[Pw_forum] phcg restart problem

[weeliat]

Dear All,

I am using phcg to calculate vibration modes at the gamma point for a 100+ atom system.

I might be wrong but I have seen a recover option for phcg.  I am facing a problem in restarting the phcg calculation after hitting my allowed walltime.

I have only added the line recover=.true. into the initial phcg input file.  The error msg for the restart run is:

*** An error occurred in MPI_Allreduce
*** on communicator MPI COMMUNICATOR 5 SPLIT FROM 3
*** MPI_ERR_TRUNCATE: message truncated
*** MPI_ERRORS_ARE_FATAL (your MPI job will now abort)

There is an output file from this restart run that reads:

     Program PHCG v.4.2         starts on  1Jan2014 at 21:43: 1 

     This program is part of the open-source Quantum ESPRESSO suite
     for quantum simulation of materials; please acknowledge
         "P. Giannozzi et al., J. Phys.:Condens. Matter 21 395502 (2009);
          URL http://www.quantum-espresso.org", 
     in publications or presentations arising from this work. More details at
     http://www.quantum-espresso.org/wiki/index.php/Citing_Quantum-ESPRESSO

     Parallel version (MPI), running on    32 processors
     R & G space division:  proc/pool =   32
     file H.pz-vbc.UPF: wavefunction(s)  1S renormalized

     Planes per process (thick) : nr3 = 216 npp =    7 ncplane = 46656

     Proc/  planes cols     G    planes cols    G      columns  G
     Pool       (dense grid)       (smooth grid)      (wavefct grid)
        1     7   1118   159298    7   1118   159298    282    19926
        2     7   1118   159298    7   1118   159298    282    19926
        3     7   1118   159298    7   1118   159298    280    19920
        4     7   1118   159298    7   1118   159298    280    19924
        5     7   1118   159298    7   1118   159298    280    19924
        6     7   1118   159298    7   1118   159298    280    19924
        7     7   1118   159294    7   1118   159294    280    19924
        8     7   1118   159294    7   1118   159294    280    19924
        9     7   1118   159294    7   1118   159294    280    19920
       10     7   1118   159294    7   1118   159294    280    19920
       11     7   1120   159300    7   1120   159300    280    19920
       12     7   1120   159300    7   1120   159300    280    19920
       13     7   1119   159299    7   1119   159299    281    19923
       14     7   1118   159290    7   1118   159290    280    19920
       15     7   1118   159290    7   1118   159290    280    19920
       16     7   1118   159298    7   1118   159298    280    19920
       17     7   1118   159298    7   1118   159298    280    19920
       18     7   1118   159298    7   1118   159298    280    19920
       19     7   1118   159298    7   1118   159298    280    19920
       20     7   1118   159298    7   1118   159298    280    19920
       21     7   1118   159298    7   1118   159298    280    19920
       22     7   1118   159294    7   1118   159294    280    19920
       23     7   1118   159294    7   1118   159294    280    19920
       24     7   1118   159294    7   1118   159294    280    19920
       25     6   1118   159294    6   1118   159294    280    19920
       26     6   1118   159294    6   1118   159294    280    19920
       27     6   1118   159294    6   1118   159294    280    19920
       28     6   1118   159294    6   1118   159294    280    19920
       29     6   1118   159294    6   1118   159294    280    19920
       30     6   1118   159294    6   1118   159294    280    19920
       31     6   1118   159294    6   1118   159294    282    19926
       32     6   1118   159294    6   1118   159294    282    19922
     tot    216  35781  5097465  216  35781  5097465   8969   637483

The output file stops here before the " ***  Starting Conjugate Gradient minimization         ***" 
No CRASH file is created.

Is this a problem with the phcg v.4.2 or something else is wrong?

Thanks,

wee-liat
Carnegie Mellon University
USA
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20140101/69bbade6/attachment.html>