slo

package
v0.12.1 Latest Latest
Warning

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

 Go to latest Published: Nov 17, 2023 License: Apache-2.0 Imports: 46 Imported by: 0

Documentation

Overview

Reference : https://github.dev/slok/sloth/tree/main/internal/alert/window.go

Index

Constants

View Source 
const (

	// alert rule names
	RecordingRuleSuffix = "-recording"
	MetadataRuleSuffix  = "-metadata"
	AlertRuleSuffix     = "-alerts"
)

Variables

View Source 
var EnabledFilters = map[string]embed.FS{"metricgroups": MetricGroups, "servicegroups": ServiceGroups}

map of directory names to their embed.FS

View Source 
var MetricGroups embed.FS
View Source 
var ServiceGroups embed.FS

Functions

func ApplyFiltersToCortexEvents added in v0.6.0 

func ApplyFiltersToCortexEvents(seriesInfo *cortexadmin.SeriesInfoList) (*sloapi.MetricGroupList, error)

func DetectActiveWindows added in v0.6.0 

func DetectActiveWindows(severity string, matrix *prommodel.Matrix) ([]*sloapi.AlertFiringWindows, error)

DetectActiveWindows

@warning Expectation is that the timestamps are ordered when traversing matrix --> sample streams --> [] values but this may not always be the case

func LeftJoinSlice added in v0.6.0 

func LeftJoinSlice[T comparable](arr1, arr2 []T) []T

func LeftJoinSliceAbstract added in v0.6.0 

func LeftJoinSliceAbstract[T any, S comparable](arr1, arr2 []T, getId func(T) S) []T

func MergeLabels 

func MergeLabels(ms ...map[string]string) map[string]string

func MergeRuleGroups added in v0.6.0 

func MergeRuleGroups(left rulefmt.RuleGroup, right *rulefmt.RuleGroup) *rulefmt.RuleGroup

func NewWindowRange added in v0.6.0 

func NewWindowRange(sloPeriod string) []string

func QuerySLOComponentByRawQuery added in v0.6.0 

func QuerySLOComponentByRawQuery(
	ctx context.Context,
	client cortexadmin.CortexAdminClient,
	rawQuery string,
	clusterId string,
) (*model.Vector, error)

func QuerySLOComponentByRawQueryRange added in v0.6.0 

func QuerySLOComponentByRawQueryRange(
	ctx context.Context,
	client cortexadmin.CortexAdminClient,
	rawQuery string,
	clusterId string,
	start time.Time,
	end time.Time,
	step time.Duration,
) (*model.Matrix, error)

func QuerySLOComponentByRecordName added in v0.6.0 

func QuerySLOComponentByRecordName(
	ctx context.Context,
	client cortexadmin.CortexAdminClient,
	recordName string,
	clusterId string,
) (*model.Vector, error)

func RegisterDatasource 

func RegisterDatasource(datasource string, sloImpl SLOStore, serviceImpl ServiceBackend)

func Scheme 

func Scheme(ctx context.Context) meta.Scheme

func TimeDurationToPromStr added in v0.6.0 

func TimeDurationToPromStr(t time.Duration) string

Pretty simple durations for prometheus.

func ToMatchingSubsetIdenticalMetric added in v0.6.0 

func ToMatchingSubsetIdenticalMetric(goodEvents, totalEvents []*sloapi.Event) (good, total []*sloapi.Event)

ToMatchingSubsetIdenticalMetric only applies when the good metric & total metric id is the same

Types

type Filter added in v0.6.0 

type Filter struct {
	Name    string        `yaml:"name"`
	Filters []FilterValue `yaml:"filters"`
	Ignore  []FilterValue `yaml:"ignore"`
}

func GetGroupConfigsFromEmbed added in v0.6.0 

func GetGroupConfigsFromEmbed(lg *slog.Logger, dirName string, dir embed.FS) []Filter

type FilterValue added in v0.6.0 

type FilterValue struct {
	Value Regexp `yaml:"value"`
	Score int    `yaml:"score"`
}

type IdentificationLabels added in v0.6.0 

type IdentificationLabels map[string]string

type LabelPair added in v0.6.0 

type LabelPair struct {
	Key  string
	Vals []string
}

type LabelPairs added in v0.6.0 

type LabelPairs []LabelPair

func (LabelPairs) Construct added in v0.6.0 

func (l LabelPairs) Construct() string

type Metric added in v0.6.0 

type Metric string

type MetricIds 

type MetricIds struct {
	Good  string
	Total string
}

type MonitoringServiceBackend 

type MonitoringServiceBackend struct {
	RequestBase
}

func (MonitoringServiceBackend) ListEvents added in v0.6.0 

func (m MonitoringServiceBackend) ListEvents() (*sloapi.EventList, error)

