Hi Jurica,
The likely explanation is that your simulation is unstable in some way and
is blowing up. Have you looked carefully at the NVT equilibration? Does it
look reasonable? Have you got vacuum bubbles formed?
Note chances are it is your restraints that are causing the problem. 50.0 is
WAY too high for anything other than minimization. A value of 1.0 to 5.0 is
probably more reasonable. Ideally you should turn the restraints off for
NPT.
What are you providing as the reference structure for the restraints? - NPT
can be VERY finicky with this. Box size changes can cause issues sometimes.
Try using the restart file from the NVT as your reference file from the
restraints. That is whatever you specify for -c should be the same as what
you specify for -ref.
Finally try setting nstlim=100,ntwx=1 and ntpr=1 and running 100 steps and
then looking carefully at any output you get in mdout and mdcrd.
Good luck,
Ross
> -----Original Message-----
> From: Jurica Novak [mailto:jnovak.irb.hr]
> Sent: Wednesday, August 17, 2011 9:23 AM
> To: AMBER Mailing List
> Subject: [AMBER] problem with NPT equilibration stage
>
> ---------- Forwarded message ----------
> Dear Amber users,
>
> after equilibrating a cyclohexane box, I have put inside a cromophore
> (78
> atoms) and did constrained minimization and NVT equilibration. Now I
> want to
> do NPT equilibration, starting from restart file representing the last
> point
> of NVT simulation. Here is my input file npt.in:
> md npt equilibration
> &cntrl
> imin = 0, irest = 1, ntx = 7,
> ntb = 2, pres0 = 1.0, ntp = 1,
> taup = 2.0,
> cut = 10.0, ntr = 1,
> ntc = 2, ntf = 2,
> tempi = 300.0, temp0 = 300.0,
> ntt = 3, gamma_ln = 1.0,
> nstlim = 100000, dt = 0.001,
> ntpr = 1000, ntwx = 1000, ntwr = 1000, ntwv = 1000
> /
> Keep cromophore restrainted
> 50.0
> RES 1 3
> END
> END
>
> Error message that is printed is:
>
> ***** Processor 7
> ***** System must be very inhomogeneous.
> ***** Readjusting recip sizes.
> In this slab, Atoms found: 14064 Allocated: 3472
>
> [eniac:17614] *** Process received signal ***
> [eniac:17614] Signal: Segmentation fault (11)
> [eniac:17614] Signal code: Address not mapped (1)
> [eniac:17614] Failing at address: 0x7f06e4f29aa8
> [eniac:17614] [ 0] /lib/libpthread.so.0 [0x7f16fc153a80]
> [eniac:17614] [ 1] sander.MPI(nb_adjust_+0x4ea) [0x520e3a]
> [eniac:17614] [ 2] sander.MPI(ewald_force_+0x12e6) [0x523976]
> [eniac:17614] [ 3] sander.MPI(force_+0x3004) [0x6df404]
> [eniac:17614] [ 4] sander.MPI(runmd_+0x1671) [0x4f8c61]
> [eniac:17614] [ 5] sander.MPI(sander_+0x61c5) [0x4c1f39]
> [eniac:17614] [ 6] sander.MPI(MAIN__+0xee7) [0x4bbce3]
> [eniac:17614] [ 7] sander.MPI(main+0x2c) [0x965a2c]
> [eniac:17614] [ 8] /lib/libc.so.6(__libc_start_main+0xe6)
> [0x7f16fbe101a6]
> [eniac:17614] [ 9] sander.MPI [0x4448a9]
> [eniac:17614] *** End of error message ***
> .
> .
> .
> [eniac:17606] [0,0,0] ORTE_ERROR_LOG: Timeout in file
> base/pls_base_orted_cmds.c at line 275
> [eniac:17606] [0,0,0] ORTE_ERROR_LOG: Timeout in file pls_rsh_module.c
> at
> line 1166
> [eniac:17606] [0,0,0] ORTE_ERROR_LOG: Timeout in file errmgr_hnp.c at
> line
> 90
> mpirun noticed that job rank 1 with PID 17610 on node eniac exited on
> signal
> 11 (Segmentation fault).
> 5 additional processes aborted (not shown)
> [eniac:17606] [0,0,0] ORTE_ERROR_LOG: Timeout in file
> base/pls_base_orted_cmds.c at line 188
> [eniac:17606] [0,0,0] ORTE_ERROR_LOG: Timeout in file pls_rsh_module.c
> at
> line 1198
> -----------------------------------------------------------------------
> ---
> mpirun was unable to cleanly terminate the daemons for this job.
> Returned
> value Timeout instead of ORTE_SUCCESS.
> -----------------------------------------------------------------------
> ---
>
> If a simulation is started on one processor only, 'Segmentation fault'
> message is thrown. I get the same error messages If I try to start NPT
> equilibration prior minimization (from input.inpcrd file generated by
> xleap), or after minimization (5000 steps, without NVT equilibration).
> What
> I have done wrong?
>
> Thanks in advance,
> Jura
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Wed Aug 17 2011 - 10:00:02 PDT
