Re: Is it correct to update db state in control file as "shutting down" during end-of-recovery checkpoint?

On Wed, Dec 8, 2021 at 9:49 AM Bossart, Nathan <bossartn@amazon.com> wrote:
>
> On 12/7/21, 5:21 PM, "Bharath Rupireddy" <bharath.rupireddyforpostgres@gmail.com> wrote:
> > On Wed, Dec 8, 2021 at 2:50 AM Bossart, Nathan <bossartn@amazon.com> wrote:
> >> I noticed that some (but not all) of the surrounding messages say
> >> "last known up at" the control file time.  I'm curious why you chose
> >> not to use that phrasing in this case.
> >
> > If state is DB_IN_END_OF_RECOVERY_CHECKPOINT that means the db was
> > interrupted while in end-of-recovery checkpoint, so I used the
> > phrasing similar to DB_IN_CRASH_RECOVERY and DB_IN_ARCHIVE_RECOVERY
> > cases. I would like to keep it as-is (in the v1 patch) unless anyone
> > has other thoughts here?
> > (errmsg("database system was interrupted while in recovery at %s",
> > (errmsg("database system was interrupted while in recovery at log time %s",
>
> I think that's alright.  The only other small suggestion I have would
> be to say "during end-of-recovery checkpoint" instead of "while in
> end-of-recovery checkpoint."

"while in" is being used by DB_IN_CRASH_RECOVERY and
DB_IN_ARCHIVE_RECOVERY messages. I don't think it's a good idea to
deviate from that and use "during".

> Another option we might want to consider is to just skip updating the
> state entirely for end-of-recovery checkpoints.  The state would
> instead go straight from DB_IN_CRASH_RECOVERY to DB_IN_PRODUCTION.  I
> don't know if it's crucial to have a dedicated control file state for
> end-of-recovery checkpoints.

Please note that end-of-recovery can take a while in production
systems (we have observed such things working with our customers) and
anything can happen during that period of time. The end-of-recovery
checkpoint is not something that gets finished momentarily. Therefore,
having a separate db state in the control file is useful.

Regards,
Bharath Rupireddy.