This archive contains answers to questions sent to Unidata support through mid-2025. Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated. For the most up-to-date information on the use of NSF Unidata software and data services, please consult the Software Documentation first.
Denise, > To: address@hidden > From: "Denise Maddix" <address@hidden> > Subject: Possibe netCDF bug in version 3.4 > Organization: LLNL > Keywords: 200006061554.e56FsMT15293 In the above message, you wrote: > I have been using netCDF 3.4 and have encountered a problem that does not > occur with netCDF 2.4, in a large dispersion code I have written. I have > written a small test code only using netCDF calls and the same problem occurs. > > I am defining two arrays, 3D and 4D with the last dimension being the > unlimited time dimension and I am using the ncsfil call not to fill the > data. I am using dynamic memory allocation to allocate space for the > two arrays. > > When I use the ncsfil command with option, "NCNOFILL", I > get no error in the code for any reason. However, when I use ncdump to > dump the netCDF file, I get the message, "NOT A NETCDF FILE." I know I > am not overflowing memory, since I have written a very small test code > to check this. The error only occurs for small sized arrays. I have > also verified the problem does not occur when I load with netCDF 2.4. > > Have you ever heard of a problem like this? No, I've never encountered this problem before. Does the test program close the netCDF file with a call to NCClOS() before exiting? > Denise Maddix > Computer Scientist > Lawrence Livermore National Laboratory > (925)- 423-4236 > address@hidden Regards, Steve Emmerson <http://www.unidata.ucar.edu>