Salesforce Outage History

Salesforce experienced performance degradation for 1 hour, causing slow performance, timeouts, and connectivity issues while some functionality remained unavailable or ran below optimal levels. Users could still access the service but encountered intermittent performance problems. The incident was investigated and resolved within the 1-hour timeframe.

Salesforce experienced performance degradation for 10 minutes, causing slow performance, timeouts, and connectivity issues while some functionality remained unavailable or suboptimal. Users could still access the service but encountered intermittent performance problems. The incident was resolved after investigation into the root cause.

Salesforce experienced performance degradation affecting document generation functionality across multiple US instances for 4.7 hours. Users encountered intermittent failures, delays, and timeouts when generating PDFs, Word documents, and PowerPoint files, with some finding success by reloading screens. The issue was resolved by isolating a problematic data partition that was causing service instability and implementing optimizations to handle specific data processing patterns that exceeded service limits.

Salesforce experienced performance degradation affecting the AUS6S component for 1.5 hours. Users could access the service but encountered slow performance, timeouts, connectivity issues, and some unavailable functionality. The incident was classified as minor with general and intermittent performance issues throughout the affected period.

A provisioning error during Salesforce's release cycle on February 20, 2026 at 07:00 UTC incorrectly locked a subset of customer environments, preventing users from logging into their Salesforce orgs and accessing business-critical data and tools. Salesforce manually restored access for all affected customers after 119.1 hours and temporarily disabled the faulty provisioning process. A permanent code fix is being deployed to prevent the system from changing org statuses during future provisioning cycles before re-enabling the process.