Horton
Menu

Re: Are ZFS snapshots unsafe when PGSQL is spreading through multiple zpools?

Bruce Momjian <bruce@momjian.us>

View thread
From: Bruce Momjian <bruce@momjian.us>
To: HECTOR INGERTO <hector_25e@hotmail.com>
Cc: Laurenz Albe <laurenz.albe@cybertec.at>, Magnus Hagander <magnus@hagander.net>, "pgsql-general@postgresql.org <pgsql-general@postgresql.org>" <pgsql-general@postgresql.org>
Date: 2023-01-17T21:38:43Z
Lists: pgsql-general 
On Tue, Jan 17, 2023 at 03:22:02PM +0000, HECTOR INGERTO wrote:
> > Another case: a transaction COMMITs, and a slightly later transaction reads
> the data
> > and sets a hint bit.  If the snapshot of the file system with the data
> directory in it
> > is slightly later than the snapshot of the file system with "pg_wal", the
> COMMIT might
> > not be part of the snapshot, but the hint bit could be.
> >
> > Then these uncommitted data could be visible if you recover from the
> snapshot.
> >
> > Yours,
> > Laurenz Albe
> 
>  
> 
> Thank you all. I have it clearer now.
> 
>  
> 
> As a last point. Making the snapshot to the WAL dataset first or last would
> make any difference?

How would you know which WAL snapshot to use, and if there is any
missing data between them?

I have often wondered if we could document rules where multiple
asychronous snapshots would be safe but have never gotten very far.

-- 
  Bruce Momjian  <bruce@momjian.us>        https://momjian.us
  EDB                                      https://enterprisedb.com

Embrace your flaws.  They make you human, rather than perfect,
which you will never be.