PLANETSCALE

🟡 PlanetScale's high availability component in us-central1 is having an identity crisis, repeatedly restarting and yeeting Postgres clusters into unplanned failovers. The irony of 'high availability' components causing unavailability is apparently lost on the universe. (Auto-resolved: no updates in 24h)

⏰ May 14, 2026 Fixed (for now) spotted via status_rss

// The official story

There is a incident in PlanetScale GCP us-central1 region impacting Postgres database availability. We are actively investigating and will update as soon as we have more information.

// Status page says

Status: Investigating A high availability component that Postgres clusters depend on is restarting unexpectedly, trigger unplanned failovers in Postgres clusters. Affected components us-central1 (Iowa) (Partial outage)

🔗 Official status page (for the PR-approved version)

https://www.planetscalestatus.com//incidents/01KRM2YSFN92N3BCVH8DB1JF2X

Planetscale's Greatest Hits

🟢 PLANETSCALE 3d ago

🟡 One availability zone in us-east-1 decided to go dark, turning replicas into time travelers and connections into fiction. PlanetScale's status page is doing that thing where 'investigating' means engineers are stress-eating at their desks. (Auto-resolved: no updates in 24h)

🟢 PLANETSCALE Sep 05, 2026

🟡 PlanetScale's Vitess clusters in GCP decided simultaneous ghosting was on brand—Iowa, Montreal, Seoul, and Virginia all taking connectivity timeouts. Investigators report the databases are present but spiritually absent. Schrödinger's data: simultaneously connected and not connected. (Auto-resolved: no updates in 24h)

🟢 PLANETSCALE Jul 24, 2026

🟠 PlanetScale's Query Insights metrics are ghosting users—reporting fewer queries than actually happened. Your databases are fine, but the telescope measuring them needs glasses. They're deploying fixes now, probably after coffee. (Auto-resolved: no updates in 24h)

🟢 PLANETSCALE Jul 18, 2026

🟡 PlanetScale's Northern Virginia servers discovered they don't actually need to accept connections. Engineering team now interrogating the infrastructure about its sudden commitment issues. Remediation in progress, which is consultant-speak for 'we're Googling this.'. (Auto-resolved: no updates in 24h)