[ncl-talk] Bug in NCL 6.3.0 vector plot?

Rick Brownrigg brownrig at ucar.edu
Mon Jun 15 13:22:10 MDT 2015


Hi Robert,

For better or worse, it works for me -- I can't duplicate the SEGV.  Can
you run it under gdb, and send me the stack trace?

Ex:

$ gdb ncl vector_2d_coord_bug.ncl
gdb) run
...segv...
gdb) bt
...stack trace...
gdb) quit

Thanks,
Rick


On Mon, Jun 15, 2015 at 10:11 AM, Robert Redl <rxschuster at gmail.com> wrote:

> Sorry, a forgot to attach the script…
>
>
>
>
> > Am 15.06.2015 um 18:09 schrieb Robert Redl <rxschuster at gmail.com>:
> >
> > Dear NCL Team,
> >
> > I have a script where in several procedures different plots are created.
> One of these procedures creates a vector map plot using 2d coordinates
> stored in lon2d/lat2d attributes. All variables used in the procedures are
> declared as local. This used to work fine for NCL 6.2.1 and earlier, but I
> get a segmentation fault with version 6.3.0.
> >
> > Attached is a minimal example witch reproduces the error.
> > I’m using NCL Version: ncl_ncarg-6.3.0.Linux_RHEL6.4_x86_64_gcc447
> >
> > Running the code without a procedure works fine, no errors.
> > Running the code without a procedure but with a final delete(wks)
> statement reproduces the error.
> > Creating an identical vector plot with 1d coordinates within a procedure
> works fine, no errors.
> >
> > Thanks for looking into that!
> >
> > Robert
> >
> >
>
>
> _______________________________________________
> ncl-talk mailing list
> ncl-talk at ucar.edu
> List instructions, subscriber options, unsubscribe:
> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.ucar.edu/pipermail/ncl-talk/attachments/20150615/193faddc/attachment.html 


More information about the ncl-talk mailing list