Comments on: Using pg_repack to Rebuild PostgreSQL Database Objects Online

By: sami lekesiz

sami lekesiz — Tue, 21 Jul 2020 21:19:04 +0000

Thanks for post. I dont understand why you use shared_preload_libraries? I use pgrepack without shared_preload_libraries.

By: Javier Zayas

Javier Zayas — Tue, 24 Dec 2019 16:13:22 +0000

Quick question. If i run a “re_pack -d database_name” do I need twice as much disk as the entire database, or do I just need twice as much disk as the largest table? I ask because I’m trying to figure out if I need to repack by table since I am limited on disk space and am wondering if the extension is smart enough to do a table at a time and clean up after itself as it goes

By: Ashutosh

Ashutosh — Thu, 07 Feb 2019 06:37:56 +0000

Nice Post. One of the disadvantage of pg_repack that we found is – If you have to cancel repack execution in middle (say due to disk space), then it will leave behind all its triggers & indexes on live table, which needs to be cleaned manually.

By: abfnt1

abfnt1 — Wed, 06 Feb 2019 17:12:47 +0000

Nice article! Another example you may want to include is running pg_repack on a production database using the D, –no-kill-backend parameter (don’t kill other backends when timed out). List of parameters are at http://reorg.github.io/pg_repack/.