Re: Giuseppe Sacco 2018-10-01 <831b8b45c4e1d4732aa39b41e276d49e3ea4bf59(dot)camel(at)eppesuigoccas(dot)homedns(dot)org>
> So, finally, my question: is pg_ctlcluster the root cause of my
> problem? Why does it loop and wait if pg_ctl will exit anyway when the
> database start or when its timeout is triggered? Should I open a bug
> report for this?
pg_ctlcluster originated years before systemd, so it is doing many
things that should nowadays really be handled by the service file, but
we haven't thrown out the old code yet. There's lots of room for
improvement, yes.
Christoph