On Sat, Jun 12, 2021 at 10:07 AM Noah Misch <noah@leadboat.com> wrote:
>
> On Fri, Jun 11, 2021 at 08:54:08PM -0500, Justin Pryzby wrote:
> > On Sat, Jun 05, 2021 at 12:08:01PM -0700, Noah Misch wrote:
> > > > > My preference is to remove pg_wait_for_backend_termination(). The use case
> > > > > that prompted this thread used pg_terminate_backend(pid, 180000); it doesn't
> > > > > need pg_wait_for_backend_termination().
> >
> > Is this an Opened Issue ?
>
> An Open Item? Not really, since there's no objective defect. Nonetheless,
> the attached is what I'd like to use.
Thanks. +1 to remove the pg_wait_for_backend_termination function. The
patch basically looks good to me. I'm attaching an updated patch. I
corrected a minor typo in the commit message, took docs and code
comment changes suggested by Justin Pryzby.
Please note that release notes still have the headline "Add functions
to wait for backend termination" of the original commit that added the
pg_wait_for_backend_termination. With the removal of it, I'm not quite
sure if we retain the the commit message or tweak it to something like
"Add optional timeout parameter to pg_terminate_backend".
<!--
Author: Magnus Hagander <magnus@hagander.net>
2021-04-08 [aaf043257] Add functions to wait for backend termination
-->
With Regards,
Bharath Rupireddy.
