Re: pg_createsubscriber --dry-run logging concerns

On Wed, Oct 1, 2025 at 1:09 PM Hayato Kuroda (Fujitsu)
<kuroda(dot)hayato(at)fujitsu(dot)com> wrote:
>
> Dear Peter,
>
> > IMO, it's not good to fool people into thinking something has happened
> > when in fact nothing happened at all. I think the logging of this tool
> > should be much more explicit wrt when it is/isn't in dry-run mode.
> > Perhaps like this:
> >
> > NORMAL
> > pg_log_info("creating publication \"%s\" in database \"%s\"", ...)
> >
> > DRY-RUN
> > pg_log_info("[dry-run] would create publication \"%s\" in database \"%s\"", ...)
>
> Per my understanding, almost all the output must be adjusted based on the mode, right?
> I feel it introduces a burden.
> Can we solve the issue if we print additional message at the beginning if the
> command runs with dry-run mode?
>

Hi Kuroda-san,

Yes, that is one way. Something is better than nothing, at least...

I think that not *everything* in dry mode is fake; some of the logged
operations are real. So, it might be good if we can show fake ones
differently. e.g. It may not take much effort to introduce a wrapper
that inserts a prefix. Use this as a replacement for just a few
specific info logs.

(code below may not work; it's just for illustrative purposes)

#define pg_log_info_checkdry(...) do {\
if (dry_run)\
pg_log_generic(PG_LOG_INFO, PG_LOG_PRIMARY, "[dry-run NOP]" __VA_ARGS__);\
else;\
pg_log_generic(PG_LOG_INFO, PG_LOG_PRIMARY, __VA_ARGS__);\
} while (0);

======
Kind Regards,
Peter Smith.
Fujitsu Australia