Use a temporary table to update scheduler metrics
When using insert into <...> select <...>, PostgreSQL disables
parallel querying. Under some circumstances (in our large production
database), this makes updating the scheduler metrics take a (very) long
time.
Parallel querying is allowed for create table <...> as select <...>,
and doing so restores the small(er) runtimes for this query (15 minutes
instead of multiple hours). To use that, we have to turn the function
into plpgsql instead of plain sql.
Related to swh/infra/sysadm-environment#3785 (closed)
Test Plan
Poked at the function in production until the runtime came back to a more sensible value. No existing tests are regressing.
Migrated from D6812 (view on Phabricator)