PDA

View Full Version : Issues with SMP client


David
04-13-09, 04:44 AM
I'm getting FILE_IO_ERROR messages. I was using the Deino client but switched to MPICH (I'm on XP 32bit) which dropped frame times by 10-15 mins.

However I still get the occasional error meaning I need to restart the client. I'm looking to eventually try SMP as a service on some office machines so I need this to work reliably. Any ideas what can cause this?

--- Opening Log file [April 13 06:32:49 UTC]


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

Folding@Home Client Version 6.23 Beta R1

http://folding.stanford.edu

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

Launch directory: C:\Program Files\Folding@Home Windows SMP Client V1.01
Executable: C:\Program Files\Folding@Home Windows SMP Client V1.01\Folding@home-Win32-x86.exe
Arguments: -smp -verbosity 9 -forceasm

Warning:
By using the -forceasm flag, you are overriding
safeguards in the program. If you did not intend to
do this, please restart the program without -forceasm.
If work units are not completing fully (and particularly
if your machine is overclocked), then please discontinue
use of the flag.

[06:32:49] - Ask before connecting: No
[06:32:49] - User name: penguininja (Team 32)
[06:32:49] - User ID: 7A11EE243E17837
[06:32:49] - Machine ID: 3
[06:32:49]
[06:32:49] Loaded queue successfully.
[06:32:49]
[06:32:49] - Autosending finished units... [April 13 06:32:49 UTC]
[06:32:49] + Processing work unit
[06:32:49] Trying to send all finished work units
[06:32:49] Work type a1 not eligible for variable processors[06:32:49] + No unsent completed units remaining.

[06:32:49] - Autosend completed
[06:32:49] Core required: FahCore_a1.exe
[06:32:49] Core found.
[06:32:49] Using generic mpiexec calls
[06:32:49] Working on queue slot 01 [April 13 06:32:49 UTC]
[06:32:49] + Working ...
[06:32:49] - Calling 'mpiexec -np 4 -channel auto -host 127.0.0.1 FahCore_a1.exe -dir work/ -suffix 01 -checkpoint 15 -forceasm -verbose -lifeline 3132 -version 623'

[06:32:50]
[06:32:50] *------------------------------*
[06:32:50] Folding@Home Gromacs SMP Core
[06:32:50] Version 1.74 (March 10, 2007)
[06:32:50]
[06:32:50] Preparing to commence simulation
[06:32:50] - Ensuring status. Please wait.
[06:33:07] - Assembly optimizations manually forced on.
[06:33:07] - Not checking prior termination.
[06:33:32] - Expanded 4803226 -> 24810145 (decompressed 516.5 percent)
[06:33:37]
[06:33:37] Project: 2665 (Run 2, Clone 154, Gen 103)
[06:33:37]
[06:33:38] Assembly optimizations on if available.
[06:33:38] Entering M.D.
[06:33:46] Could not open Sas file
[06:33:46]
[06:33:46] Folding@home Core Shutdown: FILE_IO_ERROR
[06:33:46] Finalizing output
[08:57:45] Killing all core threads
[08:57:45] Killing 4 cores
[08:57:45] Killing core 0
[08:57:45] Killing core 1
[08:57:45] Killing core 2
[08:57:45] Killing core 3

Folding@Home Client Shutdown at user request.
[08:57:45] ***** Got a SIGTERM signal (2)
[08:57:45] Killing all core threads
[08:57:45] Killing 4 cores
[08:57:45] Killing core 0
[08:57:45] Killing core 1
[08:57:45] Killing core 2
[08:57:45] Killing core 3

Folding@Home Client Shutdown.


--- Opening Log file [April 13 08:57:47 UTC]


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

Folding@Home Client Version 6.23 Beta R1

http://folding.stanford.edu

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

Launch directory: C:\Program Files\Folding@Home Windows SMP Client V1.01
Executable: C:\Program Files\Folding@Home Windows SMP Client V1.01\Folding@home-Win32-x86.exe
Arguments: -smp -verbosity 9 -forceasm

Warning:
By using the -forceasm flag, you are overriding
safeguards in the program. If you did not intend to
do this, please restart the program without -forceasm.
If work units are not completing fully (and particularly
if your machine is overclocked), then please discontinue
use of the flag.

[08:57:47] - Ask before connecting: No
[08:57:47] - User name: penguininja (Team 32)
[08:57:47] - User ID: 7A11EE243E17837
[08:57:47] - Machine ID: 3
[08:57:47]
[08:57:47] Loaded queue successfully.
[08:57:47]
[08:57:47] + Processing work unit
[08:57:47] Work type a1 not eligible for variable processors
[08:57:47] Core required: FahCore_a1.exe
[08:57:47] Core found.
[08:57:47] Using generic mpiexec calls
[08:57:47] - Autosending finished units... [April 13 08:57:47 UTC]
[08:57:47] Trying to send all finished work units
[08:57:47] + No unsent completed units remaining.
[08:57:47] - Autosend completed
[08:57:47] Working on queue slot 01 [April 13 08:57:47 UTC]
[08:57:47] + Working ...
[08:57:47] - Calling 'mpiexec -np 4 -channel auto -host 127.0.0.1 FahCore_a1.exe -dir work/ -suffix 01 -checkpoint 15 -forceasm -verbose -lifeline 312 -version 623'

