Excessive storage space used by Supabase Docker

[SyedHaizal]

Hi everyone,

I am currently facing an issue where my locally hosted Supabase instance is using an excessive amount of storage of over 750GB. The instance is running inside a Proxmox VM on Ubuntu 24.04 LTS, managed through Coolify.

Running the size check and bloat check commands from the documentation gave the following output:

size check:

pg_size_pretty
745 GB

bloat check:

type	schemaname	object_name	bloat	waste
table	public	table1	1.4	21 MB
table	public	table2	1.0	1912 kB

The rest of the bloats is less than the 1912kb.

The screenshot below shows the output of du -h of the Supabase Docker file.

I have tried the Vacuum operation as suggested by the documentation, but it only reduce the size storage slightly.

We believe the total space used by Supabase should be at maximum at around 30GB including the space of the Storage objects. Currently I am unsure if this is an issue with Supabase or Proxmox itself. How can investigate this further?

[aantti]

It's probably the logs data. See the following:

• https://github.com/orgs/supabase/discussions/40546
• Self-hosted: pg-wal files grow indefinitely because analytics replication falls behind (see #40546) #43643
• https://github.com/orgs/supabase/discussions/42038
• https://github.com/orgs/supabase/discussions/46084 (logs & analytics are optional now)