metrics

package
v0.15.6 Latest Latest
Warning

This package is not in the latest version of its module.

 Go to latest Published: Nov 29, 2024 License: Apache-2.0 Imports: 16 Imported by: 0

Documentation

Index

Constants

View Source 
const MetricPrefix = "armada_"

Variables

View Source 
var AllDescs = []*prometheus.Desc{
	QueueSizeDesc,
	QueuePriorityDesc,
	QueueResourcesDesc,
	MinQueueResourcesDesc,
	MaxQueueResourcesDesc,
	MedianQueueResourcesDesc,
	CountQueueResourcesDesc,
	MinQueueDurationDesc,
	MaxQueueDurationDesc,
	MedianQueueDurationDesc,
	MedianQueueDurationDesc,
	QueueDurationDesc,
	MinJobRunDurationDesc,
	MaxJobRunDurationDesc,
	MedianJobRunDurationDesc,
	JobRunDurationDesc,
	QueueAllocatedDesc,
	MinQueueAllocatedDesc,
	MaxQueueAllocatedDesc,
	MedianQueueAllocatedDesc,
	QueueUsedDesc,
	QueueLeasedPodCountDesc,
	ClusterCapacityDesc,
	ClusterAvailableCapacityDesc,
	QueuePriorityDesc,
	QueueLabelDesc,
}
View Source 
var ClusterAvailableCapacityDesc = prometheus.NewDesc(
	MetricPrefix+"cluster_available_capacity",
	"Cluster capacity available for Armada jobs",
	[]string{"cluster", "pool", "resourceType", "nodeType"},
	nil,
)
View Source 
var ClusterCapacityDesc = prometheus.NewDesc(
	MetricPrefix+"cluster_capacity",
	"Cluster capacity",
	[]string{"cluster", "pool", "resourceType", "nodeType"},
	nil,
)
View Source 
var ClusterCordonedStatusDesc = prometheus.NewDesc(
	MetricPrefix+"cluster_cordoned_status",
	"Cluster cordoned status",
	[]string{"cluster", "reason", "setByUser"},
	nil,
)
View Source 
var CountQueueResourcesDesc = prometheus.NewDesc(
	MetricPrefix+"queue_resource_queued_count",
	"Count of queued jobs requiring resource",
	[]string{"pool", "priorityClass", "queueName", "queue", "resourceType"},
	nil,
)
View Source 
var JobRunDurationDesc = prometheus.NewDesc(
	MetricPrefix+"job_run_time_seconds",
	"Run time for Armada jobs",
	[]string{"pool", "priorityClass", "queueName", "queue"},
	nil,
)
View Source 
var MaxJobRunDurationDesc = prometheus.NewDesc(
	MetricPrefix+"job_run_time_seconds_max",
	"Max run time for Armada jobs",
	[]string{"pool", "priorityClass", "queueName", "queue"},
	nil,
)
View Source 
var MaxQueueAllocatedDesc = prometheus.NewDesc(
	MetricPrefix+"queue_resource_allocated_max",
	"Max resource allocated by a running job",
	[]string{"pool", "priorityClass", "queueName", "queue", "resourceType"},
	nil,
)
View Source 
var MaxQueueDurationDesc = prometheus.NewDesc(
	MetricPrefix+"job_queued_seconds_max",
	"Max queue time for Armada jobs",
	[]string{"pool", "priorityClass", "queueName", "queue"},
	nil,
)
View Source 
var MaxQueueResourcesDesc = prometheus.NewDesc(
	MetricPrefix+"queue_resource_queued_max",
	"Max resource required by queued job",
	[]string{"pool", "priorityClass", "queueName", "queue", "resourceType"},
	nil,
)
View Source 
var MedianJobRunDurationDesc = prometheus.NewDesc(
	MetricPrefix+"job_run_time_seconds_median",
	"Median run time for Armada jobs",
	[]string{"pool", "priorityClass", "queueName", "queue"},
	nil,
)
View Source 
var MedianQueueAllocatedDesc = prometheus.NewDesc(
	MetricPrefix+"queue_resource_allocated_median",
	"Median resource allocated by a running job",
	[]string{"pool", "priorityClass", "queueName", "queue", "resourceType"},
	nil,
)
View Source 
var MedianQueueDurationDesc = prometheus.NewDesc(
	MetricPrefix+"job_queued_seconds_median",
	"Median queue time for Armada jobs",
	[]string{"pool", "priorityClass", "queueName", "queue"},
	nil,
)
View Source 
var MedianQueueResourcesDesc = prometheus.NewDesc(
	MetricPrefix+"queue_resource_queued_median",
	"Median resource required by queued jobs",
	[]string{"pool", "priorityClass", "queueName", "queue", "resourceType"},
	nil,
)
View Source 
var MinJobRunDurationDesc = prometheus.NewDesc(
	MetricPrefix+"job_run_time_seconds_min",
	"Min run time for Armada jobs",
	[]string{"pool", "priorityClass", "queueName", "queue"},
	nil,
)
View Source 
var MinQueueAllocatedDesc = prometheus.NewDesc(
	MetricPrefix+"queue_resource_allocated_min",
	"Min resource allocated by a running job",
	[]string{"pool", "priorityClass", "queueName", "queue", "resourceType"},
	nil,
)
View Source 
var MinQueueDurationDesc = prometheus.NewDesc(
	MetricPrefix+"job_queued_seconds_min",
	"Min queue time for Armada jobs",
	[]string{"pool", "priorityClass", "queueName", "queue"},
	nil,
)
View Source 
var MinQueueResourcesDesc = prometheus.NewDesc(
	MetricPrefix+"queue_resource_queued_min",
	"Min resource required by queued job",
	[]string{"pool", "priorityClass", "queueName", "queue", "resourceType"},
	nil,
)
View Source 
var QueueAllocatedDesc = prometheus.NewDesc(
	MetricPrefix+"queue_resource_allocated",
	"Resource allocated to running jobs of a queue",
	[]string{"cluster", "pool", "priorityClass", "queueName", "queue", "resourceType", "nodeType"},
	nil,
)
View Source 
var QueueDistinctSchedulingKeysDesc = prometheus.NewDesc(
	MetricPrefix+"queue_distinct_scheduling_keys",
	"Number of distinct scheduling keys requested by a queue",
	[]string{"queueName", "queue"},
	nil,
)
View Source 
var QueueDurationDesc = prometheus.NewDesc(
	MetricPrefix+"job_queued_seconds",
	"Queued time for Armada jobs",
	[]string{"pool", "priorityClass", "queueName", "queue"},
	nil,
)
View Source 
var QueueLabelDesc = prometheus.NewDesc(
	queueLabelMetricName,
	queueLabelMetricDescription,
	queueLabelDefaultLabels,
	nil,
)

