Re: [PATCHES] Post-special page storage TDE support - Mailing list pgsql-hackers

From Peter Geoghegan
Subject Re: [PATCHES] Post-special page storage TDE support
Date
Msg-id CAH2-WzmKZt7NxKXQBBz3BLPjRTghxLvw2CGHu+nZ2WWyxzZ3Yg@mail.gmail.com
Whole thread Raw
In response to Re: [PATCHES] Post-special page storage TDE support  (Stephen Frost <sfrost@snowman.net>)
Responses Re: [PATCHES] Post-special page storage TDE support
List pgsql-hackers
On Wed, Nov 8, 2023 at 6:04 AM Stephen Frost <sfrost@snowman.net> wrote:
> In conversations with folks (my memory specifically is a discussion with
> Peter G, added to CC, and my apologies to Peter if I'm misremembering)
> there was a pretty strong push that a page should be able to 'stand
> alone' and not depend on something else (eg: pg_control, or whatever) to
> provide info needed be able to interpret the page.  For my part, I don't
> have a particularly strong feeling on that, but that's what lead to this
> design.

The term that I have used in the past is "self-contained". Meaning
capable of being decoded more or less as-is, without any metadata, by
tools like pg_filedump.

Any design in this area should try to make things as easy to debug as
possible, for the obvious reason: encrypted data that somehow becomes
corrupt is bound to be a nightmare to debug. (Besides, we already
support tools like pg_filedump, so this isn't a new principle.)

--
Peter Geoghegan



pgsql-hackers by date:

Previous
From: Amit Kapila
Date:
Subject: Re: Synchronizing slots from primary to standby
Next
From: Michael Paquier
Date:
Subject: Re: Requiring recovery.signal or standby.signal when recovering with a backup_label