Increased error rate

All services are operational and error rates have subsided. We’re continuing to monitor our systems.

We are applying recovery steps, and the health of the database is improving. The API is still degraded, and code references and saved dashboards are disabled, but all other services are operational.

We are experiencing issues in our primary DB cluster again, leading to increased API errors and some UI unavailability.

We have finished the recovery steps and we are monitoring the results.

We are continuing in applying recovery steps and some services are recovering. We will continue to monitor the impact on all downstream services.

We are continuing in applying recovery steps and some services are recovering. Audit log has recovered and we will continue to monitor the impact on all downstream services.

We are continuing in applying recovery steps and some services are recovering. We will continue to monitor the impact on all downstream services.

We are continuing in applying recovery steps and some services are recovering. Experimental results have recovered and are no longer delayed. We will continue to monitor the impact on all downstream services.

We are continuing in applying recovery steps and some services are recovering. We will continue to monitor the impact on all downstream services.

We are continuing in applying recovery steps and some services are recovering. We will continue to monitor the impact on all downstream services.

We are still in the process of applying recovery steps and continuing to monitor the impact on all downstream services.

We are beginning to see our primary DB recover, but some of the recovery steps are still in progress. The web UI is available but we are still seeing an impact on downstream services

We are still working to restore the availability of our primary database cluster. We’re still seeing a significant impact across all our services. Experimentation results are delayed currently but we do not anticipate any data loss.

We’re still seeing a significant impact across all our services. Error rates for flag delivery (streaming and polling) are elevated, but oscillating between 1% and 10% failure rates. Most SDK clients connecting to the service should still be receiving correct flag variations.

We are continuing to work on a fix for this issue.

Our primary database cluster is unhealthy. This is causing widespread availability issues with our site, APIs and other downstream services.

We do not anticipate any data loss and are working to restore service as quickly as possible

We are still investigating the issue

We are continuing to investigate this issue.

We are currently investigating increased error rates across LaunchDarkly services.