| From: | Israel Barth Rubio <barthisrael(at)gmail(dot)com> |
|---|---|
| To: | PostgreSQL Hackers <pgsql-hackers(at)lists(dot)postgresql(dot)org> |
| Subject: | Re: Add -k/--link option to pg_combinebackup |
| Date: | 2025-01-15 18:54:28 |
| Message-ID: | CAO_rXXDTJOe=px139mkT4wBB5MvQEiFKPvHTLTZh51n9Qwu+Lg@mail.gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
One concern that I have about this --link mode, which Euler Taveira
also got concerned about: the fact that it can invalidate the original
backups if the user modifies or starts the synthetic backup without
moving it to another file system or machine.
At the moment, pg_combinebackup issues a warning about that problem as
soon as the execution of the tool is about to finish. There is also a
warning in the docs. But some users don't pay proper attention to the
output and/or docs.
I wonder if we have some way, like pg_upgrade does (rename a global
file in the original cluster), so we could increase the protection
against that issue?
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Nathan Bossart | 2025-01-15 19:15:26 | Re: convert libpgport's pqsignal() to a void function |
| Previous Message | Nathan Bossart | 2025-01-15 18:50:00 | Re: use a non-locking initial test in TAS_SPIN on AArch64 |