Horton

Re: CLUSTER on partitioned index

Michael Paquier <michael@paquier.xyz>

View thread View on postgr.es
From: Michael Paquier <michael@paquier.xyz>
To: Justin Pryzby <pryzby@telsasoft.com>
Cc: Alvaro Herrera <alvherre@alvh.no-ip.org>, 李杰(慎追) <adger.lj@alibaba-inc.com>, pgsql-hackers@lists.postgresql.org, 曾文旌(义从) <wenjing.zwj@alibaba-inc.com>, Zhihong Yu <zyu@yugabyte.com>
Date: 2022-04-13T21:11:28Z
Lists: pgsql-hackers 
On Wed, Apr 13, 2022 at 05:52:14AM -0500, Justin Pryzby wrote:
> Are you sure?  The ownership re-check in cluster_rel() occurs after acquiring
> locks.

Yep, you are right.  However, the SQL test does not check for this
blocking scenario.  In short, removing the new ACL check in
get_tables_to_cluster_partitioned() makes the test behave the same
way.  Could you implement an isolation check to make sure that the
difference is visible?  The SQL check looks useful in itself, either
way.
--
Michael

Commits

  1. Add some isolation tests for CLUSTER

  2. Have CLUSTER ignore partitions not owned by caller

  3. Remove "recheck" argument from check_index_is_clusterable()

  4. Allow CLUSTER on partitioned tables

  5. doc: Add backlinks to progress reporting documentation