On Mon, 2025-08-11 at 14:13 +0000, Ghiurea, Isabella wrote:
> I believe this is related to VACUUM FULL behavior in a streaming replication cfg ,
> for the standby server the space is not been reclaimed , But the sieze of db reported
> with pg_pretty_db_size is identical on both db size but is not the actual sum of
> /var/lib/pgsql/data1 files +dir size on the active PG were vacuum full reclaim the space.
>
> Is this a normal behavior in PG with reporting incorrect db size if vacuum full has
> not reclaim the OS disk space on both server?
VACUUM (FULL) will also reclaim the size on the standby.
There can be some delay in the case of a replication conflict.
Yours,
Laurenz Albe
