From: | Robert Haas <robertmhaas(at)gmail(dot)com> |
---|---|
To: | Alvaro Herrera <alvherre(at)alvh(dot)no-ip(dot)org> |
Cc: | Dilip Kumar <dilipbalaut(at)gmail(dot)com>, Andres Freund <andres(at)anarazel(dot)de>, "pgsql-hackers(at)postgresql(dot)org" <pgsql-hackers(at)postgresql(dot)org> |
Subject: | Re: trying again to get incremental backup |
Date: | 2023-11-20 19:27:43 |
Message-ID: | CA+TgmoYL1vq0Q-Br1uOkgP36Y0r7VfbbqA-XbSB+n=Zhz693=Q@mail.gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
On Mon, Nov 20, 2023 at 2:10 PM Robert Haas <robertmhaas(at)gmail(dot)com> wrote:
> > Is this meant to support multiple timelines each with non-overlapping
> > adjacent ranges, rather than multiple non-adjacent ranges?
>
> Correct. I don't see how non-adjacent LSN ranges could ever be a
> useful thing, but adjacent ranges on different timelines are useful.
Thinking about this a bit more, there are a couple of things we could
do here in terms of syntax. Once idea is to give up on having a
separate MANIFEST-WAL-RANGE command for each range and instead just
cram everything into either a single command:
MANIFEST-WAL-RANGES {tli} {startlsn} {endlsn}...
Or even into a single option to the BASE_BACKUP command:
BASE_BACKUP yadda yadda INCREMENTAL 'tli(at)startlsn-endlsn,...'
Or, since we expect adjacent, non-overlapping ranges, you could even
arrange to elide the duplicated boundary LSNs, e.g.
MANIFEST_WAL-RANGES {{tli} {lsn}}... {final-lsn}
Or
BASE_BACKUP yadda yadda INCREMENTAL 'tli(at)lsn,...,final-lsn'
I'm not sure what's best here. Trying to trim out the duplicated
boundary LSNs feels a bit like rearrangement for the sake of
rearrangement, but maybe it isn't really.
--
Robert Haas
EDB: http://www.enterprisedb.com
From | Date | Subject | |
---|---|---|---|
Next Message | David Steele | 2023-11-20 19:31:20 | Re: Use of backup_label not noted in log |
Previous Message | Robert Haas | 2023-11-20 19:14:00 | Re: Annoying build warnings from latest Apple toolchain |