Critical Bug - `supabase db start` failing in CI entirely

[abdulrabbani00]

Describe the bug
When I run a supabase db start in CI I get this error:

Seeding globals from roles.sql...
Applying migration 20250404200456_migrate-from-timescale.sql...
Stopping containers...
ERROR: must be owner of table users (SQLSTATE 42501)
At statement 1:                                     
-- Enable RLS                                       
ALTER TABLE auth.users ENABLE ROW LEVEL SECURITY 

Nowhere in my migrations am I running this which leads me to believe its coming from the roles.sql.

My pipeline, which was running perfectly fine 2 days ago started failing today out of nowhere. I believe if I am having this issue its going to be happening for others and this seems like a critical issue that likely requires some rollback.

If there is a way I can pin the docker version that the supabase DB uses that would be great on the CLI slide. I tried downgrading CLI Versions but that didnt help.

The version being used is: ghcr.io/supabase/postgres:15.8.1.085 - Locally I am running public.ecr.aws/supabase/postgres:15.8.1.05 and it works perfectly fine.

To Reproduce
Steps to reproduce the behavior:

1. Run an existing pipeline.
2. Perhaps don't init a pipeline, this might be working fine on fresh pipelines but perhaps having migrations afterwards is whats breaking this.
3. Code I am using

 - uses: supabase/setup-cli@v1
        with:
          version: 2.22.16
      
      - name: Start Supabase
        run: supabase db start

Expected behavior
I expect supabase to run its core migrations without a problem in CI. This works perfectly fine locally.

Screenshots
If applicable, add screenshots to help explain your problem.

System information
The crash ID: 8bf6bf1a5c0c4bc69e14ae8ee351afc3

Additional context
Running on ubuntu latest.

[abdulrabbani00]

My bad, it was an ID 10 T error.