Karpenter integration metrics
The official Datadog integration for monitoring Karpenter. It collects 210 metrics.
Official summary
Monitor the health and performance of Karpenter
Metric types
- count (occurrences in the interval): 124
- gauge (value at a point in time): 86
Metrics
Descriptions are the official definitions, unedited.
| Metric | Type | Unit |
|---|---|---|
karpenter.The total number of AWS SDK Go requests | count | - |
karpenter.Latency of AWS SDK Go requests histogram buckets | count | - |
karpenter.Count of AWS SDK Go request durations | count | - |
karpenter.Sum of AWS SDK Go request durations | count | second |
karpenter.The total number of AWS SDK Go request attempts | count | - |
karpenter.Latency of AWS SDK Go request attempts histogram buckets | count | second |
karpenter.Count of AWS SDK Go request attempt durations | count | request |
karpenter.Sum of AWS SDK Go request attempt durations | count | second |
karpenter.A metric with a constant '1' value labeled by version from which Karpenter was built. | gauge | - |
karpenter.The count of certificate reads | count | read |
karpenter.The count of certificate read errors | count | error |
karpenter.The count of client-go requests | count | request |
karpenter.Histogram buckets for client-go request durations | count | - |
karpenter.Count of client-go request durations | count | - |
karpenter.Sum of client-go request durations | count | second |
karpenter.The count of observation in the batching window histogram by upper_bound buckets | count | - |
karpenter.The count of observation in the batching window histogram | count | - |
karpenter.The sum of the duration of the batching window per batcher | count | second |
karpenter.The count of observation in the request batch histogram by upper_bound buckets | count | - |
karpenter.The count of observation in the request batch histogram | count | - |
karpenter.The sum of the size of the request batch per batcher | count | - |
karpenter.The count of observations in the duration of cloud provider histogram by upper_bound buckets,method name and provider | count | - |
karpenter.The count of observations in the duration of cloud provider histogram | count | - |
karpenter.The sum of the duration of cloud provider method calls. Labeled by the controller | count | second |
karpenter.The count of errors returned from CloudProvider calls | count | error |
karpenter.VCPUs cores for a given instance type | gauge | core |
karpenter.Memory, in bytes, for a given instance type | gauge | byte |
karpenter.Instance type offering availability, based on instance type, capacity type, and zone | gauge | - |
karpenter.Estimated hourly price used when making informed decisions on node cost calculation. This is updated once on startup and then every 12 hours | gauge | - |
karpenter.Utilization of allocatable resources by pod requests | gauge | percent |
karpenter.Current count of nodes in cluster state. | gauge | node |
karpenter.Returns 1 if cluster state is synced and 0 otherwise. Synced checks that nodeclaims and nodes that are stored in the APIServer have the same representation as Karpenter's cluster state | gauge | - |
karpenter.The time for which cluster state is not synced | gauge | second |
karpenter.Number of consistency checks that have failed | gauge | error |
karpenter.Number of currently used workers per controller | gauge | worker |
karpenter.Maximum number of concurrent reconciles per controller | gauge | - |
karpenter.The count of reconciliations per controller | count | - |
karpenter.The count of observations in the reconciliation per controller histogram by upper_bound buckets | count | - |
karpenter.The count of observations in the reconciliation per controller histogram | count | - |
karpenter.The sum of time per reconciliation per controller | count | second |
karpenter.The count of reconciliation errors per controller | count | error |
karpenter.Total number of reconciliation panics per controller | count | - |
karpenter.Total number of terminal reconciliation errors per controller | count | - |
karpenter.Total number of conversion webhook panics | count | - |
karpenter.The count of deprovisioning actions performed. Labeled by deprovisioner (Deprecated in v1.0+) | count | execution |
karpenter.Number of times the Consolidation algorithm has reached a timeout. Labeled by consolidation type (Deprecated in v1.0+) | gauge | timeout |
karpenter.Number of machines eligible for deprovisioning by Karpenter. Labeled by deprovisioner (Deprecated in v1.0+) | gauge | - |
karpenter.The count of observations in the deprovisioning evaluation histogram by upper_bound buckets (Deprecated in v1.0+) | count | - |
karpenter.The count of observations in the deprovisioning evaluation histogram (Deprecated in v1.0+) | count | - |
karpenter.The sum of the duration of the deprovisioning evaluation process in seconds (Deprecated in v1.0+) | count | second |
karpenter.The count of the observation in the replacement machine histogram by upper_bound buckets (Deprecated in v1.0+) | count | - |
karpenter.The count of the observation in the replacement machine histogram (Deprecated in v1.0+) | count | - |
