ONE OF THE PROCESSES TERMINATED BADLY,TERMINATED WITH THE WXIT STRING:KILLED (SIGNAL 9)
Posted: Sun Apr 10, 2011 4:11 am
dear master
I have tried many differnent ways,changing compiler,mkl and mpi , and gotten the parallel VASP everytime，correctly.
but，there is always a problem，
when I use the serial version ，noting goes wrong
when I use the parallel version，get some messages like：
forrtl：server （174）：SIGSEGV，segmentation fault occured
vasp 00000000005DD414 unknown unknowm unknown
vasp 00000000009D1E45 unknown unknowm unknown
vasp 0000000000449368 unknown unknowm unknown
vasp 00000000004256AC unknown unknowm unknown
libc。so。6 0000003D96A1EC5D unknown unknowm unknown
vasp 00000000004255A9 unknown unknowm unknown
ONE OF THE PROCESSES TERMINATED BADLY,TERMINATED WITH THE WXIT STRING:KILLED (SIGNAL 9)
follow the messages in this forum，I did some homework like
ulimit -s unlimited
ldd vasp
add all libraries into LD_LIBRARY_PATH
make clean
make
but the mistake happens again
the most meaningful thing is,though the VASP terminated badly，it calculated the energy correctly。
in my humble opinion，there might be some codes missed，for instance， finalize（）or something like it。
<span class='smallblacktext'>[ Edited ]</span>
I have tried many differnent ways,changing compiler,mkl and mpi , and gotten the parallel VASP everytime，correctly.
but，there is always a problem，
when I use the serial version ，noting goes wrong
when I use the parallel version，get some messages like：
forrtl：server （174）：SIGSEGV，segmentation fault occured
vasp 00000000005DD414 unknown unknowm unknown
vasp 00000000009D1E45 unknown unknowm unknown
vasp 0000000000449368 unknown unknowm unknown
vasp 00000000004256AC unknown unknowm unknown
libc。so。6 0000003D96A1EC5D unknown unknowm unknown
vasp 00000000004255A9 unknown unknowm unknown
ONE OF THE PROCESSES TERMINATED BADLY,TERMINATED WITH THE WXIT STRING:KILLED (SIGNAL 9)
follow the messages in this forum，I did some homework like
ulimit -s unlimited
ldd vasp
add all libraries into LD_LIBRARY_PATH
make clean
make
but the mistake happens again
the most meaningful thing is,though the VASP terminated badly，it calculated the energy correctly。
in my humble opinion，there might be some codes missed，for instance， finalize（）or something like it。
<span class='smallblacktext'>[ Edited ]</span>