Re: Background writer and checkpointer in crash recovery

On Tue, Aug 03, 2021 at 02:19:22PM +1200, Thomas Munro wrote:
> On Tue, Aug 3, 2021 at 9:52 AM Thomas Munro <thomas(dot)munro(at)gmail(dot)com> wrote:
> > On Tue, Aug 3, 2021 at 1:17 AM Robert Haas <robertmhaas(at)gmail(dot)com> wrote:
> > > That's great. I just realized that this leaves us with identical
> > > RequestCheckpoint() calls in two nearby places. Is there any reason
> > > not to further simplify as in the attached?
> >
> > LGTM.
>
> And pushed.

Gripe: this made the "ps" display worse than before.

7ff23c6d2 Run checkpointer and bgwriter in crash recovery.

A couple years ago, I complained that during the end-of-recovery
checkpoint, the "ps" display still said "recovering NNNNN", which made
it look like it was stuck on a particular WAL file.

That led to commit df9274adf, which updated the startup process's "ps"
to say "end-of-recovery checkpoint".

But since the start process no longer does the checkpoint, it still
says:

postgres 19738 11433 5 19:33 ? 00:00:01 postgres: startup recovering 000000010000000C000000FB
postgres 19739 11433 3 19:33 ? 00:00:00 postgres: checkpointer performing end-of-recovery checkpoint

That looks inconsistent. It'd be better if the startup process's "ps"
were cleared.

--
Justin