Since the test is failing on architectures/platform I can't reproduce
on.. I'm readapting the test in such a way to minimize the amount of
code that can be responsible for the error.
We're not looking for all system processes in order to find our
mount processes(ps -ef); we're looking directly at the mount
processes that have been spawned by our test instead.
To say that a process is no longer alive:
We try to wait for conclusion to get the processState.
If the wait results in an *ExitError, it means that either
the proc has failed to run, or it exited failing(see doc);
in both cases we're happy.. since we previously made a check
to know that those same processes were alive.
The message from the linter makes a good point..
we're storing a reference for the mount processes, but we're not
using them.. Thus the committed lines.
I copied as much code as I could from the previous tests
that seemed more alike to the one I was trying to build.
I put it inside the functional mount tests,
with the other mount-command related tests,
as I thought (as per comments on "TestFunctional()") it
could safely share a profile in parallel..
Nir Soffer
x7upLime
Akihiro Suda
Steven Powell
KallyDev
Medya Gh
Andriy Dzikh
Ilya Zuyev
Sharif Elgamal