karpenter.The sum of the time required for a replacement machine to become initialized (Deprecated in v1.0+) | count | second |
karpenter.The count of times that Karpenter failed to launch a replacement node for deprovisioning. Labeled by deprovisioner (Deprecated in v1.0+) | count | attempt |
karpenter.The count of disruption actions performed. Labeled by disruption method | count | execution |
karpenter.The number of nodes for a given NodePool that can be disrupted at a point in time. Labeled by NodePool. Note that allowed disruptions can change very rapidly, as new nodes may be created and others may be deleted at any point. | gauge | node |
karpenter.The count of times the Consolidation algorithm has reached a timeout. Labeled by consolidation type | count | timeout |
karpenter.Number of nodes eligible for disruption by Karpenter. Labeled by disruption method | gauge | node |
karpenter.The count of observations in the disruption evaluation histogram by upper_bound buckets | count | - |
karpenter.The count of observations in the disruption evaluation histogram | count | - |
karpenter.The sum of the duration of the disruption evaluation process in seconds | count | second |
karpenter.Total number of nodes disrupted. Labeled by NodePool, disruption action, method, and consolidation type. | count | node |
karpenter.Total number of reschedulable pods disrupted on nodes. Labeled by NodePool, disruption action, method, and consolidation type. | count | - |
karpenter.The number of commands that are being waited on in the disruption orchestration queue. | gauge | command |
karpenter.The number of times that Karpenter failed to launch a replacement node for disruption. Labeled by disruption method | count | attempt |
karpenter.The count of observations in the replacement nodeclaim histogram by upper_bound buckets | count | - |
karpenter.The count of observations in the replacement nodeclaim histogram | count | - |
karpenter.The sum of the amount of time required for a replacement nodeclaim to become initialized | count | second |
karpenter.The summary count of garbage collection cycles in the Karpenter instance | count | - |
karpenter.The pause duration of garbage collection cycles in the Karpenter instance by quantile | gauge | - |
karpenter.The sum of the pause duration of garbage collection cycles in the Karpenter instance | count | second |
karpenter.Number of bytes allocated and still in use | gauge | byte |
karpenter.Count of bytes allocated, even if freed. | count | byte |
karpenter.Number of bytes used by the profiling bucket hash table | gauge | byte |
karpenter.The count of frees | count | - |
karpenter.Number of bytes used for garbage collection system metadata | gauge | byte |
karpenter.Number of heap bytes allocated and still in use | gauge | byte |
karpenter.Number of heap bytes waiting to be used | gauge | byte |
karpenter.Number of heap bytes that are in use | gauge | byte |
karpenter.Number of allocated objects | gauge | object |
karpenter.Number of heap bytes released to OS | gauge | byte |
karpenter.Number of heap bytes obtained from system | gauge | byte |
karpenter.Number of seconds since 1970 of last garbage collection | gauge | second |
karpenter.The count of pointer lookups | count | - |
karpenter.The count of mallocs | count | - |
karpenter.Number of bytes in use by mcache structures | gauge | byte |
karpenter.Number of bytes used for mcache structures obtained from system | gauge | byte |
karpenter.Number of bytes in use by mspan structures | gauge | byte |
karpenter.Number of bytes used for mspan structures obtained from system | gauge | byte |
karpenter.Number of heap bytes when next garbage collection will take place | gauge | byte |
karpenter.Number of bytes used for other system allocations | gauge | byte |
karpenter.Number of bytes in use by the stack allocator | gauge | byte |
karpenter.Number of bytes obtained from system for stack allocator | gauge | byte |
karpenter.Number of bytes obtained from system | gauge | byte |
karpenter.Number of goroutines that currently exist | gauge | - |
karpenter.Information about the Go environment | gauge | - |
karpenter.Number of OS threads created | gauge | thread |
karpenter.The count of notification actions performed. Labeled by action | count | execution |
karpenter.The count of messages deleted from the SQS queue | count | message |
karpenter.The count of observations in the interruption message latency histogram by upper_bound buckets | count | - |
karpenter.The count of observations in the interruption message latency histogram | count | - |
karpenter.The sum of the length of time between message creation in queue and an action taken on the message by the controller | count | second |
karpenter.The count of messages received from the SQS queue. Broken down by message type and whether the message was actionable | count | message |
karpenter.Gauge of if the reporting system is master of the relevant lease, 0 indicates backup, 1 indicates master. 'name' is the string used to identify the lease. | gauge | - |
karpenter.The count of machines created in total by Karpenter. Labeled by reason the machine was created and the owning provisioner | count | - |
