Skip to main content
14 events
when toggle format what by license comment
Jul 19, 2023 at 20:12 answer added jjanes timeline score: 0
Jul 19, 2023 at 17:59 comment added jjanes Your partitioning scheme is not behaving as expected. Why is mdr_querystats_part_oct, for example, being scanned at all, rather than being ruled out at planning time? Maybe you also have more partitions that were ruled out, we can't know that just from the plan.
Jul 19, 2023 at 17:15 comment added jjanes Your plan seems internally inconsistent. How can the Parallel Append find only 342809 rows when the total of the children nodes is 1.2M? Maybe it is something to do with how the rows get distributed over the parallel workers? Could you repeat with max_parallel_workers_per_gather set to zero?
Jul 19, 2023 at 13:54 comment added jjanes @LaurenzAlbe I guess you can, provided you spell them correctly :)
Jul 19, 2023 at 11:46 comment added BigSmoke You may find that this kind of time-series data will be more efficient to query when you use the TimescaleDB extension.
Jul 19, 2023 at 7:50 comment added drew458 @jjanes I added the informations you mentioned (Postgres version is 14). Also there was a typo in the query aliases, sorry!
Jul 19, 2023 at 7:49 history edited drew458 CC BY-SA 4.0
added 11893 characters in body
Jul 19, 2023 at 7:46 history edited drew458 CC BY-SA 4.0
added 6 characters in body
Jul 19, 2023 at 7:02 comment added Laurenz Albe @jjanes Why can't you use aliases in the ORDER BY clause?
Jul 18, 2023 at 15:20 history edited drew458 CC BY-SA 4.0
deleted 1 character in body
Jul 18, 2023 at 14:52 history edited drew458 CC BY-SA 4.0
added 1 character in body
Jul 18, 2023 at 14:51 history edited drew458 CC BY-SA 4.0
deleted 1 character in body
S Jul 18, 2023 at 14:51 review First questions
Jul 18, 2023 at 19:52
S Jul 18, 2023 at 14:51 history asked drew458 CC BY-SA 4.0