Hi,
A 'NetCDF error: invalid argument' message indicates something is
probably going wrong at a lower level, maybe with your disk IO. We
need some more information; what version of Amber are you using and
what patches have been applied, how was Amber configured, etc. Also
you still haven't reported if the test cases pass or not
($AMBERHOME/test/test_amber_serial.sh,
$AMBERHOME/test/test_amber_parallel.sh,
$AMBERHOME/AmberTools/test/test_at_serial.sh etc).
-Dan
On Mon, May 22, 2017 at 2:28 AM, xu qingliang
<xuqingliangouc.hotmail.com> wrote:
> Dear David,
>
> Thanks for your reply ! We have tried different strategies based on your suggestions.
>
>>>>Does the error depend in any way on the specifics of the system you are
> running? Do the test cases run correctly on the national laboratory computer?>>>>>>
>
> No, there is no dependent on the systems. We have tried different systems. We have run all these systems in the machine of our own lab, and everything are working well. It seems to our system has run on the machine of the national lab. There is log file, but not the trajectory file or coordinate file. I have checked the log.out file, and I could find an error at the end. But I cannot understand.
>
> FINAL RESULTS
>
>
>
> NSTEP ENERGY RMS GMAX NAME NUMBER
> 5000 -5.1597E+05 1.1946E-01 1.1326E+01 Na+ 59780
>
> BOND = 35711.6354 ANGLE = 7251.0023 DIHED = 32097.0014
> VDWAALS = 61854.2916 EEL = -746301.8514 HBOND = 0.0000
> 1-4 VDW = 9500.0869 1-4 EEL = 83916.7562 RESTRAINT = 0.0000
> NetCDF error: Invalid argument
>
>
>
>>>>>Can you run your test case in serial mode? I note that you are using 240
> threads for you calculation--this is both likley to be very inefficient,
> and may trigger errors that we don't test for very much. So try a short
> parallel simulation with a much smaller number of threads.>>>>>
>
>
> Yes, we have tried on one cpu wtih 24 cores, but the problem is the same.
>
>
> Thanks for your help again !
>
>
> Best wishes,
>
>
>
>
>
>
>
> <http://aka.ms/weboutlook>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
--
-------------------------
Daniel R. Roe
Laboratory of Computational Biology
National Institutes of Health, NHLBI
5635 Fishers Ln, Rm T900
Rockville MD, 20852
https://www.lobos.nih.gov/lcb
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon May 22 2017 - 07:30:03 PDT
