supabase start can't pull Postgres Docker image

[AFCMS]

Describe the bug
Running supabase start fails to pull the postgres image

To Reproduce

1. supabase login (not sure if it's needed)
2. supabase init
3. supabase start
4. See error

Expected behavior
supabase start should work (last time I used the CLI 6 months ago it worked just fine)

Screenshots

System information
Rerun the failing command with --create-ticket flag.

• Ticket ID: 3352f8df129149c1a2464a0acfb47c69
• Version of OS: Fedora 41
• Version of CLI: 2.1.1 and 2.2.0
• Version of Docker: 27.4.1
• Versions of services:

        SERVICE IMAGE      │      LOCAL       │   LINKED
  ─────────────────────────┼──────────────────┼─────────────
    supabase/postgres      │ 15.8.1.021       │ 15.8.1.021
    supabase/gotrue        │ v2.167.0         │ v2.167.0
    postgrest/postgrest    │ v12.2.3          │ v12.2.3
    supabase/realtime      │ v2.33.70         │ -
    supabase/storage-api   │ v1.14.4          │ v1.14.4
    supabase/edge-runtime  │ v1.66.0          │ -
    supabase/studio        │ 20241202-71e5240 │ -
    supabase/postgres-meta │ v0.84.2          │ -
    supabase/logflare      │ 1.4.0            │ -
    supabase/supavisor     │ 1.1.56           │ -

Additional context
If applicable, add any other context about the problem here.

• Version of Node.js: 20.18.1

[sweatybridge]

There's an issue with the latest docker build. You can workaround it by unpinning the postgres version.

rm supabase/.temp/postgres-version

Apologies for any inconvenience.

[AFCMS]

I don't know if it's pure bad luck but if the workaround fixes the image pulling, supabase start still do not work.

I tried running with --debug but there wasn't any useful information either.

Created a new ticket: 116ae317d8204d53813bfad69b10d57e

[sweatybridge]

Error status 502 sounds like an issue with container health check.

Could you give this a try

pnpx supabase@beta start --ignore-health-check

[AFCMS]

Without the health check it doesn't fail and I can access studio.

But it doesn't seem I can stop having problems lol

I wanted to test auth in the local environment so I setup Discord + GitHub OAuth apps:

[auth.external.github]
enabled = true
client_id = "env(SUPABASE_AUTH_EXTERNAL_GITHUB_ID)"
secret = "env(SUPABASE_AUTH_EXTERNAL_GITHUB_SECRET)"
redirect_uri = ""
url = ""

[auth.external.discord]
enabled = true
client_id = "env(SUPABASE_AUTH_EXTERNAL_DISCORD_ID)"
secret = "env(SUPABASE_AUTH_EXTERNAL_DISCORD_SECRET)"
redirect_uri = ""
url = ""

I have put http://127.0.0.1:54321/auth/v1/callback as callback URL in both apps and when I use the Subabase client to signInWithOAuth every single try fails at the callback level with the browser just staying on the callback "page" having returned 504 with a content like {"code":504,"error_code":"request_timeout","msg":"Processing this request timed out, please retry after a moment.","error_id":"edabed98-3f09-4525-8252-5a74c6786b0f"}.

I can see in studio logs for the auth module that more than half a minute after the 504 error we get another 500 error for some reason (this is the result of one call to signInWithOAuth with the discord provider):

[avallete]

Since the original issue is resolved, and the Oauth issue is mentioned here: #3014

I'm closing this one.