func (MonitoringServiceBackend) ListMetrics added in v0.6.0 

func (m MonitoringServiceBackend) ListMetrics() (*sloapi.MetricGroupList, error)

func (MonitoringServiceBackend) ListServices added in v0.6.0 

func (m MonitoringServiceBackend) ListServices() (*sloapi.ServiceList, error)

func (*MonitoringServiceBackend) WithCurrentRequest 

func (m *MonitoringServiceBackend) WithCurrentRequest(ctx context.Context, req proto.Message) ServiceBackend

type Plugin 

type Plugin struct {
	slo.UnsafeSLOServer
	system.UnimplementedSystemPluginClient
	// contains filtered or unexported fields
}

func NewPlugin 

func NewPlugin(ctx context.Context) *Plugin

func (*Plugin) CloneSLO 

func (p *Plugin) CloneSLO(ctx context.Context, ref *corev1.Reference) (*sloapi.SLOData, error)

func (*Plugin) CloneToClusters added in v0.6.0 

func (p *Plugin) CloneToClusters(ctx context.Context, req *sloapi.MultiClusterSLO) (*sloapi.MultiClusterFailures, error)

func (*Plugin) CreateSLO 

func (p *Plugin) CreateSLO(ctx context.Context, slorequest *sloapi.CreateSLORequest) (*corev1.Reference, error)

func (*Plugin) DeleteSLO 

func (p *Plugin) DeleteSLO(ctx context.Context, req *corev1.Reference) (*emptypb.Empty, error)

func (*Plugin) GetSLO 

func (p *Plugin) GetSLO(ctx context.Context, ref *corev1.Reference) (*sloapi.SLOData, error)

func (*Plugin) ListEvents added in v0.6.0 

func (p *Plugin) ListEvents(ctx context.Context, req *sloapi.ListEventsRequest) (*sloapi.EventList, error)

func (*Plugin) ListMetrics 

func (p *Plugin) ListMetrics(ctx context.Context, req *sloapi.ListMetricsRequest) (*sloapi.MetricGroupList, error)

func (*Plugin) ListSLOs 

func (p *Plugin) ListSLOs(ctx context.Context, _ *emptypb.Empty) (*sloapi.ServiceLevelObjectiveList, error)

func (*Plugin) ListServices 

func (p *Plugin) ListServices(ctx context.Context, req *sloapi.ListServicesRequest) (*sloapi.ServiceList, error)

func (*Plugin) Preview added in v0.6.0 

func (p *Plugin) Preview(ctx context.Context, req *sloapi.CreateSLORequest) (*sloapi.SLOPreviewResponse, error)

func (*Plugin) Status 

func (p *Plugin) Status(ctx context.Context, ref *corev1.Reference) (*sloapi.SLOStatus, error)

func (*Plugin) UpdateSLO 

func (p *Plugin) UpdateSLO(ctx context.Context, req *sloapi.SLOData) (*emptypb.Empty, error)

func (*Plugin) UseAPIExtensions 

func (p *Plugin) UseAPIExtensions(intf system.ExtensionClientInterface)

func (*Plugin) UseKeyValueStore 

func (p *Plugin) UseKeyValueStore(client system.KeyValueStoreClient)

func (*Plugin) UseManagementAPI 

func (p *Plugin) UseManagementAPI(client managementv1.ManagementClient)

type Regexp added in v0.6.0 

type Regexp struct {
	*regexp.Regexp
}

Regexp adds unmarshalling from json for regexp.Regexp

func (*Regexp) MarshalText added in v0.6.0 

func (r *Regexp) MarshalText() ([]byte, error)

MarshalText marshals regexp.Regexp as string

func (*Regexp) UnmarshalText added in v0.6.0 

func (r *Regexp) UnmarshalText(b []byte) error

UnmarshalText unmarshals json into a regexp.Regexp

type RequestBase 

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

type SLO added in v0.6.0 

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

func CreateSLORequestToStruct added in v0.6.0 

func CreateSLORequestToStruct(c *sloapi.CreateSLORequest) *SLO

func NewSLO added in v0.6.0 

func NewSLO(
	sloName string,
	sloPeriod string,
	objective float64,
	svc Service,
	goodMetric Metric,
	totalMetric Metric,
	userLabels map[string]string,
	goodEvents []LabelPair,
	totalEvents []LabelPair,
) *SLO

func SLODataToStruct added in v0.6.0 

func SLODataToStruct(s *sloapi.SLOData) *SLO

func SLOFromId added in v0.6.0 

func SLOFromId(
	sloName string,
	sloPeriod string,
	objective float64,
	svc Service,
	goodMetric Metric,
	totalMetric Metric,
	userLabels map[string]string,
	goodEvents []LabelPair,
	totalEvents []LabelPair,
	id string,
) *SLO

func (*SLO) AlertPageThreshold added in v0.6.0 

func (s *SLO) AlertPageThreshold() float64

func (*SLO) ConstructAlertingRuleGroup added in v0.6.0 

func (s *SLO) ConstructAlertingRuleGroup(interval *time.Duration) rulefmt.RuleGroup

