Re: why this error?

From: Jim Phillips (jim_at_ks.uiuc.edu)
Date: Mon Aug 21 2006 - 14:55:04 CDT

Hi,

First, you're using 2.6b1 or 2.6b2, but 2.6 hasn't been released yet.
It's also useful to know what platform you're running on and if this is a
released binary or one you built yourself. Even better, just include the
initial and final portions of the log file and the command used to run.

This is the code that generates the error (in src/common.C):

void NAMD_backup_file(const char *filename, const char *extension)
{
   struct stat sbuf;
   if (stat(filename, &sbuf) == 0) {
     if ( ! extension ) extension = ".BAK";
     char *backup = new char[strlen(filename)+strlen(extension)+1];
     strcpy(backup, filename);
     strcat(backup, extension);
#if defined(WIN32) && !defined(__CYGWIN__)
     remove(backup);
#endif
     if ( rename(filename,backup) )
     {
       char errmsg[256];

       sprintf(errmsg, "Error on renaming file %s to %s",filename,backup);
       NAMD_err(errmsg);
     }
     delete [] backup;
   }
}

So, a file that existed ten lines earlier (stat succeeded) is no longer
there. Either you have a very messed up filesystem, or you're actually
running multiple independed runs instead of a single run and they're
stepping on each other. This could happen, for example, if you tried to
use mpirun on a non-MPI NAMD binary. (Most of the released binaries use
the Charm++ network interface rather than MPI.) Check the log file to see
if you're running on as many processors as you expect.

-Jim

On Mon, 21 Aug 2006, Hamsa Priya Mohana sundaram wrote:

> Hi
>
> I am using namd 2.6. I was trying to do NVT simulation for 5 amino acid
> helix in a water box of 900 water molecules
>
> After 100 ps simulation i got the following error
>
> FATAL ERROR: Error on renaming file prod_restart.xsc to
> prod_restart.xsc.old: No such file or directory
>
> Once again i did the simulation from the start. This time i got the
> following error after 892 ps
>
> WRITING EXTENDED SYSTEM TO RESTART FILE AT STEP 892000
> WRITING COORDINATES TO DCD FILE AT STEP 892000
> WRITING COORDINATES TO RESTART FILE AT STEP 892000
> FATAL ERROR: Error on renaming file prod_restart.coor to
> prod_restart.coor.old: No such file or directory
> Stack Traceback:
> [0] CmiAbort+0x35 [0x73428d]
> [1] _Z8NAMD_errPKc+0xa4 [0x49b014]
> [2] _Z16NAMD_backup_filePKcS0_+0x106 [0x49aaae]
> [3] _ZN6Output26output_restart_coordinatesEP6Vectorii+0x13d [0x61aad5]
> [4] _ZN6Output10coordinateEiiP6VectorP11FloatVectorR7Lattice+0x11f
> [0x61a7e7]
> [5] _ZN16CollectionMaster16receivePositionsEP16CollectVectorMsg+0x176
> [0x4a81e6]
> [6]
> _ZN24CkIndex_CollectionMaster39_call_receivePositions_CollectVectorMsgEPvP16CollectionMaster+0x12
> [0x4a806a]
> [7] CkDeliverMessageFree+0x30 [0x6cfbb8]
> [8] _Z15_processHandlerPvP11CkCoreState+0x4a0 [0x6d22a0]
> [9] CmiHandleMessage+0x26 [0x738ace]
> [10] CsdScheduleForever+0x4b [0x738c2b]
> [11] CsdScheduler+0x1c [0x73a2ac]
> [12] _ZN7BackEnd7suspendEv+0xe [0x4a12b6]
> [13] _ZN9ScriptTcl3runEPc+0x15b [0x656153]
> [14] main+0x222 [0x49d862]
> [15] __libc_start_main+0xdb [0x33f741c4bb]
> [16] _ZStlsISt11char_traitsIcEERSt13basic_ostreamIcT_ES5_c+0x52
> [0x49a22a]
>
>
> Could anyone help me know the reason for this error and how to rectify it?
>
> Thanks
> Hamsa
>

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:42:29 CST