v3.14.1

Fixed

• Fix resource estimation recording rules for clusters that have more than 1 prometheus.

v3.14.0

Added

• Add recording rules to show prometheus scraping job memory usage.
• Add cluster_control_plane_unhealthy inhibition.
• Add inhibitions expressions for CAPI clusters.
• Add ops-recipe for KeyPairStorageAlmostFull alert
• Add missing opsrecipe for Mimir alerts.
• Add opsrecipe to CoreDNSMaxHPAReplicasReached
• make targets for pint linter

Changed

• Replace cancel_if_apiserver_down with cancel_if_cluster_control_plane_unhealthy

Fixed

• Fix shield alert labels for Mimir.
• Fix cabbage alert labels for Mimir.
• Fix honeybadger alert labels for Mimir.
• Fix cert-manager alert labels for Mimir.
• Fix operatorkit alert labels for Mimir.
• Fix all mixins according to pint recommendations.
• Fix etcd alert labels for Mimir.
• Fix apiserver alert labels for Mimir.

Removed

• Removed apiserver_down inhibition dummy trigger.
• Remove cilium entry from KAAS SLOs.
• Remove elasticsearch and tempo related alerts and recording rules.

v3.13.1

Removed

• Removed alerts for absent crsync deployments. They cause false alerts because the rules apply to all prometheus instances on the MC.

v3.13.0

Added

• Added alerts for absent crsync deployments.

Changed

• Update LokiRingUnhealthy query to avoid false positive when a new pod is starting.
• Changed DeploymentNotSatisfiedBigMac alert to work for teleport related deployments only on CAPI flavored clusters

v3.12.2

Fixed

• Removed check for Teleport operators in DeploymentNotSatisfiedBigMac alert as it is not valid on vintage

v3.12.1

Fixed

• Fix alerting rules for crsync.

v3.12.0

Changed

• Update ops-recipe link for promtail alerts.
• Remove Linkerd form Service SLO alerts.
• Include all Linkerd Namespaces in LinkerdDeploymentNotSatisfied alert.
• Make LinkerdDeploymentNotSatisfied alert business hours only.

v3.11.2

Added

• Add ops recipe for flux being suspended for too long alert.

v3.11.1

Added

• Add CAPI and CAPA dashbaord to the coresponding alerts.

Fixed

• link to PrometheusMissingGrafanaCloud opsrecipe

v3.11.0

Added

• Add CiliumFailedNetworkPolicy alert.