| From: | "Andrus" <kobruleht2(at)hot(dot)ee> |
|---|---|
| To: | "Tom Lane" <tgl(at)sss(dot)pgh(dot)pa(dot)us>, "David G(dot) Johnston" <david(dot)g(dot)johnston(at)gmail(dot)com> |
| Cc: | "Michael Paquier" <michael(at)paquier(dot)xyz>, "Adrian Klaver" <adrian(dot)klaver(at)aklaver(dot)com>, "pgsql-general" <pgsql-general(at)postgresql(dot)org> |
| Subject: | Re: Query returns no rows in pg_basebackup cluster |
| Date: | 2020-05-25 05:47:32 |
| Message-ID: | F43B6DD2AB7F44D7BDC16F51CF8D88F4@dell2 |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-general |
Hi!
>>> No. Physical copies need to be based on the same platform.
>> Does the O/S that the client software runs on really affect this?
>To the extent that the O/S determines text sort order, yes; see thread.
>The short answer here is that we aren't going to support such cases.
>If you try to replicate across platforms, and it works, you're in luck.
>If it doesn't work, you get to keep both pieces; we will not accept
>that as a bug.
In 2017 Peter wrote that ICU-based collations will offered alongside the libc-based collations (1)
Currently it still requires re-compilation of Postgres for all binary replication platforms.
Maybe ICU locale will selected during installation automatically in Postgres 13 . Using same ICU locale in all replication platforms
will hopefully fix the issue.
Currently option is to use ucs_basic as default collation when creating cluster.
(1) https://www.2ndquadrant.com/en/blog/icu-support-postgresql-10/
Andrus.
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Andrus | 2020-05-25 06:02:49 | Re: Query returns no rows in pg_basebackup cluster |
| Previous Message | Nico De Ranter | 2020-05-25 05:30:21 | Re: pg_dump crashes |