Minor impactResolvedCloud

Fly.io incident: Slow image pulls delaying Machine starts in IAD

Fly.io reported this minor-impact incident on its official status page on Oct 30, 2026, 20:24 UTC. It was resolved after 1h 28m.

Right now Fly.io is operational. Live Fly.io status →

🔔 Get alerted when Fly.io has issuesFree · email alerts · no credit card
Oct 30, 2026, 20:24 UTC
started
Oct 30, 2026, 21:52 UTC
resolved
1h 28m
duration
Minor
impact

Timeline

updates from Fly.io's status page ↗
  1. resolvedOct 30, 2026, 21:52 UTC

    Image pull times in IAD have returned to normal.

  2. monitoringOct 30, 2026, 21:23 UTC

    We have implemented a fix and are monitoring results.

  3. identifiedOct 30, 2026, 21:07 UTC

    We have identified the cause of the issue and are working on a fix.

  4. investigatingOct 30, 2026, 20:24 UTC

    We are investigating slow image pulls within Fly Machines in IAD. Newly created or updated Machines are taking longer to start, which may cause deployments to time out. Machines start successfully after the delay. Consider creating Machines in alternate regions, if possible, or running `fly deploy --wait-timeout 10m` to give Machines up to 10 minutes to prepare images and start.

What APIStatus.watch saw

  • Oct 30, 2026, 20:29 UTCoperational → degraded
  • Oct 30, 2026, 21:54 UTCdegraded → operational

Other services with incidents at the same time

Overlapping incidents aren't necessarily related.

How often does Fly.io have incidents?

Fly.io reported 31 incidents in the last 90 days, 15 of them major or critical. A typical incident lasted 1h. See Fly.io's uptime and incident history

← Previous: Elevated issues with the Machines API (Oct 29, 2026)