Project instance dead for 6+ days (522 on all requests, pause fails, restarts don't help) — ticket SU-421252

[monitoringdarek]

My free-plan project has been completely unreachable since ~July 14 and I've run out of options on my side. Hoping someone from the team can escalate.

Project ref: ktfjdngmvrnqkzjxvzoc

Symptoms:

• Every request to the project returns Cloudflare error 522 (example Ray ID: a1e3b228290dbbe4, 2026-07-20 17:31 UTC)
• Dashboard Table Editor / SQL Editor: "Failed to run sql query: Connection terminated due to connection timeout"
• API Gateway logs show even Supabase's own infra requests failing: POST /admin/v1/walg/check-async-backup → 522 and POST /admin/v1/network-bans/retrieve → 522 (source: @supabase-infra)

What I've tried:

• Multiple restarts — no effect
• Pausing the project — fails with "Failed to verify backup status before pausing"

• Checked usage — quotas not exceeded (DB size 27 MB / 500 MB, egress 8%)

Support ticket: SU-421252 (opened July 14, first reply suggested compute resources / slow DB, which doesn't match — the instance is dead, not slow). Today I re-submitted via the dashboard support form with logs attached.

This is a production project. Is there anything else I can do, or can someone flag this instance for manual recovery? Thanks!

[GaryAustin1]

When you say "first reply" was this from a support person?
What do you mean you resubmitted via dashboard how did you submit first ticket?

Have you stopped your project from hitting the API? Could be disk I/O if still being hit.

Supabase does not usually monitor Github for project issues and has to be done thru support.

[monitoringdarek]

Thank you for your response.

The first response came from the official Supabase Support team after I submitted ticket SU-421252 through the support form in the Supabase Dashboard. I did not submit it through the API.

I tried restarting the project several times from the Dashboard, but it had no effect. I also tried to pause the project, but the operation failed with the following message:

“Failed to pause project: Failed to verify backup status before pausing.”

Therefore, I was unable to complete the pause and restore procedure. The project has been inaccessible for more than six days, and requests return 522 errors. The usage limits have not been exceeded.

An I/O or storage-related issue seems possible, especially because the system cannot verify the backup status. I have contacted Supabase Support again and asked them to escalate the case or perform a manual recovery.

This is a production project, so any additional advice would be greatly appreciated.

[GaryAustin1]

If your app is constantly retrying then you may never get enough disk i/o to start up. But assuming that is not the case then sounds like support is the only path to go thru.

[monitoringdarek]

Thank you. The application may have been retrying Supabase requests automatically. I will stop the Garmin synchronization container and any other services connected to the project, wait for a while, and then try restarting the project once.

If this does not help, I will continue the case with Supabase Support under ticket SU-421252.

[brhymnmyr-creator]

Hello please help me