RE: Are ZFS snapshots unsafe when PGSQL is spreading through multiple zpools? - Mailing list pgsql-general

From HECTOR INGERTO
Subject RE: Are ZFS snapshots unsafe when PGSQL is spreading through multiple zpools?
Date
Msg-id GV1P189MB20363DCA2B387BB3E4B2643FF5C69@GV1P189MB2036.EURP189.PROD.OUTLOOK.COM
Whole thread Raw
In response to Re: Are ZFS snapshots unsafe when PGSQL is spreading through multiple zpools?  (Laurenz Albe <laurenz.albe@cybertec.at>)
Responses Re: Are ZFS snapshots unsafe when PGSQL is spreading through multiple zpools?  (Bruce Momjian <bruce@momjian.us>)
Re: Are ZFS snapshots unsafe when PGSQL is spreading through multiple zpools?  (Laurenz Albe <laurenz.albe@cybertec.at>)
List pgsql-general

> 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?

 

pgsql-general by date:

Previous
From: Laurenz Albe
Date:
Subject: Re: Tablespace OID, database OID, relfilenode
Next
From: Wiwwo Staff
Date:
Subject: Re: Tablespace OID, database OID, relfilenode