karpenter.The count of machines disrupted in total by Karpenter. Labeled by disruption type of the machine and the owning provisioner | count | - |
karpenter.The count of machine drifted reasons in total by Karpenter. Labeled by drift type of the machine and the owning provisioner | count | - |
karpenter.The count of machines initialized in total by Karpenter. Labeled by the owning provisioner | count | - |
karpenter.The count of machines launched in total by Karpenter. Labeled by the owning provisioner | count | - |
karpenter.The count of machines registered in total by Karpenter. Labeled by the owning provisioner | count | - |
karpenter.The count of machines terminated in total by Karpenter. Labeled by reason the machine was terminated and the owning provisioner | count | - |
karpenter.Number of nodeclaims created in total by Karpenter. Labeled by reason the nodeclaim was created and the owning nodepool | gauge | - |
karpenter.Number of nodeclaims disrupted in total by Karpenter. Labeled by disruption type of the nodeclaim and the owning nodepool | gauge | - |
karpenter.Number of nodeclaims drifted reasons in total by Karpenter. Labeled by drift type of the nodeclaim and the owning nodepool | gauge | - |
karpenter.Number of nodeclaims initialized in total by Karpenter. Labeled by the owning nodepool | gauge | - |
karpenter.Histogram buckets for CloudProvider Instance termination duration | count | - |
karpenter.Count of CloudProvider Instance termination observations | count | - |
karpenter.Sum of CloudProvider Instance termination durations | count | second |
karpenter.Number of nodeclaims launched in total by Karpenter. Labeled by the owning nodepool | gauge | - |
karpenter.Number of nodeclaims registered in total by Karpenter. Labeled by the owning nodepool | gauge | - |
karpenter.Number of nodeclaims terminated in total by Karpenter. Labeled by reason the nodeclaim was terminated and the owning nodepool | gauge | - |
karpenter.Histogram buckets for NodeClaim termination duration | count | - |
karpenter.Count of NodeClaim termination duration observations | count | - |
karpenter.Sum of NodeClaim termination durations | count | second |
karpenter.The nodepool limits are the limits specified on the provisioner that restrict the quantity of resources provisioned. Labeled by nodepool name and resource type | gauge | - |
karpenter.The nodepool usage is the amount of resources that have been provisioned by a particular nodepool. Labeled by nodepool name and resource type | gauge | - |
karpenter.The amount of resources allocatable by nodes | gauge | - |
karpenter.The count of nodes created in total by Karpenter. Labeled by owning provisioner | count | node |
karpenter.The number of pods that are waiting for a successful eviction in the eviction queue. | gauge | - |
karpenter.The count of deleted leaked leases | count | - |
karpenter.The resources reserved for system overhead, the difference between the nodes capacity and allocatable values are reported by the status. | gauge | - |
karpenter.The count of nodes terminated in total by Karpenter. Labeled by owning provisioner | count | node |
karpenter.The count of observations in the nodes termination time seconds summary | count | - |
karpenter.The time taken between a nodes deletion request and the removal of its finalizer by quantile | gauge | - |
karpenter.The sum of the time taken between a nodes deletion request and the removal of its finalizer | count | second |
karpenter.Total resources specified by DaemonSet pod limits | gauge | - |
karpenter.Total resources requested by DaemonSet pods | gauge | - |
karpenter.Total pod resources specified by non-DaemonSet pod limits | gauge | - |
karpenter.Total pod resources requested by non-DaemonSet pods bound | gauge | - |
karpenter.Current lifetime of nodes in seconds | gauge | second |
karpenter.Count of nodes drained | count | node |
karpenter.Count of node eviction requests | count | request |
karpenter.Histogram buckets for node lifetime durations | count | - |
karpenter.Count of node lifetime durations | count | - |
karpenter.Sum of node lifetime durations | count | second |
karpenter.Time current status condition has been active for ec2nodeclass | gauge | second |
karpenter.Count of status condition transitions for ec2nodeclass | count | - |
karpenter.Number of conditions for ec2nodeclass | gauge | - |
karpenter.Time current status condition has been active for node | gauge | second |
karpenter.Count of status condition transitions for node | count | - |
karpenter.Histogram of condition state durations for node | count | second |
karpenter.Number of conditions for node | gauge | - |
karpenter.Histogram buckets for node termination durations | count | second |
karpenter.Time current status condition has been active for nodeclaim | gauge | second |
karpenter.Count of status condition transitions for nodeclaim | count | - |
karpenter.Histogram of condition state durations for nodeclaim | count | second |
karpenter.Number of conditions for nodeclaim | gauge | - |
karpenter.Histogram buckets for nodeclaim termination durations | count | - |
