From: Ruchi Sachdeva (ruchi.namd_at_gmail.com)
Date: Fri Feb 27 2009 - 03:55:30 CST
I am trying to carry out md simulation on a linux cluster. After running
smoothly for a few pico seconds and output files being made, all namd jobs
are getting exit with an error message:
"Fatal error on PE 0> FATAL ERROR: Error writing XSC restart file
smd.restart.xsc: Interrupted system call"
Jobs exit at random time point with error in writing restart files,
sometimes it is .restart.xsc or .restart.vel or restart.coor file. If I set
the option 'binaryrestart no' then I get another error : "Fatal error on PE
0> FATAL ERROR: EOF in PDB::write line 26213 - file system full?:
Interrupted system call"
However jobs complete successfully when restart files are either not being
written or they are written only 4-5 times during the whole run. Although
this way I can made my jobs run to completion but I want to know the reason
for above error messages and how to fix them. Can anybody please help? I
could not find anything relevant in the archive.
This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:52:25 CST