QueueLabelDesc so it can be added to AllDescs which makes Describe() work properly 

actual describe for this metric is generated dynamically as the labels are dynamic
View Source 
var QueueLeasedPodCountDesc = prometheus.NewDesc(
	MetricPrefix+"queue_leased_pod_count",
	"Number of leased pods",
	[]string{"cluster", "pool", "queueName", "queue", "phase", "nodeType"},
	nil,
)
View Source 
var QueuePriorityDesc = prometheus.NewDesc(
	MetricPrefix+"queue_priority",
	"Queue priority factor",
	[]string{"queueName", "queue"},
	nil,
)
View Source 
var QueueResourcesDesc = prometheus.NewDesc(
	MetricPrefix+"queue_resource_queued",
	"Resource required by queued jobs",
	[]string{"pool", "priorityClass", "queueName", "queue", "resourceType"},
	nil,
)
View Source 
var QueueSizeDesc = prometheus.NewDesc(
	MetricPrefix+"queue_size",
	"Number of jobs in a queue",
	[]string{"queueName", "queue"},
	nil,
)
View Source 
var QueueUsedDesc = prometheus.NewDesc(
	MetricPrefix+"queue_resource_used",
	"Resource actually being used by running jobs of a queue",
	[]string{"cluster", "pool", "queueName", "queue", "resourceType", "nodeType"},
	nil,
)