karpenter.Time current status condition has been active for nodepool | gauge | second |
karpenter.Count of status condition transitions for nodepool | count | - |
karpenter.Number of conditions for nodepool | gauge | - |
karpenter.Histogram buckets for ec2nodeclass status condition transitions | count | - |
karpenter.Count of ec2nodeclass status condition transitions | count | - |
karpenter.Sum of ec2nodeclass status condition transitions | count | - |
karpenter.Current time for ec2nodeclass termination | gauge | second |
karpenter.Histogram buckets for ec2nodeclass termination durations | count | - |
karpenter.Count of ec2nodeclass termination durations | count | - |
karpenter.Sum of ec2nodeclass termination durations | count | - |
karpenter.Current time for node termination | gauge | second |
karpenter.Current time for nodeclaim termination | gauge | second |
karpenter.Histogram buckets for nodepool status condition transitions | count | - |
karpenter.Count of nodepool status condition transitions | count | - |
karpenter.Sum of nodepool status condition transitions | count | - |
karpenter.Current time for nodepool termination | gauge | second |
karpenter.Histogram buckets for nodepool termination durations | count | - |
karpenter.Count of nodepool termination durations | count | - |
karpenter.Sum of nodepool termination durations | count | - |
karpenter.The count of the observations in the pod startup summary | count | - |
karpenter.The time taken between pod creation and the pod being in a running state by quantile | gauge | - |
karpenter.The sum of the time from pod creation and the pod being in a running state | count | second |
karpenter.Pod state is the current state of pods. This metric can be used several ways as it is labeled by the pod name, namespace, owner, node, provisioner name, zone, architecture, capacity type, instance type and pod phase. | gauge | - |
karpenter.Total user and system CPU time spent in seconds | count | second |
karpenter.Maximum number of open file descriptors | gauge | - |
karpenter.Number of open file descriptors | gauge | - |
karpenter.Resident memory size in bytes | gauge | byte |
karpenter.Start time of the process since unix epoch in seconds | gauge | second |
karpenter.Maximum amount of virtual memory available in bytes | gauge | byte |
karpenter.Virtual memory size in bytes | gauge | byte |
karpenter.The limits specified on the provisioner that restrict the quantity of resources provisioned. Labeled by provisioner name and resource type | gauge | - |
karpenter.The count of observations in the provisioner scheduling histogram by upper_bound buckets | count | - |
karpenter.The count of observations in the provisioner scheduling histogram | count | - |
karpenter.The sum of the duration of scheduling process in seconds. Broken down by provisioner and error | count | second |
karpenter.The number of pods that are waiting to be scheduled. | gauge | - |
karpenter.The count of observations in the provisioner scheduling simulation histogram by upper_bound buckets | count | - |
karpenter.The count of observations in the provisioner scheduling simulation histogram | count | - |
karpenter.The sum of the duration of scheduling simulations used for deprovisioning and provisioning in seconds | count | second |
karpenter.The amount of resources that have been provisioned by a particular provisioner. Labeled by provisioner name and resource type | gauge | - |
karpenter.The percentage of each resource used based on the resources provisioned and the limits that have been configured in the range [0,100]. Labeled by provisioner name and resource type | gauge | percent |
karpenter.Count of HTTP requests, partitioned by status code, method, and host. | count | request |
karpenter.The amount of seconds the longest running processor for workqueue been running | gauge | second |
karpenter.The count of observations in the workqueue queue duration histogram by upper_bound buckets | count | - |
karpenter.The count of observations in the workqueue queue duration histogram | count | - |
karpenter.The sum of the duration of how long in seconds an item stays in workqueue before being requested | count | second |
karpenter.The amount of seconds of work that has been done that is in progress and hasn't been observed by work_duration. Large values indicate stuck threads. One can deduce the number of stuck threads by observing the rate at which this increases | gauge | - |
karpenter.The count of observations in the workqueue work duration histogram by upper_bound buckets | count | - |
karpenter.The count of observations in the workqueue work duration histogram | count | - |
karpenter.The sum of the amount of seconds spent processing an item from workqueue takes | count | second |
karpenter.The count of adds handled by workqueue | count | - |
karpenter.Current depth of workqueue | gauge | - |
karpenter.The count of retries handled by workqueue | count | attempt |
Official documentation
Metric names and descriptions are the official Datadog integration definitions (Copyright Datadog, Inc., BSD-3-Clause), reproduced unedited. The full license terms are on the About page.