Incident History

Engineering has released a fix and as of 9:10am UTC, customers should no longer experience an issue with AzureAD login. At this time, we are considering this issue resolved.

Loki experienced a read outage from ~21:40-22:00 UTC today. We identified the issue, applied remediation, and continue to monitor for any further disruptions. At this time, we are considering this issue resolved.

The connections have been stable since 14:50 UTC and all 4 probes operating normally during that time. Transitioning this incident from monitoring to resolved.

This incident has been resolved.

Today from 20:15-21:30 UTC a small sub-set of customers in the prod-us-east-0 region could see missing recording rule samples from this time period. This incident is currently resolved.

The incident has been resolved. The cause was one database server being under heavy CPU load - that caused database queries to either take a long time to complete or fail altogether for Grafana Cloud instances using that specific database server. As an outcome, it made some instances unavailable, and it also made new instance startups fail.

This incident has been resolved.

The service has been restored.

A fix has been implemented and we are monitoring the results. All affected instances should now be online and functioning normally.

A fix has been implemented and we are monitoring the results. All affected instances should now be online and functioning normally.