Windows SMP Client V1.01
Executable: C:\Program Files (x86)\Folding@Home Windows SMP Client V1.01\[email protected]
Arguments: -smp 

[06:25:39] - Ask before connecting: No
[06:25:39] - User name: smileyscout (Team 33)
[06:25:39] - User ID: 6A1ECBA01EBAB75C
[06:25:39] - Machine ID: 1
[06:25:39] 
[06:25:39] Could not open work queue, generating new queue...
[06:25:39] - Preparing to get new work unit...
[06:25:39] + Attempting to get work packet
[06:25:39] - Connecting to assignment server
[06:25:40] - Successful: assigned to (171.64.65.63).
[06:25:40] + News From Folding@Home: Welcome to Folding@Home
[06:25:40] Loaded queue successfully.
[06:25:43] + Closed connections
[06:25:43] 
[06:25:43] + Processing work unit
[06:25:43] Work type a1 not eligible for variable processors
[06:25:43] Core required: FahCore_a1.exe
[06:25:43] Core not found.
[06:25:43] - Core is not present or corrupted.
[06:25:43] - Attempting to download new core...
[06:25:43] + Downloading new core: FahCore_a1.exe
[06:25:44] + 10240 bytes downloaded
snip...
[06:25:45] + 789667 bytes downloaded
[06:25:45] Verifying core Core_a1.fah...
[06:25:45] Signature is VALID
[06:25:45] 
[06:25:45] Trying to unzip core FahCore_a1.exe
[06:25:45] Decompressed FahCore_a1.exe (2035712 bytes) successfully
[06:25:50] + Core successfully engaged
[06:25:56] 
[06:25:56] + Processing work unit
[06:25:56] Work type a1 not eligible for variable processors
[06:25:56] Core required: FahCore_a1.exe
[06:25:56] Core found.
[06:25:56] Using generic mpiexec calls
[06:25:56] Working on queue slot 01 [August 3 06:25:56 UTC]
[06:25:56] + Working ...
[06:25:57] 
[06:25:57] *------------------------------*
[06:25:57] Folding@Home Gromacs SMP Core
[06:25:57] Version 1.74 (March 10, 2007)
[06:25:57] 
[06:25:57] Preparing to commence simulation
[06:25:57] - Ensuring status. Please wait.
[06:25:57] - Starting from initial work packet
[06:25:57] 
[06:25:57] Project: 3062 (Run 1, Clone 148, Gen 33)
[06:25:57] 
[06:25:57] Assembly optimizations on if available.
[06:25:57] Entering M.D.
[06:26:14]  percent)
[06:26:14] - Starting from initial work packet
[06:26:14] 
[06:26:14] Project: 3062 (Run 1, Clone 148, Gen 33)
[06:26:14] 
[06:26:14] Entering M.D.
[06:26:20] Rejecting checkpoint
[06:26:21] a SSE boost OK.
[06:26:21] ambda5_99sbExtra SSE boost OK.
[06:26:21] 
[06:26:21] Extra SSE boost OK.
[06:26:21] Writing local files
[06:26:21] Completed 0 out of 5000000 steps  (0 percent)
[06:41:18] Writing local files
[06:41:18] Completed 50000 out of 5000000 steps  (1 percent)
[06:56:25] Writing local files
[06:56:25] Completed 100000 out of 5000000 steps  (2 percent)
[07:09:54] Writing local files
[07:09:54] Completed 150000 out of 5000000 steps  (3 percent)
[07:23:22] Writing local files
[07:23:22] Completed 200000 out of 5000000 steps  (4 percent)
[07:36:53] Writing local files
[07:36:53] Completed 250000 out of 5000000 steps  (5 percent)
[07:52:01] Writing local files
[07:52:01] Completed 300000 out of 5000000 steps  (6 percent)
[08:05:35] Writing local files
[08:05:36] Completed 350000 out of 5000000 steps  (7 percent)
[08:19:08] Writing local files
[08:19:08] Completed 400000 out of 5000000 steps  (8 percent)
[08:34:09] Writing local files
[08:34:09] Completed 450000 out of 5000000 steps  (9 percent)
[08:47:58] Writing local files
[08:47:58] Completed 500000 out of 5000000 steps  (10 percent)
[08:56:55] Warning:  long 1-4 interactions
[08:56:55] Gromacs cannot continue further.
[08:56:55] Going to send back what have done.
[08:56:55] logfile size: 21803
[08:56:55] - Writing 22339 bytes of core data to disk...
[08:56:55]   ... Done.
[08:56:55] - Failed to delete work/wudata_01.xtc
[08:56:55] No C.P. to delete.
[08:56:55] - Failed to delete work/wudata_01.sas
[08:56:55] - Failed to delete work/wudata_01.goe
[08:56:55] - Failed to delete work/wudata_01.pdo
[08:56:55] - Failed to delete work/wudata_01.xvg
[08:56:55] Warning:  check for stray files
[08:56:55] 
[08:56:55] Folding@home Core Shutdown: EARLY_UNIT_END
[08:56:55] 
[08:56:55] Folding@home Core Shutdown: EARLY_UNIT_END
[08:56:59] CoreStatus = 7B (123)
[08:56:59] Client-core communications error: ERROR 0x7b
[08:56:59] This is a sign of more serious problems, shutting down.

http://foldingforum.org/viewtopic.php?f=46&t=4494&p=45199#p45199

It doesn't look like I'm the only one having issues either. Any ideas why 5.91 would run fine and 6.22 -smp explode like this? I've been running 2x8800 GT GPU clients stable for the last twelve hours plus. My Q6600 isn't overclocked either.

 
To fix that, delete queue.dat and the work folder. I guess the 6.22 client botched the cleanup routine so it cannot clean the files properly, hence this error.

 
Well make sure you do the install also... both times I've attempted to install 6.22 without reinstalling the MPI service it has had critical errors... nto sure if this was relic's problem also, but that solved it for me. I now have 3 boxes running on the new 6.22 and 6.20 combo without issue.

 
I forgot to install MPI as well and got the same core communication error. After installing MPI again I've had no issues for 12 hours or so.

 
they really screwed this up.. i'm having the same issues.

stable fine for 6 months, no errors....

i switch to 6.22 with the -smp flag, and now I get this error every 2-3 days.

"Folding@home Core Shutdown: MISSING_WORK_FILES"

"CoreStatus = 1 (1)"

"Folding@home has run into a serious error running the core. and will shutdown."
Click to expand...
 
prosper0 said:
they really screwed this up.. i'm having the same issues.

stable fine for 6 months, no errors....

i switch to 6.22 with the -smp flag, and now I get this error every 2-3 days.
Click to expand...

What has worked for me is deleting the Work directory and the queue.dat and the Fahcore file... that usually does it...

Still odd it has these issues....

I do need to upgrade to R3... but honestly some 20 clients... I am not looking forward to upgrading atm.



 
The R3 upgrade isn't that bad. I just changed 3 and it took a total of about 5 minutes. All I do is take a snapshot backup of the folders that will be affected. Stop the client. Check task manager and manually close any stray F@H cores still running (my experience is that there will always be some). Restore the backup. The SMPD.exe file will not restore, this is not an issue, just skip it. Download the client.exe file if you haven't already. Create a new shortcut. Start the client up and walla, you're done. Even with 20 clients (PITA, I'm sure), it shouldn't take more than 30 minutes.

 
You must log in or register to reply here.
Back
Top