[CIG-MC] citcoms - question about tracers

[tan2]

On Sun, Apr 1, 2012 at 2:51 PM, Alexander Evseev <alevseev at gmail.com> wrote:
> Good day, Eh,
>
> I have a question about CitcomS. I tried to restart it from 5400 step
> checkpoint with tracers generated randomly.
>
> Parameter tracer_ic_method was set to 0. So I expect tracers to be generated
> randomly, but instead of it, i get the following error (see below).
>
> When i tried the same input file, but set restart to 0 (so CitcomS started
> from beginning) it is all ok with tracers.
>
> Could you please help, what should i do to get randomly generated tracers
> with restart.
>
> ...................
> Input parameters taken from file 'input3.sample'
> Initialization complete after 37.9371 seconds
>
> read_checkpoint: restarting from
> /home/alevseev/citcoms/run/bin/case2/fulltest.chkpt.0.5400
> Error in reading checkpoint file: wrong sentinel, me=3
> Error in reading checkpoint file: wrong sentinel, me=4
> Error in reading checkpoint file: wrong sentinel, me=5
> Error in reading checkpoint file: wrong sentinel, me=7
> Error in reading checkpoint file: wrong sentinel, me=1
> Error in reading checkpoint file: wrong sentinel, me=2
> Error in reading checkpoint file: wrong sentinel, me=9
> Error in reading checkpoint file: wrong sentinel, me=11
> call read_checkpoint functioncall read_checkpoint functioncall
> read_checkpoint functioncall read_checkpoint functioncall read_checkpoint
> functioncall read_checkpoint functioncall read_checkpoint functionError in
> reading checkpoint file: wrong sentinel, me=6
> Error in reading checkpoint file: wrong sentinel, me=0
> call read_checkpoint functioncall read_checkpoint functioncall
> read_checkpoint functioncall read_checkpoint functioncall read_checkpoint
> functionError in reading checkpoint file: wrong sentinel, me=8
> Error in reading checkpoint file: wrong sentinel, me=10
> --------------------------------------------------------------------------
> mpirun has exited due to process rank 3 with PID 9224 on
> node romus exiting without calling "finalize". This may
> have caused other processes in the application to be
> terminated by signals sent by mpirun (as reported here).
>
> Best regards,
> Alexander
>


Hi Alex,
(cc to cig-mc list)

Sorry for the late reply as I am in vacation and don't have regular
internet access.

The task you were trying to achieve cannot be done with restart. The
restart is intended for resuming an interrupted run (e.g. hardware
failure, power loss) and should have identical input parameters as the
original run.

To achieve your task, you can save the temperature output at 5400th
step, then, start a new run with tracer parameters and "tic_method=-1"
to read the saved temperature.

Cheers,
Eh