Comment 5 for bug 1279048

Quoting Seth Arnold (<email address hidden>):
> A few other small things that I noticed:

Hi Seth,

> - Some failure cases don't have logging:
> setup_base_path() !base_path
> set_clone_children() snprintf()
> set_use_hierarchy() snprintf()
>
> - do_move_pid_main() fprintf() error message slightly incorrect

thanks, fixed these in git.

> - get_pid_scm_complete() char *output is not marked nih_local, this looks
> like a memory leak.

Drat, I was hoping you'd found a memory leak, however this isn't
one. It should be better commented at get_pid_scm_complete(),
but output will be nih_strduped() with the parent we pass in,
which is the socket data struct. So it will be freed by nih
when the socket data struct is freed. (I've added a comment in
git)