[08:57:47]
[08:57:47] *------------------------------*
[08:57:47] Folding@Home Gromacs SMP Core
[08:57:47] Version 1.74 (March 10, 2007)
[08:57:47]
[08:57:47] Preparing to commence simulation
[08:57:47] - Ensuring status. Please wait.
[08:58:04] - Assembly optimizations manually forced on.
[08:58:04] - Not checking prior termination.
[08:58:30] - Expanded 4803226 -> 24810145 (decompressed 516.5 percent)
[08:58:31]
[08:58:31] Project: 2665 (Run 2, Clone 154, Gen 103)
[08:58:31]
[08:58:35] Assembly optimizations on if available.
[08:58:35] Entering M.D.
[08:58:41] Calling FAH init
[08:58:41] file
[08:58:41]
[08:58:41] Folding@home Core Shutdown: FILE_IO_ERROR
[08:58:41] Finalizing output
[08:58:52] Killing all core threads
[08:58:52] Killing 4 cores
[08:58:52] Killing core 0
[08:58:52] Killing core 1
[08:58:52] Killing core 2
[08:58:52] Killing core 3

Folding@Home Client Shutdown at user request.
[08:58:52] ***** Got a SIGTERM signal (2)
[08:58:52] Killing all core threads
[08:58:52] Killing 4 cores
[08:58:52] Killing core 0
[08:58:52] Killing core 1
[08:58:52] Killing core 2
[08:58:52] Killing core 3

Folding@Home Client Shutdown.


--- Opening Log file [April 13 08:58:54 UTC]


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

Folding@Home Client Version 6.23 Beta R1

http://folding.stanford.edu

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

Launch directory: C:\Program Files\Folding@Home Windows SMP Client V1.01
Executable: C:\Program Files\Folding@Home Windows SMP Client V1.01\Folding@home-Win32-x86.exe
Arguments: -smp -verbosity 9 -forceasm

Warning:
By using the -forceasm flag, you are overriding
safeguards in the program. If you did not intend to
do this, please restart the program without -forceasm.
If work units are not completing fully (and particularly
if your machine is overclocked), then please discontinue
use of the flag.

[08:58:54] - Ask before connecting: No
[08:58:54] - User name: penguininja (Team 32)
[08:58:54] - User ID: 7A11EE243E17837
[08:58:54] - Machine ID: 3
[08:58:54]
[08:58:55] Loaded queue successfully.
[08:58:55]
[08:58:55] + Processing work unit
[08:58:55] - Autosending finished units... [April 13 08:58:55 UTC]
[08:58:55] Work type a1 not eligible for variable processors
[08:58:55] Trying to send all finished work units
[08:58:55] Core required: FahCore_a1.exe
[08:58:55] + No unsent completed units remaining.
[08:58:55] - Autosend completed
[08:58:55] Core found.
[08:58:55] Using generic mpiexec calls
[08:58:55] Working on queue slot 01 [April 13 08:58:55 UTC]
[08:58:55] + Working ...
[08:58:55] - Calling 'mpiexec -np 4 -channel auto -host 127.0.0.1 FahCore_a1.exe -dir work/ -suffix 01 -checkpoint 15 -forceasm -verbose -lifeline 3392 -version 623'

[08:58:55]
[08:58:55] *------------------------------*
[08:58:55] Folding@Home Gromacs SMP Core
[08:58:55] Version 1.74 (March 10, 2007)
[08:58:55]
[08:58:55] Preparing to commence simulation
[08:58:55] - Ensuring status. Please wait.
[08:59:12] - Assembly optimizations manually forced on.
[08:59:12] - Not checking prior termination.
[08:59:37] - Expanded 4803226 -> 24810145 (decompressed 516.5 percent)
[08:59:38]
[08:59:38] Project: 2665 (Run 2, Clone 154, Gen 103)
[08:59:38]
[08:59:42] Assembly optimizations on if available.
[08:59:42] Entering M.D.
[08:59:48] Calling FAH init
[08:59:50] Read topology
[08:59:51] s
[08:59:51] Writing local files
[08:59:51] int)
[08:59:51] Read checkpoint
[08:59:51] Protein: HGG with glycosylations
[08:59:51] Writing local files
[08:59:51] Completed 133369 out of 250000 steps (53 percent)
[09:00:03] Extra SSE boost OK.
[09:14:53] Timered checkpoint triggered.
[09:29:36] Writing local files
[09:29:37] Completed 135000 out of 250000 steps (54 percent)

Jolly-Swagman
04-13-09, 05:22 AM
That is usually a hardware conflict error, Sometimes even just a simple Defrag of HDD will help or a PC Rebboot

If it was the Core it should also have a Core error code there too

Adak
04-13-09, 05:45 AM
David, did you perhaps install the MPICH without removing all the DEINO and FAH stuff from your rig's registry?

I got this error once when a HD data cable went bad, but far more commonly, I'll have swapped some directories around, or installed without removing the previous client, and that stuffs it.