ConstructAlertingRuleGroup

Note: first two are expected to be the recording rules Note: second two are expected to be the alerting rules

func (*SLO) ConstructCortexRules added in v0.6.0 

func (s *SLO) ConstructCortexRules(interval *time.Duration) (sli, metadata, alerts rulefmt.RuleGroup)

func (*SLO) ConstructMetadataRules added in v0.6.0 

func (s *SLO) ConstructMetadataRules(interval *time.Duration) rulefmt.RuleGroup

func (*SLO) ConstructRawAlertQueries added in v0.6.0 

func (s *SLO) ConstructRawAlertQueries() (yaml.Node, yaml.Node)

func (*SLO) ConstructRecordingRuleGroup added in v0.6.0 

func (s *SLO) ConstructRecordingRuleGroup(interval *time.Duration) rulefmt.RuleGroup

func (*SLO) GetId added in v0.6.0 

func (s *SLO) GetId() string

func (*SLO) GetName added in v0.6.0 

func (s *SLO) GetName() string

func (*SLO) GetObjective added in v0.6.0 

func (s *SLO) GetObjective() float64

func (*SLO) GetPeriod added in v0.6.0 

func (s *SLO) GetPeriod() string

func (*SLO) GetPrometheusRuleFilterByIdLabels added in v0.6.0 

func (s *SLO) GetPrometheusRuleFilterByIdLabels() (string, error)

func (*SLO) RawBudgetRemainingQuery added in v0.6.0 

func (s *SLO) RawBudgetRemainingQuery() string

func (*SLO) RawCurrentBurnRateQuery added in v0.6.0 

func (s *SLO) RawCurrentBurnRateQuery() string

RawCurrentBurnRateQuery ratioRate : slo:sli_error:ratio_rate<some-period-string>

func (*SLO) RawDashboardInfoQuery added in v0.6.0 

func (s *SLO) RawDashboardInfoQuery() string

func (*SLO) RawErrorBudgetQuery added in v0.6.0 

func (s *SLO) RawErrorBudgetQuery() string

func (*SLO) RawGoodEventsQuery added in v0.6.0 

func (s *SLO) RawGoodEventsQuery(w string) (string, error)

func (*SLO) RawObjectiveQuery added in v0.6.0 

func (s *SLO) RawObjectiveQuery() string

func (*SLO) RawPeriodBurnRateQuery added in v0.6.0 

func (s *SLO) RawPeriodBurnRateQuery() string

func (*SLO) RawPeriodDurationQuery added in v0.6.0 

func (s *SLO) RawPeriodDurationQuery() string

func (*SLO) RawSLIQuery added in v0.6.0 

func (s *SLO) RawSLIQuery(w string) (string, error)

func (*SLO) RawTotalEventsQuery added in v0.6.0 

func (s *SLO) RawTotalEventsQuery(w string) (string, error)

func (*SLO) SetId added in v0.6.0 

func (s *SLO) SetId(id string)

func (*SLO) SetName added in v0.6.0 

func (s *SLO) SetName(input string)

type SLOLogging 

type SLOLogging struct {
	RequestBase
}

type SLOMonitoring 

type SLOMonitoring struct {
	RequestBase
}

func (SLOMonitoring) Clone 

func (s SLOMonitoring) Clone(clone *sloapi.SLOData) (*corev1.Reference, *sloapi.SLOData, error)

func (SLOMonitoring) Create 

func (s SLOMonitoring) Create() (*corev1.Reference, error)

func (SLOMonitoring) Delete 

func (s SLOMonitoring) Delete(existing *sloapi.SLOData) error

func (SLOMonitoring) MultiClusterClone added in v0.6.0 

func (s SLOMonitoring) MultiClusterClone(
	base *sloapi.SLOData,
	inputClusters []*corev1.Reference,
	svcBackend ServiceBackend,
) ([]*corev1.Reference, []*sloapi.SLOData, []error)

func (SLOMonitoring) Preview added in v0.6.0 

func (s SLOMonitoring) Preview(slo *SLO) (*sloapi.SLOPreviewResponse, error)

func (SLOMonitoring) Status 

func (s SLOMonitoring) Status(existing *sloapi.SLOData) (*sloapi.SLOStatus, error)

