| From: | Lukasz Biegaj <lukasz(dot)biegaj(at)unity(dot)pl> |
|---|---|
| To: | pgsql-general(at)lists(dot)postgresql(dot)org |
| Subject: | data=writeback and safety of WAL files |
| Date: | 2021-01-27 07:46:58 |
| Message-ID: | 569a2a1d-4613-449e-6fa4-5cc685592d69@unity.pl |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-general |
Hi,
The <https://www.postgresql.org/docs/13/wal-intro.html> document states,
that in case of ext3 filesystems, filesystem journaling can be
configured with data=writeback option to increase performance (in
specific cases).
Per my understanding this causes some of the filesystem integrity
mechanism to be loosened and in case of PostgreSQL data partition it
would be okay because WAL files replicate this functionality at a higher
level.
What about the storage of WAL files itself? Wouldn't it be possible that
in case of unexpected reboot and dirty file system these files get
corrupted, making the database state impossible to bring to operational
level?
--
Lukasz Biegaj | Unity Group | https://www.unitygroup.com/
System Architect, AWS Solutions Architect
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Kai Daguerre | 2021-01-27 13:08:47 | Custom FDW - the results of a nested query/join not being passed as qual to the outer query |
| Previous Message | Ron | 2021-01-27 05:20:30 | Re: Postgres 9.3 service no longer starts on Windows 8.1 |