Hi,
I've been using Charles Parnot's excellent Gridstuffer to manage my
xgrid job submissions. I'm running into problems whilst attempting
to redirect output written to stdout.
The first few lines of my Gridstuffer submission file look like:
[blank line for no prototype]
-so data/VN_2_0.00001_0.00001.txt ./exe -repeats=10000 -
maxgen=20000 -
patchmax=10 -N=2 -datafreq=182 -lambdaf=0.00001 -lambdam=0.00001 -
seed=1
-so data/VN_4_0.00001_0.00001.txt ./exe -repeats=10000 -
maxgen=20000 -
patchmax=10 -N=4 -datafreq=182 -lambdaf=0.00001 -lambdam=0.00001 -
seed=2
-so data/VN_8_0.00001_0.00001.txt ./exe -repeats=10000 -
maxgen=20000 -
patchmax=10 -N=8 -datafreq=182 -lambdaf=0.00001 -lambdam=0.00001 -
seed=3
-so data/VN_16_0.00001_0.00001.txt ./exe -repeats=10000 -maxgen=20000
-patchmax=10 -N=16 -datafreq=182 -lambdaf=0.00001 -lambdam=0.00001 -
seed=4
....
Using this submission file produces the expected output file for each
job, on my client computer, in data/VN_2_0.00001_0.00001.txt etc.
The problem is that the output file is empty!
If I do not specify the -so flag, the stdout data is successfully
collected and saved back onto the client, and organised in the
standard gridstuffer way.
For my convenience I would like to be able to specify the filename
for saving the stdout of each job (my post-processing relies on the
filenames).
Can anyone advise me what I'm doing wrong?