Status Only return errors here that should be considered severe InternalServerErrors - Check if enough time has passed to evaluate the rules - First Checks if it has NoData - If it has Data, check if it is within budget - If is within budget, check if any alerts are firing

func (SLOMonitoring) Update 

func (s SLOMonitoring) Update(existing *sloapi.SLOData) (*sloapi.SLOData, error)

func (*SLOMonitoring) WithCurrentRequest 

func (s *SLOMonitoring) WithCurrentRequest(ctx context.Context, req proto.Message) SLOStore

type SLOStore 

type SLOStore interface {
	// This method has to handle storage of the SLO in the KVStore itself
	// since there can be partial successes inside the method
	Create() (*corev1.Reference, error)
	Update(existing *sloapi.SLOData) (*sloapi.SLOData, error)
	Delete(existing *sloapi.SLOData) error
	Clone(clone *sloapi.SLOData) (*corev1.Reference, *sloapi.SLOData, error)
	MultiClusterClone(
		base *sloapi.SLOData,
		clusters []*corev1.Reference,
		svcBackend ServiceBackend,
	) ([]*corev1.Reference, []*sloapi.SLOData, []error)
	Status(existing *sloapi.SLOData) (*sloapi.SLOStatus, error)
	Preview(s *SLO) (*sloapi.SLOPreviewResponse, error)
	WithCurrentRequest(ctx context.Context, req proto.Message) SLOStore
}

func NewSLOMonitoringStore 

func NewSLOMonitoringStore(p *Plugin, lg *slog.Logger) SLOStore

type Service added in v0.6.0 

type Service string

type ServiceBackend 

type ServiceBackend interface {
	ListServices() (*sloapi.ServiceList, error)
	ListMetrics() (*sloapi.MetricGroupList, error)
	ListEvents() (*sloapi.EventList, error)
	WithCurrentRequest(ctx context.Context, req proto.Message) ServiceBackend
}

func NewMonitoringServiceBackend 

func NewMonitoringServiceBackend(p *Plugin, lg *slog.Logger) ServiceBackend

type SliQueryInfo added in v0.6.0 

type SliQueryInfo struct {
	GoodQuery  string
	TotalQuery string
}

SliQueryInfo used for filling sli query templates

type SloFiltersInfo added in v0.6.0 

type SloFiltersInfo struct {
	SloIdLabel      string
	SloServiceLabel string
	SloNameLabel    string
	SloId           string
	SloService      string
	SloName         string
}

type StorageAPIs 

type StorageAPIs struct {
	SLOs     storage.KeyValueStoreT[*slo.SLOData]
	Services storage.KeyValueStoreT[*slo.Service]
	Metrics  storage.KeyValueStoreT[*slo.Metric]
}

type UserLabels added in v0.6.0 

type UserLabels []string

type Window 

type Window struct {
	// ErrorBudgetPercent is the error budget % consumed for a full time window.
	// Google gives us some defaults in its SRE workbook that work correctly most of the times:
	// - Page quick:   2%
	// - Page slow:    5%
	// - Ticket quick: 10%
	// - Ticket slow:  10%
	ErrorBudgetPercent float64
	// ShortWindow is the small window used on the alerting part to stop alerting
	// during a long window because we consumed a lot of error budget but the problem
	// is already gone.
	ShortWindow time.Duration
	// LongWindow is the long window used to alert based on the errors happened on that
	// long window.
	LongWindow time.Duration
}

func (Window) Validate 

func (w Window) Validate() error

type Windows 

type Windows struct {
	SLOPeriod   time.Duration
	PageQuick   Window
	PageSlow    Window
	TicketQuick Window
	TicketSlow  Window
}

func GenerateGoogleWindows 

func GenerateGoogleWindows(budgetingInterval time.Duration) *Windows

https://sre.google/workbook/alerting-on-slos/

budgeting interval is the shortest interval to monitor in a window

func WindowDefaults 

func WindowDefaults(period time.Duration) *Windows

https://sre.google/workbook/alerting-on-slos/

func (Windows) GetBurnRateFactor 

func (w Windows) GetBurnRateFactor(totalWindow time.Duration, errorBudgetPercent float64, consumptionWindow time.Duration) float64

getBurnRateFactor calculates the burnRateFactor (speed) needed to consume all the error budget available percent in a specific time window taking into account the total time window.

func (Windows) GetSpeedPageQuick 

func (w Windows) GetSpeedPageQuick() float64

Error budget speeds based on a full time window, however once we have the factor (speed) the value can be used with any time window.

func (Windows) GetSpeedPageSlow 

func (w Windows) GetSpeedPageSlow() float64

func (Windows) GetSpeedTicketQuick 

func (w Windows) GetSpeedTicketQuick() float64

func (Windows) GetSpeedTicketSlow 

func (w Windows) GetSpeedTicketSlow() float64

func (Windows) Validate 

func (w Windows) Validate() error

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.