AMBER: segmentation fault occured doing MD

From: priya priya <priyaanand_27.yahoo.co.in>
Date: Fri, 26 Oct 2007 12:51:31 +0100 (BST)

Hi

I am trying to MD for a simple peptide, with all usual amino acids. Minimization in implicit solvent was running fine, but when heated the system, the system blow off, with velocities exceeding the limit, that it is not written in rst file, ******* .
Part of rst file:

NASP
   59 0.5000000E+01
1942.45790201908.7496988************1943.17131281909.4646415************
1943.17127991909.4646726************1943.17129151909.4646601************
1908.32378701894.50209451740.65808271909.08865581895.27863631740.6657391
1

 In the output file it says that vlimit exceeded for step 416(and so on); vmax = 20.1241.
output file is like:
 
 NSTEP = 7550 TIME(PS) = 0.755 TEMP(K) =********* PRESS = 0.0
  Etot = 491282380.9662 EKtot = 181063.2625 EPtot = 491101317.7037
  BOND = 18.0335 ANGLE = 1584011.0917 DIHED = 0.0000
  1-4 NB = 0.0000 1-4 EEL = 0.0000 VDWAALS = 5887.1084
  EELEC = 168.2903 EGB = 0.0000 RESTRAINT = 0.0000
  DV/DL = 198.3142
 vlimit exceeded for step 7551(and so on); vmax = 20.1241.

I have tried to decrease the dt =0.0005 to 0.001 also, but still of no use.

Input file was as follows:
Stage 1 heating of test 0 to 50K
 &cntrl
  imin=0, irest=0, ntx=1,
  nstlim=10000, dt=0.0005,
  ntc=2, ntf=2,
  ntt=1, tautp=1.0,
  tempi=0.0, temp0=50.0,
  ntpr=50, ntwx=50,
  ntb=0, igb=1,
  cut=999.,rgbmax=999.
 /


the error in the error file is:
forrtl: severe (174): SIGSEGV, segmentation fault occurred
Image PC Routine Line Source
sander.MPI 00000000004A2F42 Unknown Unknown Unknown

Stack trace terminated abnormally.
forrtl: error (78): process killed (SIGTERM)
PSIlogger: Child with rank 1 exited with status 174.
PSIlogger: Child with rank 0 exited with status 1.

I have seen in the archive about this error, but the suggestions are not working.
Would anybody suggest what is wrong.


Regards
Priya

       
---------------------------------
 Save all your chat conversations. Find them online.
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber.scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo.scripps.edu
Received on Sun Oct 28 2007 - 06:07:36 PDT
Custom Search