> On Oct 6, 2021, at 4:12 PM, Peter Geoghegan <pg@bowt.ie> wrote:
>
>>
>> Ok, excellent, that was probably the only thing that had me really hung up. I thought you were still asking for
pg_amcheckto filter out the --parent-check option when in recovery, but if you're not asking for that, then I might
haveenough to go on now.
>
> Sorry about that. I realized my mistake (not specifically addressing
> pg_is_in_recovery()) after I hit "send", and should have corrected the
> record sooner.
>
>> I was using "downgrading" to mean downgrading from bt_index_parent_check() to bt_index_check() when
pg_is_in_recovery()is true, but you've clarified that you're not requesting that downgrade, so I think we've now gotten
pastthe last sticking point about that whole issue.
>
> Right. I never meant anything like making a would-be
> bt_index_parent_check() call into a bt_index_check() call, just
> because of the state of the system (e.g., it's in recovery). That
> seems awful, in fact.
Please find attached the latest version of the patch which includes the changes we discussed.
—
Mark Dilger
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company