*when "touch abc"*
*abc file was created*
*i think the old directory is "readonly" for some reasons when I had this
directory "MD_Template" .. anyhow, I created another test run with the
followings:*
kwu.kwucs426:/hdd/kwu/abc$ ls -lt
total 9328
-rw-r--r-- 1 root root 7588646 Jun 25 08:41 water.prmtop
-rwxr-xr-x 1 root root 153 Jun 25 08:41 run_md_run01.sh
-rw-r--r-- 1 root root 484 Jun 25 08:41 md01.inp
-rw-r--r-- 1 root root 1951056 Jun 25 08:39 300.rst
kwu.kwucs426:/hdd/kwu/abc$ ./run_md_run01.sh
Unit 6 Error on OPEN:
md01.out
STOP PMEMD Terminated Abnormally!
kwu.kwucs426:/hdd/kwu/abc$ ls
300.rst md01.inp run_md_run01.sh water.prmtop
kwu.kwucs426:/hdd/kwu/abc$ cd ..
kwu.kwucs426:/hdd/kwu$ ls
abc GPU0_DATA MD_Template
kwu.kwucs426:/hdd/kingsley$ ls -lt
total 12
drwxr-xr-x 2 root root 4096 Jun 25 08:41 abc
drwxrwxr-x 9 kwu kwu 4096 Jun 24 21:46 MD_Template
drwxr-xr-x 2 kwu kwu 4096 Jun 21 18:56 GPU0_DATA
kwu.kwucs426:/hdd/kwu$ sudo chown kwu:kwu abc
kwu.kwucs426:/hdd/kwu$ ls -lt
total 12
drwxr-xr-x 2 kwu kwu 4096 Jun 25 08:41 abc
drwxrwxr-x 9 kwu kwu 4096 Jun 24 21:46 MD_Template
drwxr-xr-x 2 kwu kwu 4096 Jun 21 18:56 GPU0_DATA
kwu.kwucs426:/hdd/kwu$ cd abc/
kwu.kwucs426:/hdd/kwu/abc$ ls
300.rst md01.inp run_md_run01.sh water.prmtop
kwu.kwucs426:/hdd/kwu/abc$ ./run_md_run01.sh * It
was running for a couple minutes before the following error message came
out*
cudaMemcpy GpuBuffer::Download failed the launch timed out and was
terminated
kwu.kwucs426:/hdd/kwu/abc$ ls
300.rst md01.inp md01.out mden run_md_run01.sh
amd.log md01.mdcrd md01.rst mdinfo water.prmtop
kwu.kwucs426:/hdd/kwu/abc$
*it seemed working as my output file, mdcrd file an rst file were
created... I do not know what about:*
"cudaMemcpy GpuBuffer::Download failed the launch timed out and was
terminated" was meant for
On Mon, Jun 25, 2018 at 4:59 AM, David A Case <david.case.rutgers.edu>
wrote:
> On Sun, Jun 24, 2018, Lod King wrote:
> >
> > I installed cuda-8.0 in my ubuntu 16 with amber 16 and nvidia driver
> 384..
> >
> > however, when I tested run...it gave me the following error
> >
> > At line 122 of file master_setup.F90 (unit = 6, file =
> 'all_min_fixnoH.out')
> > Fortran runtime error: Cannot write to file opened for READ
>
> This is odd, but I'm suspecting that you might be on a read-only file
> system, or you are in a directory where you don't have write permission.
> Especially since something simple like ambpdb (which has nothing to do
> with cuda) gives this error:
>
> > $AMBERHOME/bin/ambpdb -p ../00.prep/helix2.prmtop -c all_min_nofix.rst >
> > min_helix2.pdb
> >
> > ./run_min_protein.sh: line 9: min_helix2.pdb: Permission denied
>
> What happens if you type "touch abc" in the directory where you are
> trying to run your jobs?
>
> ....dac
>
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Jun 25 2018 - 09:30:02 PDT
