Re: pg_bulkload slowing down

On 2025-Jul-28, Sbob wrote:

> All;
> 
> 
> We are loading a series of files into a single table, each file has approx
> 500million rows, the first file took ~ 1.25 hours, now on file number 8 it
> has been running for more than 10 hours, there is only a PK index on the
> table.

I bet it's trying to find unused OIDs for toast pointers -- there are
only 2^32-1 of those available in any one toast table, and by the eighth
file of 500 million rows, you'd run out.

You could workaround this by partitioning the table.

-- 
Álvaro Herrera               48°01'N 7°57'E  —  https://www.EnterpriseDB.com/