• [31805] Decreased the number of Pushgateway replicas to one to fix the issue with metrics being split into two pieces without replica control because Pushgateway does not support clustering.

  • [31655] Enhanced the StackLight upgrade Jenkins pipeline job by changing the Elasticsearch timeout from five minutes to two hours to avoid the cases when Elasticsearch has not enough time to become green after an update.

  • [31032] Fixed the permissions wrongly set for the Prometheus configuration directory.

  • [31454] Fixed the issue with the Distribution of PGs per OSD panel of the Ceph OSD overview Grafana dashboard displaying no Ceph OSDs because of a query timeout. Also, updated the Distribution of PGs per OSD panel to display the data in bars instead of lines.

  • [31241] Fixed the logs severity in the raise condition of the RabbitmqErrorLogsTooHigh alert.

  • [31017] Fixed the calculation for the API availability metrics.

  • [31586] Fixed the issue with the Prometheus formula failing to run with Salesforce notifications enabled.