Storage stuck in "Coming up…" for 1+ month after restart — project can't reach ACTIVE_HEALTHY; pause/restore disabled (no self-recovery) [ref: abxgeyabzwzrorecsjcd, ap-northeast-2]

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 After a routine restart, every service on my project came back Healthy except Storage. Storage has been stuck in "Coming up…" for over a month, which keeps the whole project pinned in COMING_UP and it never reaches ACTIVE_HEALTHY. Because the project is not in a PAUSED state, the Pause button is disabled and the restore_project API is rejected — so I have no self-recovery path and depend entirely on the infra team. Verified that Storage is the only broken service: - Auth healthy: GET /auth/v1/health → 200 - DB API ready: HEAD /rest-admin/v1/ready → 200; Postgres is reachable (select 1; works) - Storage broken: - health endpoint returns ABORTED REQ - public objects return HTTP 544, e.g. GET /storage/v1/object/public/asset/kakao_login_logo.png - storage.buckets and storage.objects are both empty (0 rows) via SQL → the storage catalog appears lost / reset Possibly a wider reset (please confirm what the backend sees): via SQL the public schema has 0 application tables, supabase_migrations.schema_migrations does not exist, and auth.users has only 1 row left. Support ticket SU-426011 was opened ~2.5 weeks ago (account linked as requested) with no human reply, and no confirmation it reached the infra queue. To Reproduce 1. Have a Free-plan project running normally in ap-northeast-2. 2. Restart the project. 3. All services return Healthy except Storage, which stays in "Coming up…". 4. The project stays in COMING_UP indefinitely (1+ month) and never reaches ACTIVE_HEALTHY. 5. Try to recover: Pause is disabled and restore_project is rejected because the state is COMING_UP, not PAUSED → no self-service recovery. 6. Storage requests fail: health = ABORTED REQ, public objects = HTTP 544; storage.buckets/storage.objects are empty in SQL. Expected behavior - Storage should finish bringing up so the pr…