Closed nickjer closed 9 years ago
This may be due to the fact that VFTSolid doesn't delete all the files in the subdirectories before it tries to remove the subdirectories.
Files that still remain are:
$ ll
total 184
drwxr-xr-x 2 jnicklas appl 4096 Jun 30 09:29 ./
drwxr-xr-x 4 jnicklas appl 4096 Jun 30 09:29 ../
-rw-r--r-- 1 jnicklas appl 6268 Jun 30 09:28 ctsp.out
-rw-r--r-- 1 jnicklas appl 0 Jun 30 09:28 heating-time.out
-rw-r--r-- 1 jnicklas appl 163996 Jun 30 09:28 tec-fem.out
I can take a look at this.
This is fixed in the del-ctsp-subdir branch. It works in both Windows and Linux. It basically completely removes all CTSPsubdXXX
sub-directories before it creates the CTSP data.
Waiting on Ed's approval or suggestions.
Thanks. Did you want to go ahead and Delete() those *.out? Actually, they are not used anymore so maybe we should remove them from CTSP.x Best wishes, EdP
The fix I implemented deletes all files irrespective of their names in the CTSPsubdXXX sub-directories. So the *.out are also deleted now.
The ctsp.out
is important as it has the log information for the CTSP run in that sub-directory. That is generated when you run CTSP in the VFT iHPC web app on OSC.
I will go ahead and merge in the branch with the fix. I'll leave it up to you if you want edit CTSP's actual code so that it doesn't generate heating-time.out
and tec-fem.out
. But with the fix, they won't cause anymore trouble.
Seems after you export CTSP from VFTSolid with some number of cores and later reduce the number of cores, there are still extra
CTSPsubdxxx
sub-directories laying around.Is this the intended feature or a bug?