new public beta client for linux, now 6.24

[pcowley]

Thanks, the alternate version fixed the problem.

Cheers
Pete

[Ivoshiee]

Still no 32-bit build. Why not?

[pcfxer]

I am unable to run the client on FreeBSD 7.1 - RELEASE 32-bit. I have run brandelf -t Linux fah6 and mpiexec (that latter was after my normal procedure of just running brandelf on the fah app), chmod 777 fah6 and mpiexec.

The error and command:

Code: Select all

%./fah6 -freeBSD -local -forceasm -verbosity 9
./fah6: Exec format error. Binary file not executable.


This is while I'm running two 6.02 single core clients so it isn't my method that is wrong, unless there is a flag change that I don't know about....

Regards,
Brodey

[smoking2000]

I am unable to run the client on FreeBSD 7.1 - RELEASE 32-bit. I have run brandelf -t Linux fah6 and mpiexec (that latter was after my normal procedure of just running brandelf on the fah app), chmod 777 fah6 and mpiexec.

The new v6.24 client is a 64 bit Linux binary and can therefor not be run on 32 bit Linux nor on 32 bit FreeBSD or OpenBSD using their Linux emulation layer.

[pcfxer]

Riiight, as per the post above mine asking about a 32-bit build. Gotcha'.

[kasson]

We can build 32-bit, but right now we're having issues with dependencies on libstdc++. So the build would be somewhat kernel-specific. We'd like to have a completely static binary. Hence the delay. (And we have a bunch of things to work on right now.)

SMP cores are still 64-bit only for Linux, also.

[pcfxer]

I'd love to help but I don't think the "cores" are open source are they?

Always appreciated, thanks for the response! Keep up the good work on the project!

[torswin]

Are there any news on me and Ivoshiee's problem with glibc and starting the client? I'm not authorised to see the thread Ivoshiee linked to, so if anyone could give me an update it would be nice

[kasson]

See my post above--no change from then. We'll post when we have something.

[dutchmm]

I downloaded the version to which your OP refers, and this is the result of running it on Mandriva 2009.0 (64 bit) with the 2.6.27.10-desktop-1mnb kernel.

[mike@uwless164 Download]$ ./fah6-2 -smp

Note: Please read the license agreement (fah6-2 -license). Further
use of this software requires that you have read and accepted this agreement.

2 cores detected


--- Opening Log file [February 21 07:48:08 UTC]


# Linux SMP Console Edition ###################################################
###############################################################################

Folding@Home Client Version 6.24beta

http://folding.stanford.edu

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

Launch directory: /home/mike/Download
Executable: ./fah6-2
Arguments: -smp

[07:48:08] - Ask before connecting: No
[07:48:08] - User name: Dutchmm (Team 31574)
[07:48:08] - User ID: 42905BD5336CB79C
[07:48:08] - Machine ID: 1
[07:48:08]
[07:48:08] Loaded queue successfully.
[07:48:08] - Preparing to get new work unit...
[07:48:08] + Attempting to get work packet
[07:48:08] - Connecting to assignment server
Floating point exception
[mike@uwless164 Download]$ uname -a
Linux uwless164 2.6.27.10-desktop-1mnb #1 SMP Thu Jan 29 11:16:18 EST 2009 x86_64 Intel(R) Core(TM)2 Duo CPU E6750 @ 2.66GHz GNU/Linux
[mike@uwless164 Download]$

Which should we be running?

[bollix47]

@dutchmm

Could you please restart the client using the -verbosity 9 switch in case there is more info related to what is happening?

Code: Select all

./fah6-2 -smp -verbosity 9

[dutchmm]

Could you please restart the client using the -verbosity 9 switch in case there is more info related to what is happening?

I came to the conclusion this would not finish until the early hours, and I could lose another day anyway. So here we have a WU, 69% completed with fah6_alt, now attempting to start with the current 6.24 beta

--- Opening Log file [February 21 21:25:01 UTC]


# Linux Console Edition #######################################################
###############################################################################

Folding@Home Client Version 6.24beta

http://folding.stanford.edu

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

Launch directory: /home/mike/Download
Executable: ./fah6-2
Arguments: -verbosity 9

[21:25:01] - Ask before connecting: No
[21:25:01] - User name: Dutchmm (Team 31574)
[21:25:01] - User ID: 42905BD5336CB79C
[21:25:01] - Machine ID: 1
[21:25:01]
[21:25:01] Loaded queue successfully.
[21:25:01] - Autosending finished units... [February 21 21:25:01 UTC]
[21:25:01] Trying to send all finished work units
[21:25:01] + No unsent completed units remaining.
[21:25:01] - Autosend completed
[21:25:01]
[21:25:01] + Processing work unit
[21:25:01] Core required: FahCore_a2.exe
[21:25:01] Core found.
[21:25:01] Working on queue slot 01 [February 21 21:25:01 UTC]
[21:25:01] + Working ...
[21:25:01] - Calling './FahCore_a2.exe -dir work/ -suffix 01 -checkpoint 15 -verbose -lifeline 30821 -version 624'

