Incident report for October 9 and 10, 2026 - Cross-data-center network and database connection failures
Blog post from Inngest
Inngest reported two service disruptions on October 9 and 10, 2026, both linked to its reliance on connectivity among three data centers and a central database. On October 9, a provider network outage, reportedly related to a DDoS-protection configuration change, isolated one data center from external services and other facilities, halting affected queue processing and creating a shared event-pipeline backlog that delayed execution for most accounts until connectivity returned. On October 10, PgBouncer database connection poolers twice lost their usable database connections after short client deadlines and query cancellations caused connections to be closed faster than they could be recreated, with cron load spikes and newly deployed lookups worsening the issue. A rerouting mitigation restored processing but introduced intermittent delays when network address translation gateways silently dropped connections. Some runs affected in both events remained permanently marked as Running because internal failures consumed retries without correctly transitioning run status. Inngest has changed database timeout behavior, updated run-failure handling, improved gateway routing and incident communications, and is migrating toward a single-data-center cell architecture to remove cross-data-center dependencies.
No tracked trend matches for this post yet.
Use this post, company, and trend context to find content marketing opportunities, perform competitive analysis, or address product feature gaps via the Plushcap MCP server or the Plushcap API.