--- title: Prometheus Metrics Reference pubDate: 2026-08-17T09:53:44.266Z antora: editUrl: https://github.com/couchbase/docs-operator/edit/release/2.9/modules/ROOT/pages/reference-prometheus-metrics.adoc xref: xref:operator::reference-prometheus-metrics.adoc[] --- [Consult the llms.txt file for a full list of contents](/llms.txt) [View original HTML](/operator/current/reference-prometheus-metrics.html) # Prometheus Metrics Reference > This page captures the metrics supplied to Prometheus by the Couchbase Kubernetes Operator and links reference pages of a number of additional metrics that are exported by third party libraries. ## [](#operator-metrics)Operator Metrics | Metric | Type | Unit | Labels | Optional Labels | Stability | Added | | ------------------------------------------------------------------------------------------------------------------------------------------ | --------- | ------------ | ----------------------------- | --------------------------- | --------- | ----- | | backup\_jobs\_created\_total Total number of backup jobs that have been created by the operator | counter | | namespace,backup\_type | cluster\_uuid,cluster\_name | committed | 2.8.0 | | cluster\_manual\_intervention Indicates whether manual intervention is required for the cluster | gauge | | namespace,name | cluster\_uuid,cluster\_name | committed | 2.9.0 | | cpu\_under\_management Total cpu requests for operator managed pods in k8s cpu units | gauge | | namespace,name | cluster\_uuid,cluster\_name | committed | 2.8.0 | | in\_place\_upgrade\_failures The number of times in place upgrades have failed | counter | | name | cluster\_uuid,cluster\_name | committed | 2.7.0 | | in\_place\_upgrades\_total Total number of in place upgrades performed by operator | counter | | name | cluster\_uuid,cluster\_name | committed | 2.7.0 | | kubernetes\_api\_request\_failures Total failed requests to the Kubernetes API by the operator | counter | | method,host | | committed | 2.8.0 | | kubernetes\_api\_requests\_time\_milliseconds Length of time per request to the Kubernetes API | histogram | milliseconds | method,host | | committed | 2.8.0 | | kubernetes\_api\_requests\_total Total requests made to the Kubernetes API by the operator | counter | | method,host | | committed | 2.8.0 | | memory\_under\_management\_bytes Total memory requests for operator managed pods in bytes | gauge | bytes | namespace,name | cluster\_uuid,cluster\_name | committed | 2.8.0 | | pod\_readiness\_duration The time it takes for a pod to enter a ready state | gauge | milliseconds | name,serverClass | cluster\_uuid,cluster\_name | committed | 2.7.0 | | pod\_recoveries\_total Total number of times operator has recovered a pod when the pod has been down | counter | | name,podName | cluster\_uuid,cluster\_name | committed | 2.7.0 | | pod\_recovery\_failures\_total Total number of times operator has failed to recover a pod | counter | | name,podName | cluster\_uuid,cluster\_name | committed | 2.7.0 | | pod\_replacements\_failed Total number of times pods have failed to be recovered by the operator | counter | | name | cluster\_uuid,cluster\_name | committed | 2.7.0 | | pod\_replacements\_total The amount of times operator has replaced a couchbase server pod due to a change in a couchbase cluster resources | counter | | name | cluster\_uuid,cluster\_name | committed | 2.7.0 | | reconcile\_failures Total failed reconcile operations performed on a specific cluster | counter | | namespace,name | cluster\_uuid,cluster\_name | committed | 2.3.0 | | reconcile\_time\_seconds Length of time per reconcile for a specific cluster | histogram | seconds | namespace,name | cluster\_uuid,cluster\_name | committed | 2.3.0 | | reconcile\_total Total reconcile operations performed on a specific cluster | counter | | namespace,name,result | cluster\_uuid,cluster\_name | committed | 2.3.0 | | server\_http\_request\_codes\_total Total HTTP requests to Couchbase Server for a specific cluster, method and status code returned | counter | | name,method,code,service,host | name,namespace | committed | 2.3.0 | | server\_http\_request\_failures Total failed HTTP requests to Couchbase Server for a specific cluster | counter | | name,method,service,host | name,namespace | committed | 2.3.0 | | server\_http\_requests\_time\_milliseconds Length of time per request for a specific cluster | histogram | milliseconds | name,method,service,host | name,namespace | committed | 2.3.0 | | server\_http\_requests\_total Total HTTP requests to Couchbase Server for a specific cluster | counter | | name,method,service,host | name,namespace | committed | 2.3.0 | | swap\_rebalance\_failures Total number of times swap rebalances have failed | counter | | name | cluster\_uuid,cluster\_name | committed | 2.7.0 | | swap\_rebalances\_total Total number of swap rebalances performed by the operator | counter | | name | cluster\_uuid,cluster\_name | committed | 2.7.0 | | upgrade\_duration The time taken to perform an upgrade | | milliseconds | name | cluster\_uuid,cluster\_name | committed | 2.7.0 | | volume\_expansions\_total Total number of times the size of volumes have been increased under management | counter | | name,volumeName | cluster\_uuid,cluster\_name | committed | 2.7.0 | | volume\_size\_under\_management\_bytes Total memory claimed by volumes under management by the operator in bytes | gauge | bytes | namespace,name | cluster\_uuid,cluster\_name | committed | 2.8.0 | ## [](#additional-metrics)Additional Metrics * [Kubebuilder](https://book.kubebuilder.io/reference/metrics-reference)