Re: [PATCH] server: assert on failure to open /dev/null (Coverity)

[Alexandre Julliard]

Eric Pouech  writes:
> Marcus,
>
> IMO,fatal_error() should be preferred over assert() (hoping that
> fatal_error() is actually marked as no_return)

There's no reason to have an error here, it will still work fine even if
we can't open /dev/null.