← Back to the carnage
SNOWFLAKE
🔴 Snowflake in US West decided to play hide-and-seek with customer data. Engineers are investigating why 'processing efficiency' is code for 'we broke it.' Meanwhile, your queries are getting real acquainted with timeout errors. (Auto-resolved: no updates in 24h)
// The official story
AWS - US West (Oregon): INC0159680
// Status page says
Jun 22 , 21:48 UTC Update - Current status: We are continuing to investigate the underlying source of the issue. In parallel, we are working to improve processing efficiency in a portion of our service infrastructure as a potential mitigation for the increased downstream load affecting related services. Customer experience: Customers hosted in the specified regions may be unable to access or use multiple core
🔗 Official status page (for the PR-approved version)
https://status.snowflake.com/incidents/zn8xxp2qzpr1Snowflake's Greatest Hits
🟢 SNOWFLAKE
Sep 01, 2026
🟠 Snowflake found the bug hiding in plain sight and is slowly letting customers back in like a bouncer checking IDs. Most folks can login again, but they're taking their sweet time rolling out the fix to everyone else. ETA: another hour of watching the status page like it's a reality show. (Auto-resolved: no updates in 24h)
🟢 SNOWFLAKE
Aug 31, 2026
🟠 Snowflake went full 'Elsa had a bad day' across multiple regions, then ungracefully thawed most things back. Snowsight and Streaming still doing their best impression of ghosts. Recovery theater continues. (Auto-resolved: no updates in 24h)
🟢 SNOWFLAKE
Aug 28, 2026
🟠 Snowflake's Snowpark Container Services ghosted an entire region—turns out their cloud provider's capacity evaporated. They're now playing musical chairs with compute resources, prioritizing Business Critical accounts while everyone else watches from the timeout corner. (Auto-resolved: no updates in 24h)
🟢 SNOWFLAKE
Aug 14, 2026
🟡 Snowflake decided to play dead in certain regions—users getting 503s faster than a failed API call. The warehouse is now a very expensive interactive fiction game where the goal is to log in. Engineering team's 30-minute update window: the corporate way of saying 'we're still googling the error code.'. (Auto-resolved: no updates in 24h)