2023-03-08 Incident: Infrastructure connectivity issue affecting multiple regions

On March 8-10, 2023, Datadog experienced a multi-regional outage affecting platform access, monitoring, and data ingestion. The incident was triggered by an automatic systemd update that interfered with network routing, impacting tens of thousands of nodes. Datadog resolved the issue by disabling the problematic update channel, restoring compute capacity, and backfilling historical data, utilizing a large incident response team and extensive communication with customers.