Re: Reducing power consumption on idle servers

On Mon, Nov 21, 2022 at 2:43 AM Thomas Munro <thomas(dot)munro(at)gmail(dot)com> wrote:
>
> On Mon, Nov 21, 2022 at 3:35 AM Simon Riggs
> <simon(dot)riggs(at)enterprisedb(dot)com> wrote:
> > On Sat, 19 Nov 2022 at 10:59, Simon Riggs <simon(dot)riggs(at)enterprisedb(dot)com> wrote:
> > > New version attached.
> >
> > Fix for doc xref
>
> I removed a stray variable declaration from xlogrecovery.h, and wrote
> a draft commit message. I'll wait 24 hours before committing, to
> provide a last chance for anyone who wants to complain about dropping
> promote_trigger_file.
>
> (We could of course change it so that the timeout based wakeup only
> happens if you have actually set promote_trigger_file, but I think
> we've established that we just don't want the filesystem polling
> feature so I'm whispering this in parentheses.)

Thanks. The v11 patch mostly looks good to me and it passes cirrus-ci
tests - https://github.com/BRupireddy/postgres/tree/do_away_with_promote_trigger_file.

I have a comment:

- * Wait for more WAL to arrive. Time out after 5 seconds
- * to react to a trigger file promptly and to check if the
- * WAL receiver is still active.
+ * Wait for more WAL to arrive, when we will be woken
+ * immediately by the WAL receiver. Use of trigger file
+ * via promote_trigger_file is now fully removed.
*/
Do we need to introduce reference to removal of promote_trigger_file
in the code? If left as-is, it'll lie there for many years. Isn't it
enough to specify in appendix-obsolete-recovery-config.sgml?

--
Bharath Rupireddy
PostgreSQL Contributors Team
RDS Open Source Databases
Amazon Web Services: https://aws.amazon.com