Resolved
Load balancer latency in the AP-South region has returned to normal levels. All health checks are passing and performance metrics are within expected thresholds.
Update โ Investigation
Our team has identified a routing misconfiguration affecting the secondary load balancer pool in AP-South. A fix is being deployed now. ETA for resolution is 30 minutes.
Investigating โ Elevated Latency
We're detecting elevated response times (300โ800ms) on load balancers in the AP-South (Mumbai) region. Our engineering team is actively investigating the root cause. All other regions remain unaffected.
Resolved
Database connection stability in US-East has been fully restored. The connection pooling fix has been applied and monitoring shows normal behavior across all managed database instances.
Update โ Fix Deployed
A patch has been deployed to the connection pooling layer. Database instances are being reconnected automatically. Customers may need to restart their application connections.
Investigating
Several customers reported intermittent connection resets on managed PostgreSQL databases in the US-East region. We're investigating and have escalated to the database engineering team.
Resolved
Cache purge propagation delays have been resolved. All edge nodes are now processing purge requests within the expected 60-second window.
Investigating
CDN cache purge requests are taking longer than expected to propagate across edge nodes. Purges that normally complete in 60 seconds are taking up to 5 minutes. We're working on a fix.
Scheduled Maintenance Window
We will be performing rolling upgrades to the managed database clusters in the EU-West (Ireland) region. During this maintenance window, you may experience brief connection resets (2โ3 seconds per instance). Read replicas will remain available. Failover will not be affected. We recommend performing any batch operations outside this window.
Resolved
Rate limiting thresholds have been restored to normal. All API endpoints are processing requests within expected limits. No further action is required.
Investigating โ 429 Errors
Some API consumers are receiving unexpected 429 (Too Many Requests) errors. Our rate limiting configuration was inadvertently modified during a deployment. We're rolling back the change now.
๐ Get Status Notifications
Subscribe to receive email or SMS alerts when incidents occur or maintenance is scheduled.