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

[Support #PAD-341482]: Failed test in netCDF build

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 Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated.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: [Support #PAD-341482]: Failed test in netCDF build
  • Date: Fri, 22 Aug 2008 14:51:10 -0600
Hi Peter,

> *** testing nf_put_att_double ...
> Message Catalog System: corrupt file.Message Catalog System: corrupt
> file.Message Catalog System: corrupt file.forrtl: severe (180):
> SIGBUS, bus error occurred
> Image PC Routine Line Source
> nf_test 000C93B7 Unknown Unknown Unknown
> nf_test 000C8853 Unknown Unknown Unknown
> nf_test 0009F4FA Unknown Unknown Unknown
> nf_test 0007A57A Unknown Unknown Unknown
> nf_test 00081309 Unknown Unknown Unknown
> libSystem.B.dylib 925FA09B Unknown Unknown Unknown
>
> Stack trace terminated abnormally.

The above messages are not from the netCDF test.

I would recommend using Disk Utility to verufy the disk on which the
tests were run, to see whether there is a problem with the disk or
file system.  Also, open up the Console application in Applications/Utilities
and see if there is a console message about a disk problem.  If so
and this is your boot disk, you will probably have to attempt a repair
with Disk Utility run from the Leopard install CD.

It also seems possible this could be an indication of a bug in the Intel
I/O runtime you are using, since I see someone on the Intel site has reported
an identical symptom when compiling a different library:

  
http://softwarecommunity.intel.com/isn/Community/en-US/forums/thread/30248363.aspx

You could distinguish between these two possibilities by trying to build
netCDF again in a different directory.  If it gets the same error in the
same place, it looks like there's something wrong with the Intel runtime
I/O library underneath netCDF.  If it now builds fine, or gets an I/O
error in a different place, it's more likely a file system problem.

We have Mac OS X Leopard, but the only Intel compiler we have to test with is
on a Linux system, and it builds fine on that.  We also have reports that
version 3.6.2 builds OK on Mac OS X with from Intel:

  http://www.intel.com/support/performancetools/sb/CS-027812.htm#13

Please let us know if you can determine the problem from the above.

--Russ

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



Ticket Details
===================
Ticket ID: PAD-341482
Department: Support netCDF
Priority: Normal
Status: Closed