On Wed, Jul 17, 2019 at 9:57 AM Tom Lane <tgl@sss.pgh.pa.us> wrote:
> It looks that way, but how would a broken non-shared index have held up
> autovacuuming in other databases? Maybe, as this one's xmin horizon
> got further and further behind, the launcher eventually stopped
> considering launching workers into any other databases? That seems
> like a bad thing; it's postponing work that will need to be done
> eventually.
I don't know exactly how the launcher would behave offhand, but it's
clear that not being able to VACUUM one table in one database (because
it has a corrupt index) ultimately risks the availability of every
database in the cluster. Many installations receive little to no
supervision, so it may just be a matter of time there. That is
certainly a bad thing.
--
Peter Geoghegan