Project: schetopro (ID: ad050ce2-368c-4d7f-8bef-963aabdd1ecb)

Service: api (ID: 6b5f0c68-2d80-4bdf-820c-e1b13feff92f)

Environment: production

Following up on my previous report. Your diagnostic said: "Try redeploying to reschedule onto a healthy builder instance" - I redeployed, but the new deployment (a539ecfe-d6e9-457c-96e8-1fcaa1f25e39) still landed on the same unhealthy builder, builder-zxzvnh, and failed the same way - 2-3 seconds, no output, before Railpack even started:

scheduling build on Metal builder "builder-zxzvnh"

scheduling build on Metal builder "builder-zxzvnh"

That's now 10 consecutive failures over ~1 hour, all pinned to builder-zxzvnh, including the one after your suggested fix. It looks like my service/project may be sticky-assigned to this specific unhealthy builder rather than being load-balanced across the pool - a plain redeploy isn't enough to get us off it.

Failed deployment IDs (chronological): 22633407, 34a5e141, 05103f1a, 71916979, ac298b90, 9f0ffcdc, ee9768a5, 05096ed4, 60a135db, a539ecfe (latest, post-diagnosis).

Could you manually reassign my service to a different builder instance, or otherwise unstick it from builder-zxzvnh? The service is still serving fine on the last successful deploy (5487234f, 2026-07-15 23:26 UTC) - this is only blocking new deploys, not causing a live outage.

Your builds are failing before your code is reached, so this is not caused by anything in your project or configuration. Cancel any queued builds to avoid compounding the issue, then wait 30-60 minutes before retrying. Additional redeploys while the condition persists will land the same way and add to the queue.