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

[netCDF #QVT-485828]: netCDF4 make check on AIX

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 Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated.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: [netCDF #QVT-485828]: netCDF4 make check on AIX
  • Date: Tue, 12 Aug 2008 13:56:21 -0600
> Hi,
>
> Some of our users are planning to use netCDF4's parallel I/O features on a
> AIX 5.3 system. I managed to get HDF5 1.8.1 compiled for parallel I/O with
> all checks successful. NetCDF4 also compiled but many checks fail.
>
> Some errors are just of the following kind:
>
> *** Testing netcdf-4 mapped variable functions.
> *** testing mapping with 2x2 variable...ok.
> *** testing mapping with 2x3 variable...ok.
> *** testing simple example from C Users' Guide...ok.
> *** testing transposed rh example from C Users' Guide...ok.
> *** Tests successful!
> ERROR: 0032-184 MPI was not finalized  in routine unknown, task 0
> FAIL: tst_varms
>
> Here the culprit seems to be that exit (in macro FINAL_RESULTS) is called
> before MPI_Finalize().
>
> ...
> FINAL_RESULTS;
> #ifdef USE_PARALLEL
> MPI_Finalize();
> #endif
> }
>
>
>
> Others trigger HDF5 errors:
>
>
> ...
>
> *** Testing netcdf-4 dimensions.
> *** Checking that netcdf-4 dimids inq works on netcdf-3 file...ok.
> *** Checking that netcdf-4 dimids inq works on more complex netcdf-3
file...ok.
> *** Testing file with just one dimension...ok.
> *** Testing renaming of one dimension...ok.
> *** Testing file with just one unlimited dimension...ok.
> *** Testing file with two unlimited dimensions...ok.
> *** Testing ordering of dimensions...ok.
> HDF5-DIAG: Error detected in HDF5 (1.8.1) MPI-process 0:
> #000: H5Gdeprec.c line 872 in H5Gget_objinfo(): cannot stat object
> major: Invalid arguments to routine
> minor: Unable to initialize object
> #001: H5Gdeprec.c line 999 in H5G_get_objinfo(): name doesn't exist
> major: Symbol table
> minor: Object already exists
> #002: H5Gtraverse.c line 861 in H5G_traverse(): internal path traversal
> failed
> major: Symbol table
> minor: Object not found
> #003: H5Gtraverse.c line 691 in H5G_traverse_real(): traversal operator
> failed
> major: Symbol table
> minor: Callback failed
> #004: H5Gdeprec.c line 903 in H5G_get_objinfo_cb(): name doesn't exist
> major: Symbol table
> minor: Object not found
> #005: H5Gobj.c line 1096 in H5G_obj_lookup(): can't locate object
> major: Symbol table
> minor: Object not found
> #006: H5Gcompact.c line 527 in H5G_compact_lookup(): object not found
> major: Symbol table
> minor: Object not found
> *** Testing file with just one unlimited dimension and one var...ok.
>
> ...
>
> Is there something wrong on my side or should I simply ignore those errors
> for now?
>
> Cheers,
> Mathis
>
> --
> Mathis Rosenhauer
> Deutsches Klimarechenzentrum
> www.dkrz.de
>
>

Did you build with --enable-parallel-tests? And did the parallel tests work?
(Like nc_test/tst_parallel.c?)

Thanks,

Ed

Ticket Details
===================
Ticket ID: QVT-485828
Department: Support netCDF
Priority: Normal
Status: Open