Horton

Re: ALTER TABLE .. DETACH PARTITION CONCURRENTLY

Michael Paquier <michael@paquier.xyz>

View thread View on postgr.es
From: Michael Paquier <michael@paquier.xyz>
To: Amit Langote <amitlangote09@gmail.com>
Cc: Alvaro Herrera <alvherre@2ndquadrant.com>, Pg Hackers <pgsql-hackers@lists.postgresql.org>
Date: 2020-10-01T03:50:39Z
Lists: pgsql-hackers 
On Thu, Sep 24, 2020 at 12:51:52PM +0900, Amit Langote wrote:
> Also, I noticed that looking up a parent's partitions via
> RelationBuildPartitionDesc or directly will inspect inhdetached to
> include or exclude partitions, but checking if a child table is a
> partition of a given parent table via get_partition_parent doesn't.
> Now if you fix get_partition_parent() to also take into account
> inhdetached, for example, to return InvalidOid if true, then the
> callers would need to not consider the child table a valid partition.
> So, RelationGetPartitionQual() on a detached partition should actually
> return NIL, making my earlier claim about not needing the defensive
> CHECK constraint invalid.  But maybe that's fine if all places agree
> on a detached partition not being a valid partition anymore?

It would be good to get that answered, and while on it please note
that the patch needs a rebase.
--
Michael

Commits

  1. Make detach-partition-concurrently-4 less timing sensitive

  2. Improve documentation on DETACH PARTITION lock levels

  3. Track detached partitions more accurately in partdescs

  4. Allow a partdesc-omitting-partitions to be cached

  5. Fix relcache inconsistency hazard in partition detach

  6. Don't add a redundant constraint when detaching a partition

  7. ALTER TABLE ... DETACH PARTITION ... CONCURRENTLY

  8. Let ALTER TABLE Phase 2 routines manage the relation pointer

  9. Check default partitions constraints while descending