On Wed, Apr 1, 2020 at 3:43 PM Andres Freund <and...@anarazel.de> wrote: > The thing that makes me really worried is that the contents of the time > mapping seem very wrong. I've reproduced query results in a REPEATABLE > READ transaction changing (pruned without triggering an error). And I've > reproduced rows not getting removed for much longer than than they > should, according to old_snapshot_threshold.
I think it would be a good idea to add a system view that shows the contents of the mapping. We could make it a contrib module, if you like, so that it can even be installed on back branches. We'd need to move the structure definition from snapmgr.c to a header file, but that doesn't seem like such a big deal. Maybe that contrib module could even have some functions to simulate aging without the passage of any real time. Like, say you have a function or procedure old_snapshot_pretend_time_has_passed(integer), and it moves oldSnapshotControl->head_timestamp backwards by that amount. Maybe that would require updating some other fields in oldSnapshotControl too but it doesn't seem like we'd need to do a whole lot. -- Robert Haas EnterpriseDB: http://www.enterprisedb.com The Enterprise PostgreSQL Company
