Re: pg_repack in cluster

On 1/10/20 10:01 AM, dagamier wrote:
> I have two databases that are clustered. One is my primary (DB1) and the
> other is my secondary (DB2). Both have the same tables and schemas. Could I
> use pg_repack against each of these separately (I am wanting to do this at a
> "by table" level) to clean up dead space that hasn't been returned? Should I
> be doing it this way? I would assume that dead space in DB1 would also exist
> in DB2 as deletes are performed as part of the replication process, but the
> reclamation of space back to the system wouldn't take place? Just curious if
> the fact that the database is being replicated would prevent the needed
> steps from occurring as it's my understanding that you can't drop a table
> that's being replicated between two instances.

Why not use VACUUM?

>
>
>
> --
> Sent from: https://www.postgresql-archive.org/PostgreSQL-general-f1843780.html
>
>

--
Adrian Klaver
adrian(dot)klaver(at)aklaver(dot)com