Functions

func CollectQueueMetrics 

func CollectQueueMetrics(queueCounts map[string]int, queueDistinctSchedulingKeyCounts map[string]int, metricsProvider QueueMetricProvider) []prometheus.Metric

func Describe 

func Describe(out chan<- *prometheus.Desc)

func NewClusterAvailableCapacity added in v0.3.66 

func NewClusterAvailableCapacity(value float64, cluster string, pool string, resource string, nodeType string) prometheus.Metric

func NewClusterCordonedStatus added in v0.15.2 

func NewClusterCordonedStatus(value float64, cluster string, reason string, setByUser string) prometheus.Metric

func NewClusterTotalCapacity added in v0.3.66 

func NewClusterTotalCapacity(value float64, cluster string, pool string, resource string, nodeType string) prometheus.Metric

func NewCountQueueResources 

func NewCountQueueResources(value uint64, pool string, priorityClass string, queue string, resource string) prometheus.Metric

func NewJobRunRunDuration 

func NewJobRunRunDuration(count uint64, sum float64, buckets map[float64]uint64, pool string, priorityClass string, queue string) prometheus.Metric

func NewMaxJobRunDuration 

func NewMaxJobRunDuration(value float64, pool string, priorityClass string, queue string) prometheus.Metric

func NewMaxQueueAllocated 

func NewMaxQueueAllocated(value float64, pool string, priorityClass string, queue string, resource string) prometheus.Metric

func NewMaxQueueDuration 

func NewMaxQueueDuration(value float64, pool string, priorityClass string, queue string) prometheus.Metric

func NewMaxQueueResources 

func NewMaxQueueResources(value float64, pool string, priorityClass string, queue string, resource string) prometheus.Metric

func NewMedianJobRunDuration 

func NewMedianJobRunDuration(value float64, pool string, priorityClass string, queue string) prometheus.Metric

func NewMedianQueueAllocated 

func NewMedianQueueAllocated(value float64, pool string, priorityClass string, queue string, resource string) prometheus.Metric

func NewMedianQueueDuration 

func NewMedianQueueDuration(value float64, pool string, priorityClass string, queue string) prometheus.Metric

func NewMedianQueueResources 

func NewMedianQueueResources(value float64, pool string, priorityClass string, queue string, resource string) prometheus.Metric

func NewMinJobRunDuration 

func NewMinJobRunDuration(value float64, pool string, priorityClass string, queue string) prometheus.Metric

func NewMinQueueAllocated 

func NewMinQueueAllocated(value float64, pool string, priorityClass string, queue string, resource string) prometheus.Metric

func NewMinQueueDuration 

func NewMinQueueDuration(value float64, pool string, priorityClass string, queue string) prometheus.Metric

func NewMinQueueResources 

func NewMinQueueResources(value float64, pool string, priorityClass string, queue string, resource string) prometheus.Metric

func NewQueueAllocated added in v0.3.66 

func NewQueueAllocated(value float64, queue string, cluster string, pool string, priorityClass string, resource string, nodeType string) prometheus.Metric

func NewQueueDistinctSchedulingKeyMetric added in v0.4.52 

func NewQueueDistinctSchedulingKeyMetric(value int, queue string) prometheus.Metric

func NewQueueDuration 

func NewQueueDuration(count uint64, sum float64, buckets map[float64]uint64, pool string, priorityClass string, queue string) prometheus.Metric

func NewQueueLabelsMetric added in v0.10.0 

func NewQueueLabelsMetric(queue string, labels map[string]string) prometheus.Metric

func NewQueueLeasedPodCount added in v0.3.66 

func NewQueueLeasedPodCount(value float64, cluster string, pool string, queue string, phase string, nodeType string) prometheus.Metric

func NewQueuePriorityMetric added in v0.8.7 

func NewQueuePriorityMetric(value float64, queue string) prometheus.Metric

func NewQueueResources 

func NewQueueResources(value float64, pool string, priorityClass string, queue string, resource string) prometheus.Metric

func NewQueueSizeMetric 

func NewQueueSizeMetric(value int, queue string) prometheus.Metric

func NewQueueUsed added in v0.3.66 

func NewQueueUsed(value float64, queue string, cluster string, pool string, resource string, nodeType string) prometheus.Metric

