On 4/16/24 22:33, Tomas Vondra wrote: > On 4/15/24 20:35, Tomas Vondra wrote: >> On 4/15/24 10:18, Tomas Vondra wrote: > > ... > > That is, no parallel index builds on temporary tables. Which means the > test is not actually testing anything :-( Much more stable, but not very > useful for finding issues. > > I think the best way to stabilize the test is to just not delete the > rows. That means we won't have any "empty" ranges (runs of pages without > any tuples). >
I just pushed a revert and a patch to stabilize the test in a different way - Matthias mentioned to me off-list that DELETE is not the only way to generate empty ranges in a BRIN index, because partial indexes have the same effect. After playing with that a bit, that seems to work fine (works with parallel builds, not affected by cleanup), so done that way. regards -- Tomas Vondra EnterpriseDB: http://www.enterprisedb.com The Enterprise PostgreSQL Company
