[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: 20040330:unaligned access pid=##### va =0x

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 Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated.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.

  • Subject: Re: 20040330:unaligned access pid=##### va =0x
  • Date: Thu, 01 Apr 2004 11:16:40 -0700
>To: <address@hidden>,
>To: <address@hidden>
>From: "Joseph Benckert" <address@hidden>
>Subject: Re: 20040330:unaligned access pid=##### va =0xnetcdf troubles
>Organization: UCAR/Unidata
>Keywords: 200403291443.i2TEhsrV014206

Joseph

> Russ, thanks for all the good help.  I'm off and running now.
> 
> I was wondering if you had any examples dealing with the unlimited
> dimension and record variables.  Our aim is to create 5 to 7 variables in
> one dataset and each of these variables will be "large" ( at least > 2gig
> and possiblly much bigger).  It's my understanding that the unlimited
> dimension is the way to accomplish this and I see examples of how to
> create a dataset with the unlimited dimension and a warning about writing
> to it without first creating a record but I don't really see any examples
> on it's use.

Yes, there are a couple of examples in the "Large File Support"
section of the F90 User's Guide (which I've got to get into the C and
Fortran-77 User's Guides soon):

  
http://www.unidata.ucar.edu/packages/netcdf/f90/Documentation/f90-html-docs/guide9.html#2236524

The second example shows how to use record variables to get lots of
data in a file, but the first example is only relevant for one very
large variable.

> Also, did you ever get a chance to run ncdump on that file that was giving
> me the errors on the PSC cluster(in the subject line)?  Any idea wha that
> error is?

Yes, sorry I forgot to tell you that I ran ncdump on it and it worked
fine, with no errors.  It showed "missing values" for all the data,
but the file is a valid netCDF file:

  $ ncdump UG5sf_1000.0000
  netcdf UG5sf_1000 {
  dimensions:
          x = 160 ;
          y = 20 ;
  variables:
          float Density(y, x) ;
          float Pressure(y, x) ;
          float XVelocity(y, x) ;
          float YVelocity(y, x) ;
          float XScale(x) ;
          float YScale(y) ;
  data:

   Density =
    _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, 
      _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, 

    ...
  }

--Russ
_____________________________________________________________________

Russ Rew                                         UCAR Unidata Program
address@hidden          http://www.unidata.ucar.edu/staff/russ