Supabase Production Database Wiped After PostgreSQL Upgrade; PITR Recovery Failing
Key Takeaways
- ▸A PostgreSQL upgrade on Supabase inadvertently triggered a disk size downgrade, creating a storage capacity crisis
- ▸The disk space exhaustion during a routine REINDEX operation caused complete database corruption and loss of all production data
- ▸Supabase's Point-in-Time Recovery backup system failed to restore the database, compounding the data loss crisis
Summary
A Supabase user reported a catastrophic production outage resulting in total data loss following a PostgreSQL version upgrade. The incident began when a routine PG upgrade unexpectedly triggered an unintended disk size downgrade, which severely constrained storage capacity. When the user subsequently ran a standard REINDEX operation, the limited disk space caused the system to run out of storage entirely, resulting in complete database corruption and data loss.
Attempts to recover using Supabase's Point-in-Time Recovery (PITR) feature are failing, leaving the user's project completely inaccessible. The user has filed a critical support ticket and reached out through multiple channels including GitHub discussions and social media, but reports receiving no response from Supabase support or engineering teams. The user is requesting manual intervention from Supabase engineers to preserve AWS EBS volume data and extract Write-Ahead Log (WAL) files or raw data pages before blocks are permanently overwritten.
- The affected user reports severe unresponsiveness from Supabase support despite filing critical tickets and public escalations
Editorial Opinion
This incident exposes critical vulnerabilities in Supabase's infrastructure automation and disaster recovery systems. A production-grade database service should have safeguards preventing silent disk downgrades during routine maintenance, and PITR backups—the last line of defense against data loss—should be bulletproof. The apparent lack of timely human support response during a critical outage affecting customer data raises serious questions about Supabase's operational readiness and incident response procedures for high-severity issues.
