Password Auth failing

[sideout100]

Bug report

• [ x] I confirm this is a bug with Supabase, not with my own application.
• [ x] I confirm I have searched the Docs, GitHub Discussions, and Discord.

Describe the bug

Password auth fails with error 28P01 on Transaction Pooler after multiple password resets. Project ref: bifbnytzuiaxyvmusbzi.

To Reproduce

Steps to reproduce the behavior, please provide code snippets or a repository:

1. Go to '…'
2. Click on '…'
3. Scroll down to '…'
4. See error

Expected behavior

restart the Supavisor instance for my project

Screenshots

If applicable, add screenshots to help explain your problem.

System information

• OS: [e.g. macOS, Windows] Windows
• Browser (if applies) [e.g. chrome, safari] Chrome
• Version of supabase-js: [e.g. 6.0.2] Latest
• Version of Node.js: [e.g. 10.10.0] Latest

Additional context

Add any other context about the problem here.

[coderabbitai[bot]]

⚠️ Possible Duplicate Issue(s)

• Transaction pooler rejects password after database password reset (aws-1-us-east-1) #44210

🔗 Related PRs

#46789 - chore(docs): add details and troubleshooting steps for auth schema er… [merged]
#46866 - makes it possible to reset password from connect dialog [merged]

---

📝 Issue Planner

_{Check the box below or use the @coderabbitai plan command to generate an implementation plan and prompts that you can use with your favorite coding assistant.}

• Create Plan

---

🧪 Issue enrichment is currently in open beta.

You can configure auto-planning by selecting labels in the issue_enrichment configuration.

To disable automatic issue enrichment, add the following to your .coderabbit.yaml:

issue_enrichment:
  auto_enrich:
    enabled: false

💬 Have feedback or questions? Drop into our discord!

[Hallidayo]

Seems more an local issue rather than a bug so moved over to a discussion.

[wi5nuu]

Error 28P01 is PostgreSQL's "password authentication failed" code. When it appears specifically on the Transaction Pooler (Supavisor) after multiple password resets, the root cause is almost always a stale credential cache inside Supavisor — not a wrong password on your end.

Why this happens

Supavisor caches the database credentials it uses to open connections to the underlying Postgres instance. When you reset the database password, Supavisor's cached password becomes invalid. It keeps trying to authenticate with the old password until the cache is cleared or the instance is restarted — which is why the error persists even after a successful reset.

Things to try yourself first

1. Pause and resume the project

This is the fastest self-serve way to force Supavisor to restart and pick up the new credentials:

Dashboard → Settings → General → Pause project → Resume project

2. Reset the database password once more after resuming

Go to Dashboard → Settings → Database and reset the password again after the project comes back up. This ensures Supavisor starts fresh with the correct credentials.

3. Verify you are connecting to the correct pooler string

After any password reset, the connection string in Dashboard → Settings → Database → Connection string is updated automatically. Make sure your app is using the latest string — an old hardcoded connection string in a .env file is a common cause of continued 28P01 after a reset.

# Transaction Pooler format (port 6543)
postgresql://postgres.[project-ref]:[NEW-PASSWORD]@aws-0-[region].pooler.supabase.com:6543/postgres

⚠️ Do not mix the Session Pooler (port 5432) and Transaction Pooler (port 6543) connection strings — they authenticate differently and using the wrong one after a reset is another common source of 28P01.

If the above does not fix it

If pausing/resuming and re-entering the password still does not resolve the error, Supavisor may need a manual restart on Supabase's side. You can request this by:

• Opening a support ticket at https://supabase.com/dashboard/support/new and including your project ref (bifbnytzuiaxyvmusbzi)
• Asking in the Supabase Discord #help channel for a faster response

Include your project ref bifbnytzuiaxyvmusbzi in any support request so the team can identify the Supavisor instance immediately.

[SufyExpert]

✅ Solution: Resolving 28P01 Auth Error on Supavisor

Root Cause:
Error 28P01 (Invalid Password) on the Transaction Pooler (port 6543) typically indicates a credential desync. After a password reset, the primary database is updated, but the Supavisor pooling layer may still be caching the legacy credentials.

---

1. Immediate Self-Service Fix

Force a configuration refresh by toggling the pooler:

1. Go to Dashboard > Settings > Database.
2. Find Connection Pooling.
3. Disable the pooler, wait 30 seconds, and Enable it again.
   • This forces Supavisor to fetch the updated postgres password.

2. Requested Maintainer Action

If the toggle does not resolve the sync issue for project bifbnytzuiaxyvmusbzi:

• Action: Manually restart the Supavisor/GoTrue instances associated with this project.
• Reason: Clear the internal cache to allow the new password to propagate to the pooling layer.

---

Technical Context:

• Error Code: 28P01
• Impacted Service: Supavisor (Transaction Pooler)
• Target Project: bifbnytzuiaxyvmusbzi

category: auth status: fix-identified priority: high