[21:25:01]
[21:25:01] *------------------------------*
[21:25:01] Folding@Home Gromacs SMP Core
[21:25:01] Version 2.04 (Thu Jan 29 16:43:57 PST 2009)
[21:25:01]
[21:25:01] Preparing to commence simulation
[21:25:01] - Looking at optimizations...
[21:25:01] - Files status OK
[21:25:02] - Expanded 4845807 -> 24003985 (decompressed 495.3 percent)
[21:25:02] Called DecompressByteArray: compressed_data_size=4845807 data_size=24003985, decompressed_data_size=24003985 diff=0
[21:25:02] - Digital signature verified
[21:25:02]
[21:25:02] Project: 2672 (Run 0, Clone 171, Gen 70)
[21:25:02]
[21:25:02] Assembly optimizations on if available.
[21:25:02] Entering M.D.
[21:25:08] Will resume from checkpoint file
NNODES=1, MYRANK=0, HOSTNAME=uwless164
G R O M A C S (-:

Groningen Machine for Chemical Simulation

VERSION 4.0.3_pre (-:


Written by David van der Spoel, Erik Lindahl, Berk Hess, and others.
Copyright (c) 1991-2000, University of Groningen, The Netherlands.
Copyright (c) 2001-2008, The GROMACS development team,
check out http://www.gromacs.org for more information.


mdrun (-:

Reading file work/wudata_01.tpr, VERSION 3.3.99_development_20070618 (single precision)
Note: tpx file_version 48, software version 58
starting mdrun '22890 system'
17750002 steps, 35500.0 ps (continuing from step 17500002, 35000.0 ps).

-------------------------------------------------------
Program mdrun, VERSION 4.0.3_pre
Source code file: md.c, line: 1107

Fatal error:
Checkpoint error on step 17672520

-------------------------------------------------------

Thanx for Using GROMACS - Have a Nice Day

Halting program mdrun

gcq#0: Thanx for Using GROMACS - Have a Nice Day

[unset]: aborting job:
application called MPI_Abort(MPI_COMM_WORLD, -1) - process 0
[21:25:10] Resuming from checkpoint
[21:25:10] fcSaveRestoreState: I/O failed dir=0, var=00007FD0A6A03010, varsize=1762308
[21:25:10] fcCheckPointResume: failure in call to fcSaveRestoreState() to restore state.
[21:25:10] CoreStatus = FF (255)
[21:25:10] Sending work to server
[21:25:10] Project: 2672 (Run 0, Clone 171, Gen 70)
[21:25:10] - Read packet limit of 540015616... Set to 524286976.
[21:25:10] - Error: Could not get length of results file work/wuresults_01.dat
[21:25:10] - Error: Could not read unit 01 file. Removing from queue.
[21:25:10] Trying to send all finished work units
[21:25:10] + No unsent completed units remaining.
[21:25:10] - Preparing to get new work unit...
[21:25:10] + Attempting to get work packet
[21:25:10] - Will indicate memory of 3966 MB
[21:25:10] - Connecting to assignment server
[21:25:10] Connecting to http://assign.stanford.edu:8080/
Floating point exception

[preet.to]

I am running 6.24 Beta on 64 Bit Fedora. I have now had 3 WU fail as follows:

Code: Select all

[17:44:37] Completed 2000000 out of 2000000 steps  (100 percent)
[17:44:37] Writing final coordinates.
[17:44:37] Past main M.D. loop
[17:44:41] CoreStatus = 0 (0)
[17:44:41] Sending work to server
[17:44:41] Project: 5101 (Run 0, Clone 154, Gen 70)
[17:44:41] - Error: Could not get length of results file work/wuresults_02.dat
[17:44:41] - Error: Could not read unit 02 file. Removing from queue.
[17:44:41] Trying to send all finished work units
[17:44:41] + No unsent completed units remaining.
[17:44:41] - Preparing to get new work unit...


I have run memtest86 and found no memory problems. I read up about corestatus = 0. That was not helpful. I ran qfix and tried to upload that unit. I have two now in the queue that are frozen in time.

What do I do? I have lost this machine for any production whatsoever.

[314159]

I have two now in the queue that are frozen in time

May we assume that you deleted the slots containing the "orphaned" results prior to running qfix?

[preet.to]

Yes absolutely. It was a delete, qfix, send X sequence that is failing each time.