| From: | Jérémie Grauer <jeremie(dot)grauer(at)cosium(dot)com> |
|---|---|
| To: | pgsql-hackers(at)lists(dot)postgresql(dot)org |
| Subject: | new option to allow pg_rewind to run without full_page_writes |
| Date: | 2022-11-03 15:54:13 |
| Message-ID: | 367d01a7-90bb-9b70-4cda-248e81cc475c@cosium.com |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
Hello,
Currently pg_rewind refuses to run if full_page_writes is off. This is
to prevent it to run into a torn page during operation.
This is usually a good call, but some file systems like ZFS are
naturally immune to torn page (maybe btrfs too, but I don't know for
sure for this one).
Having the option to use pg_rewind without the cost associated with
full_page_writes when using a system immune to torn page is beneficial:
increased performance and more compact WAL.
This patch adds a new option "--no-ensure-full-page-writes" to pg_rewind
for this situation, as well as patched documentation.
Regards,
Jeremie Grauer
| Attachment | Content-Type | Size |
|---|---|---|
| v1-0001-adds-the-option-no-ensure-full-page-writes-to-pg_rew.patch | text/x-patch | 9.6 KB |
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Melanie Plageman | 2022-11-03 17:00:24 | Re: pg_stat_bgwriter.buffers_backend is pretty meaningless (and more?) |
| Previous Message | Peter Eisentraut | 2022-11-03 15:49:51 | Re: Privileges on PUBLICATION |