On Tue, 5 Mar 2019 at 03:01, Derek Hans <derek.h...@gmail.com> wrote: > Based on a reply to reporting this as a bug, moving rows out of foreign > partitions is not yet implemented so this is behaving as expected. There's a > mention of this limitation in the Notes section of the Update docs.
(Moving this discussion to -Hackers) In [1], Derek reports that once a row is inserted into a foreign partition that an UPDATE does not correctly route it back out into the correct partition. I didn't really follow the foreign partition code when it went in, but do recall being involved in the documentation about the limitations of partitioned tables in table 5.10.2.3 in [2]. Unfortunately, table 5.10.2.3 does not seem to mention this limitation at all. As Derek mentions, there is a brief mention in [3] in the form of: "Currently, rows cannot be moved from a partition that is a foreign table to some other partition, but they can be moved into a foreign table if the foreign data wrapper supports it." I don't quite understand what a "foreign table to some other partition" is meant to mean. Partitions don't have foreign tables, they can only be one themselves. I've tried to put all this right again in the attached. However, I was a bit unsure of what "but they can be moved into a foreign table if the foreign data wrapper supports it." is referring to. Copying Robert and Etsuro as this was all added in 3d956d9562aa. Hopefully, they can confirm what is meant by this. [1] https://www.postgresql.org/message-id/cagrp7a3xc1qy_b2wjcgad8uqts_ndcjn06o5mts_ne1nyhb...@mail.gmail.com [2] https://www.postgresql.org/docs/devel/ddl-partitioning.html#DDL-PARTITIONING-DECLARATIVE-LIMITATIONS [3] https://www.postgresql.org/docs/devel/sql-update.html -- David Rowley http://www.2ndQuadrant.com/ PostgreSQL Development, 24x7 Support, Training & Services
doc_confirm_foreign_partition_limitations.patch
Description: Binary data
