Resolved after 30 minutes, on January 26, 2026.
**Summary** We experienced a brief global incident that caused the reviews feed to be unavailable for a small number of users. The issue lasted approximately 30 minutes and has been fully resolved. **Root Cause** A database connection limit was reached due to a MySQL configuration issue. The service user responsible for reply-to-review operations inherited a global connection limit, which was exceeded during a short load spike. This resulted in temporary errors. We are also investigating a possible accumulation of idle database connections that may have increased the likelihood of hitting this limit. **Resolution** The connection limit for the affected service user was increased and applied consistently across all databases. This immediately removed the bottleneck and restored normal operation. **Prevention Measures** To avoid similar incidents in the future, we are: * Explicitly configuring database connection limits per service and environment * Adding monitoring and alerts for per-user database connection usage * Reviewing database connection pooling and autoscaling behavior * Investigating and addressing potential connection leaks in backend services We apologize for the disruption and appreciate your patience while we worked to resolve the issue.
Update timeline
- resolved Jan 28, 2026, 01:56 PM UTC
We experienced a global incident lasting about 30 minutes, during which the reviews feed may have been unavailable for a small number of users. The issue has been resolved, and all services are operating normally.
- postmortem Jan 28, 2026, 01:59 PM UTC
**Summary** We experienced a brief global incident that caused the reviews feed to be unavailable for a small number of users. The issue lasted approximately 30 minutes and has been fully resolved. **Root Cause** A database connection limit was reached due to a MySQL configuration issue. The service user responsible for reply-to-review operations inherited a global connection limit, which was exceeded during a short load spike. This resulted in temporary errors. We are also investigating a possible accumulation of idle database connections that may have increased the likelihood of hitting this limit. **Resolution** The connection limit for the affected service user was increased and applied consistently across all databases. This immediately removed the bottleneck and restored normal operation. **Prevention Measures** To avoid similar incidents in the future, we are: * Explicitly configuring database connection limits per service and environment * Adding monitoring and alerts for per-user database connection usage * Reviewing database connection pooling and autoscaling behavior * Investigating and addressing potential connection leaks in backend services We apologize for the disruption and appreciate your patience while we worked to resolve the issue.
Your product runs on more than AppFollow
Add the vendors you depend on and get one message when any of them reports trouble. Free for 3 vendors, no card.
Amazon Web Services