namd 2.10 crashes when using more than 1 CPU for FEP simulation

From: Tien Huynh (gdesktop_at_us.ibm.com)
Date: Tue Oct 27 2015 - 21:58:14 CDT

• Next message: Nikhil Maroli: "Including two parameters for protein-ligand simulation"
• Previous message: Giacomo Fiorin: "Re: Biasing potential of a Metadynamics Simulation"
• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]

Hi,
namd version 2.10 and the nightly built version always cause "Segmentation
fault' without generating a core dumped file when using more than 1 CPU
for FEP simulations.
Thanks!

OPENING FEP ENERGY OUTPUT FILE
OPENING EXTENDED SYSTEM TRAJECTORY FILE
LDB: ============= START OF LOAD BALANCING ============== 5.34365
LDB: ============== END OF LOAD BALANCING =============== 5.34373
Info: useSync: 0 useProxySync: 0
LDB: =============== DONE WITH MIGRATION ================ 5.34394
LDB: ============= START OF LOAD BALANCING ============== 9.17656
Segmentation fault (core dumped)

• Next message: Nikhil Maroli: "Including two parameters for protein-ligand simulation"
• Previous message: Giacomo Fiorin: "Re: Biasing potential of a Metadynamics Simulation"
• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]

This archive was generated by hypermail 2.1.6 : Tue Dec 27 2016 - 23:21:29 CST