Barry Smith <bsmith@mcs.anl.gov> writes:

>   test:
>     suffix: restart_0
>     requires: hdf5
>     args: -run_type test -refinement_limit 0.0    -bc_type dirichlet -interpolate 1 -petscspace_order 1 -dm_view hdf5:sol.h5 -vec_view hdf5:sol.h5::append
>
>   test:
>     suffix: restart_1
>     requires: hdf5
>     args: -run_type test -refinement_limit 0.0    -bc_type dirichlet -interpolate 1 -petscspace_order 1 -f sol.h5 -restart
>
> See a problem?
>
> Should the same run of the example view the files and then load them back in? versus trying to read in a data file from another run that may not even have been created before and even if it was, the file was definitely created  in a different directory?

So if write only is broken, do you want both to fail?  I think it's
better to read and write separately, with comparison using h5diff, since
that independently tests read vs write and establishes backward
compatibility, which you'd really like the test system to make you deal
with explicitly.