Types

type FloatMetrics 

type FloatMetrics struct {
	// contains filtered or unexported fields
}

func (*FloatMetrics) GetBuckets 

func (d *FloatMetrics) GetBuckets() map[float64]uint64

func (*FloatMetrics) GetCount 

func (d *FloatMetrics) GetCount() uint64

func (*FloatMetrics) GetMax 

func (d *FloatMetrics) GetMax() float64

func (*FloatMetrics) GetMedian 

func (d *FloatMetrics) GetMedian() float64

func (*FloatMetrics) GetMin 

func (d *FloatMetrics) GetMin() float64

func (*FloatMetrics) GetSum 

func (d *FloatMetrics) GetSum() float64

type FloatMetricsRecorder 

type FloatMetricsRecorder struct {
	// contains filtered or unexported fields
}

func NewDefaultJobDurationMetricsRecorder 

func NewDefaultJobDurationMetricsRecorder() *FloatMetricsRecorder

func NewFloatMetricsRecorder 

func NewFloatMetricsRecorder(buckets ...float64) *FloatMetricsRecorder

func (*FloatMetricsRecorder) GetMetrics 

func (d *FloatMetricsRecorder) GetMetrics() *FloatMetrics

func (*FloatMetricsRecorder) Record 

func (d *FloatMetricsRecorder) Record(value float64)

type HttpMetricsProvider added in v0.3.91 

type HttpMetricsProvider struct {
	// contains filtered or unexported fields
}

HttpMetricsProvider is a metrics provider scraping metrics from a url.

func NewHttpMetricsProvider added in v0.3.91 

func NewHttpMetricsProvider(url string, client *http.Client) *HttpMetricsProvider

func (*HttpMetricsProvider) Collect added in v0.3.91 

func (srv *HttpMetricsProvider) Collect(ctx context.Context, _ *logrus.Entry) (map[string]float64, error)

type JobMetricsRecorder 

type JobMetricsRecorder struct {
	// contains filtered or unexported fields
}

func NewJobMetricsRecorder 

func NewJobMetricsRecorder() *JobMetricsRecorder

func (*JobMetricsRecorder) Metrics 

func (r *JobMetricsRecorder) Metrics() []*QueueMetrics

func (*JobMetricsRecorder) RecordJobRuntime 

func (r *JobMetricsRecorder) RecordJobRuntime(pool string, priorityClass string, jobRuntime time.Duration)

func (*JobMetricsRecorder) RecordResources 

func (r *JobMetricsRecorder) RecordResources(pool string, priorityClass string, resources armadaresource.ComputeResourcesFloat)

type ManualMetricsProvider added in v0.3.91 

type ManualMetricsProvider struct {
	// contains filtered or unexported fields
}

func (*ManualMetricsProvider) Collect added in v0.3.91 

func (srv *ManualMetricsProvider) Collect(_ context.Context, _ *logrus.Entry) (map[string]float64, error)

func (*ManualMetricsProvider) WithCollectionDelay added in v0.3.91 

func (srv *ManualMetricsProvider) WithCollectionDelay(d time.Duration) *ManualMetricsProvider

func (*ManualMetricsProvider) WithMetrics added in v0.3.91 

func (srv *ManualMetricsProvider) WithMetrics(metrics map[string]float64) *ManualMetricsProvider

type MetricsProvider added in v0.3.91 

type MetricsProvider interface {
	Collect(context.Context, *logrus.Entry) (map[string]float64, error)
}

type QueueMetricProvider 

type QueueMetricProvider interface {
	GetQueuedJobMetrics(queueName string) []*QueueMetrics
	GetRunningJobMetrics(queueName string) []*QueueMetrics
	GetAllQueues() []*api.Queue
}

type QueueMetrics 

type QueueMetrics struct {
	Pool          string
	PriorityClass string
	Resources     ResourceMetrics
	Durations     *FloatMetrics
}

type QueueMetricsRecorder 

type QueueMetricsRecorder struct {
	Pool          string
	PriorityClass string
	// contains filtered or unexported fields
}

type ResourceMetrics 

type ResourceMetrics map[string]*FloatMetrics

type ResourceMetricsRecorder 

type ResourceMetricsRecorder struct {
	// contains filtered or unexported fields
}

func NewResourceMetricsRecorder 

func NewResourceMetricsRecorder() *ResourceMetricsRecorder

func (*ResourceMetricsRecorder) GetMetrics 

func (d *ResourceMetricsRecorder) GetMetrics() ResourceMetrics

func (*ResourceMetricsRecorder) Record 

func (d *ResourceMetricsRecorder) Record(value armadaresource.ComputeResourcesFloat)

Source Files

View all Source files

Jump to

Keyboard shortcuts

? : This menu
/ : Search site
f or F : Jump to
y or Y : Canonical URL
go.dev uses cookies from Google to deliver and enhance the quality of its services and to analyze traffic. Learn more.