SigNoz is an open-source, OpenTelemetry-native observability platform for your team and their AI agents. Get logs, metrics, and traces in one tool with features like APM, distributed tracing, log management, infra monitoring, etc. Combined with SigNoz MCP and a native AI teammate (in SigNoz Cloud) it helps you build more resilient apps.
TypeScript apmapplication-monitoringdistributed-tracinggogood-first-issuejaegerloglogsmetricsmonitoringnextjsobservabilityopen-sourceopentelemetryprometheusreactreactjsself-hostedtracingtypescript
feat(infra-monitoring): filter by pod status (#12278)
Nikhil Mantri85bf5ce64499
diff --git a/docs/api/openapi.yml b/docs/api/openapi.ymlindex 74b62c5c60..8ce1533c32 100644--- a/docs/api/openapi.yml+++ b/docs/api/openapi.yml@@ -4212,6 +4212,21 @@ components:- missingOptionalMetrics- missingRequiredAttributestype: object+ InframonitoringtypesClusterFilter:+ properties:+ expression:+ type: string+ filterByNodeReadiness:+ items:+ $ref: '#/components/schemas/InframonitoringtypesNodeCondition'+ nullable: true+ type: array+ filterByPodStatus:+ items:+ $ref: '#/components/schemas/InframonitoringtypesPodStatus'+ nullable: true+ type: array+ type: objectInframonitoringtypesClusterRecord:properties:clusterCPU:@@ -4349,6 +4364,16 @@ components:- containerCannotRun- unknowntype: object+ InframonitoringtypesContainerFilter:+ properties:+ expression:+ type: string+ filterByContainerStatus:+ items:+ $ref: '#/components/schemas/InframonitoringtypesContainerStatus'+ nullable: true+ type: array+ type: objectInframonitoringtypesContainerReady:enum:- ready@@ -4448,6 +4473,16 @@ components:- total- endTimeBeforeRetentiontype: object+ InframonitoringtypesDaemonSetFilter:+ properties:+ expression:+ type: string+ filterByPodStatus:+ items:+ $ref: '#/components/schemas/InframonitoringtypesPodStatus'+ nullable: true+ type: array+ type: objectInframonitoringtypesDaemonSetRecord:properties:currentNodes:@@ -4520,6 +4555,16 @@ components:- total- endTimeBeforeRetentiontype: object+ InframonitoringtypesDeploymentFilter:+ properties:+ expression:+ type: string+ filterByPodStatus:+ items:+ $ref: '#/components/schemas/InframonitoringtypesPodStatus'+ nullable: true+ type: array+ type: objectInframonitoringtypesDeploymentRecord:properties:availablePods:@@ -4661,6 +4706,16 @@ components:- total- endTimeBeforeRetentiontype: object+ InframonitoringtypesJobFilter:+ properties:+ expression:+ type: string+ filterByPodStatus:+ items:+ $ref: '#/components/schemas/InframonitoringtypesPodStatus'+ nullable: true+ type: array+ type: objectInframonitoringtypesJobRecord:properties:activePods:@@ -4784,6 +4839,16 @@ components:- message- documentationLinktype: object+ InframonitoringtypesNamespaceFilter:+ properties:+ expression:+ type: string+ filterByPodStatus:+ items:+ $ref: '#/components/schemas/InframonitoringtypesPodStatus'+ nullable: true+ type: array+ type: objectInframonitoringtypesNamespaceRecord:properties:counts:@@ -4865,6 +4930,21 @@ components:- ready- notReadytype: object+ InframonitoringtypesNodeFilter:+ properties:+ expression:+ type: string+ filterByNodeReadiness:+ items:+ $ref: '#/components/schemas/InframonitoringtypesNodeCondition'+ nullable: true+ type: array+ filterByPodStatus:+ items:+ $ref: '#/components/schemas/InframonitoringtypesPodStatus'+ nullable: true+ type: array+ type: objectInframonitoringtypesNodeRecord:properties:condition:@@ -4981,6 +5061,16 @@ components:- shutdown- unexpectedAdmissionErrortype: object+ InframonitoringtypesPodFilter:+ properties:+ expression:+ type: string+ filterByPodStatus:+ items:+ $ref: '#/components/schemas/InframonitoringtypesPodStatus'+ nullable: true+ type: array+ type: objectInframonitoringtypesPodRecord:properties:meta:@@ -5080,7 +5170,7 @@ components:format: int64type: integerfilter:- $ref: '#/components/schemas/Querybuildertypesv5Filter'+ $ref: '#/components/schemas/InframonitoringtypesClusterFilter'groupBy:items:$ref: '#/components/schemas/Querybuildertypesv5GroupByKey'@@ -5106,7 +5196,7 @@ components:format: int64type: integerfilter:- $ref: '#/components/schemas/Querybuildertypesv5Filter'+ $ref: '#/components/schemas/InframonitoringtypesContainerFilter'groupBy:items:$ref: '#/components/schemas/Querybuildertypesv5GroupByKey'@@ -5132,7 +5222,7 @@ components:format: int64type: integerfilter:- $ref: '#/components/schemas/Querybuildertypesv5Filter'+ $ref: '#/components/schemas/InframonitoringtypesDaemonSetFilter'groupBy:items:$ref: '#/components/schemas/Querybuildertypesv5GroupByKey'@@ -5158,7 +5248,7 @@ components:format: int64type: integerfilter:- $ref: '#/components/schemas/Querybuildertypesv5Filter'+ $ref: '#/components/schemas/InframonitoringtypesDeploymentFilter'groupBy:items:$ref: '#/components/schemas/Querybuildertypesv5GroupByKey'@@ -5210,7 +5300,7 @@ components:format: int64type: integerfilter:- $ref: '#/components/schemas/Querybuildertypesv5Filter'+ $ref: '#/components/schemas/InframonitoringtypesJobFilter'groupBy:items:$ref: '#/components/schemas/Querybuildertypesv5GroupByKey'@@ -5236,7 +5326,7 @@ components:format: int64type: integerfilter:- $ref: '#/components/schemas/Querybuildertypesv5Filter'+ $ref: '#/components/schemas/InframonitoringtypesNamespaceFilter'groupBy:items:$ref: '#/components/schemas/Querybuildertypesv5GroupByKey'@@ -5262,7 +5352,7 @@ components:format: int64type: integerfilter:- $ref: '#/components/schemas/Querybuildertypesv5Filter'+ $ref: '#/components/schemas/InframonitoringtypesNodeFilter'groupBy:items:$ref: '#/components/schemas/Querybuildertypesv5GroupByKey'@@ -5288,7 +5378,7 @@ components:format: int64type: integerfilter:- $ref: '#/components/schemas/Querybuildertypesv5Filter'+ $ref: '#/components/schemas/InframonitoringtypesPodFilter'groupBy:items:$ref: '#/components/schemas/Querybuildertypesv5GroupByKey'@@ -5314,7 +5404,7 @@ components:format: int64type: integerfilter:- $ref: '#/components/schemas/Querybuildertypesv5Filter'+ $ref: '#/components/schemas/InframonitoringtypesStatefulSetFilter'groupBy:items:$ref: '#/components/schemas/Querybuildertypesv5GroupByKey'@@ -5365,6 +5455,16 @@ components:- list- grouped_listtype: string+ InframonitoringtypesStatefulSetFilter:+ properties:+ expression:+ type: string+ filterByPodStatus:+ items:+ $ref: '#/components/schemas/InframonitoringtypesPodStatus'+ nullable: true+ type: array+ type: objectInframonitoringtypesStatefulSetRecord:properties:currentPods:diff --git a/frontend/src/api/generated/services/sigNoz.schemas.ts b/frontend/src/api/generated/services/sigNoz.schemas.tsindex 1b5fb4c475..da04376e50 100644--- a/frontend/src/api/generated/services/sigNoz.schemas.ts+++ b/frontend/src/api/generated/services/sigNoz.schemas.ts@@ -5648,6 +5648,47 @@ export interface InframonitoringtypesChecksDTO {type: InframonitoringtypesCheckTypeDTO;}+export enum InframonitoringtypesNodeConditionDTO {+ ready = 'ready',+ not_ready = 'not_ready',+ no_data = 'no_data',+}+export enum InframonitoringtypesPodStatusDTO {+ pending = 'pending',+ running = 'running',+ failed = 'failed',+ unknown = 'unknown',+ crashloopbackoff = 'crashloopbackoff',+ imagepullbackoff = 'imagepullbackoff',+ errimagepull = 'errimagepull',+ createcontainerconfigerror = 'createcontainerconfigerror',+ containercreating = 'containercreating',+ oomkilled = 'oomkilled',+ completed = 'completed',+ error = 'error',+ containercannotrun = 'containercannotrun',+ evicted = 'evicted',+ nodeaffinity = 'nodeaffinity',+ nodelost = 'nodelost',+ shutdown = 'shutdown',+ unexpectedadmissionerror = 'unexpectedadmissionerror',+ no_data = 'no_data',+}+export interface InframonitoringtypesClusterFilterDTO {+ /**+ * @type string+ */+ expression?: string;+ /**+ * @type array,null+ */+ filterByNodeReadiness?: InframonitoringtypesNodeConditionDTO[] | null;+ /**+ * @type array,null+ */+ filterByPodStatus?: InframonitoringtypesPodStatusDTO[] | null;+}+export type InframonitoringtypesClusterRecordDTOCounts = {/*** @type integer@@ -5923,21 +5964,6 @@ export interface InframonitoringtypesContainerCountsByStatusDTO {waiting: number;}-export enum InframonitoringtypesContainerReadyDTO {- ready = 'ready',- not_ready = 'not_ready',- no_data = 'no_data',-}-export type InframonitoringtypesContainerRecordDTOMetaAnyOf = {- [key: string]: string;-};--/**- * @nullable- */-export type InframonitoringtypesContainerRecordDTOMeta =- InframonitoringtypesContainerRecordDTOMetaAnyOf | null;-export enum InframonitoringtypesContainerStatusDTO {running = 'running',waiting = 'waiting',@@ -5954,6 +5980,32 @@ export enum InframonitoringtypesContainerStatusDTO {unknown = 'unknown',no_data = 'no_data',}+export interface InframonitoringtypesContainerFilterDTO {+ /**+ * @type string+ */+ expression?: string;+ /**+ * @type array,null+ */+ filterByContainerStatus?: InframonitoringtypesContainerStatusDTO[] | null;+}++export enum InframonitoringtypesContainerReadyDTO {+ ready = 'ready',+ not_ready = 'not_ready',+ no_data = 'no_data',+}+export type InframonitoringtypesContainerRecordDTOMetaAnyOf = {+ [key: string]: string;+};++/**+ * @nullable+ */+export type InframonitoringtypesContainerRecordDTOMeta =+ InframonitoringtypesContainerRecordDTOMetaAnyOf | null;+export interface InframonitoringtypesContainerRecordDTO {containerCountsByReady: InframonitoringtypesContainerCountsByReadyDTO;containerCountsByStatus: InframonitoringtypesContainerCountsByStatusDTO;@@ -6025,6 +6077,17 @@ export interface InframonitoringtypesContainersDTO {warning?: Querybuildertypesv5QueryWarnDataDTO;}+export interface InframonitoringtypesDaemonSetFilterDTO {+ /**+ * @type string+ */+ expression?: string;+ /**+ * @type array,null+ */+ filterByPodStatus?: InframonitoringtypesPodStatusDTO[] | null;+}+export type InframonitoringtypesDaemonSetRecordDTOMetaAnyOf = {[key: string]: string;};@@ -6110,6 +6173,17 @@ export interface InframonitoringtypesDaemonSetsDTO {warning?: Querybuildertypesv5QueryWarnDataDTO;}+export interface InframonitoringtypesDeploymentFilterDTO {+ /**+ * @type string+ */+ expression?: string;+ /**+ * @type array,null+ */+ filterByPodStatus?: InframonitoringtypesPodStatusDTO[] | null;+}+export type InframonitoringtypesDeploymentRecordDTOMetaAnyOf = {[key: string]: string;};@@ -6272,6 +6346,17 @@ export interface InframonitoringtypesHostsDTO {warning?: Querybuildertypesv5QueryWarnDataDTO;}+export interface InframonitoringtypesJobFilterDTO {+ /**+ * @type string+ */+ expression?: string;+ /**+ * @type array,null+ */+ filterByPodStatus?: InframonitoringtypesPodStatusDTO[] | null;+}+export type InframonitoringtypesJobRecordDTOMetaAnyOf = {[key: string]: string;};@@ -6357,6 +6442,17 @@ export interface InframonitoringtypesJobsDTO {warning?: Querybuildertypesv5QueryWarnDataDTO;}+export interface InframonitoringtypesNamespaceFilterDTO {+ /**+ * @type string+ */+ expression?: string;+ /**+ * @type array,null+ */+ filterByPodStatus?: InframonitoringtypesPodStatusDTO[] | null;+}+export type InframonitoringtypesNamespaceRecordDTOCounts = {/*** @type integer@@ -6433,11 +6529,21 @@ export interface InframonitoringtypesNamespacesDTO {warning?: Querybuildertypesv5QueryWarnDataDTO;}-export enum InframonitoringtypesNodeConditionDTO {- ready = 'ready',- not_ready = 'not_ready',- no_data = 'no_data',+export interface InframonitoringtypesNodeFilterDTO {+ /**+ * @type string+ */+ expression?: string;+ /**+ * @type array,null+ */+ filterByNodeReadiness?: InframonitoringtypesNodeConditionDTO[] | null;+ /**+ * @type array,null+ */+ filterByPodStatus?: InframonitoringtypesPodStatusDTO[] | null;}+export type InframonitoringtypesNodeRecordDTOMetaAnyOf = {[key: string]: string;};@@ -6499,6 +6605,17 @@ export interface InframonitoringtypesNodesDTO {warning?: Querybuildertypesv5QueryWarnDataDTO;}+export interface InframonitoringtypesPodFilterDTO {+ /**+ * @type string+ */+ expression?: string;+ /**+ * @type array,null+ */+ filterByPodStatus?: InframonitoringtypesPodStatusDTO[] | null;+}+export type InframonitoringtypesPodRecordDTOMetaAnyOf = {[key: string]: string;};@@ -6509,27 +6626,6 @@ export type InframonitoringtypesPodRecordDTOMetaAnyOf = {export type InframonitoringtypesPodRecordDTOMeta =InframonitoringtypesPodRecordDTOMetaAnyOf | null;-export enum InframonitoringtypesPodStatusDTO {- pending = 'pending',- running = 'running',- failed = 'failed',- unknown = 'unknown',- crashloopbackoff = 'crashloopbackoff',- imagepullbackoff = 'imagepullbackoff',- errimagepull = 'errimagepull',- createcontainerconfigerror = 'createcontainerconfigerror',- containercreating = 'containercreating',- oomkilled = 'oomkilled',- completed = 'completed',- error = 'error',- containercannotrun = 'containercannotrun',- evicted = 'evicted',- nodeaffinity = 'nodeaffinity',- nodelost = 'nodelost',- shutdown = 'shutdown',- unexpectedadmissionerror = 'unexpectedadmissionerror',- no_data = 'no_data',-}export interface InframonitoringtypesPodRecordDTO {/*** @type object,null@@ -6606,7 +6702,7 @@ export interface InframonitoringtypesPostableClustersDTO {* @format int64*/end: number;- filter?: Querybuildertypesv5FilterDTO;+ filter?: InframonitoringtypesClusterFilterDTO;/*** @type array,null*/@@ -6633,7 +6729,7 @@ export interface InframonitoringtypesPostableContainersDTO {* @format int64*/end: number;- filter?: Querybuildertypesv5FilterDTO;+ filter?: InframonitoringtypesContainerFilterDTO;/*** @type array,null*/@@ -6660,7 +6756,7 @@ export interface InframonitoringtypesPostableDaemonSetsDTO {* @format int64*/end: number;- filter?: Querybuildertypesv5FilterDTO;+ filter?: InframonitoringtypesDaemonSetFilterDTO;/*** @type array,null*/@@ -6687,7 +6783,7 @@ export interface InframonitoringtypesPostableDeploymentsDTO {* @format int64*/end: number;- filter?: Querybuildertypesv5FilterDTO;+ filter?: InframonitoringtypesDeploymentFilterDTO;/*** @type array,null*/@@ -6741,7 +6837,7 @@ export interface InframonitoringtypesPostableJobsDTO {* @format int64*/end: number;- filter?: Querybuildertypesv5FilterDTO;+ filter?: InframonitoringtypesJobFilterDTO;/*** @type array,null*/@@ -6768,7 +6864,7 @@ export interface InframonitoringtypesPostableNamespacesDTO {* @format int64*/end: number;- filter?: Querybuildertypesv5FilterDTO;+ filter?: InframonitoringtypesNamespaceFilterDTO;/*** @type array,null*/@@ -6795,7 +6891,7 @@ export interface InframonitoringtypesPostableNodesDTO {* @format int64*/end: number;- filter?: Querybuildertypesv5FilterDTO;+ filter?: InframonitoringtypesNodeFilterDTO;/*** @type array,null*/@@ -6822,7 +6918,7 @@ export interface InframonitoringtypesPostablePodsDTO {* @format int64*/end: number;- filter?: Querybuildertypesv5FilterDTO;+ filter?: InframonitoringtypesPodFilterDTO;/*** @type array,null*/@@ -6843,13 +6939,24 @@ export interface InframonitoringtypesPostablePodsDTO {start: number;}+export interface InframonitoringtypesStatefulSetFilterDTO {+ /**+ * @type string+ */+ expression?: string;+ /**+ * @type array,null+ */+ filterByPodStatus?: InframonitoringtypesPodStatusDTO[] | null;+}+export interface InframonitoringtypesPostableStatefulSetsDTO {/*** @type integer* @format int64*/end: number;- filter?: Querybuildertypesv5FilterDTO;+ filter?: InframonitoringtypesStatefulSetFilterDTO;/*** @type array,null*/diff --git a/pkg/modules/inframonitoring/implinframonitoring/clusters.go b/pkg/modules/inframonitoring/implinframonitoring/clusters.goindex 59f9b8454d..48b329a474 100644--- a/pkg/modules/inframonitoring/implinframonitoring/clusters.go+++ b/pkg/modules/inframonitoring/implinframonitoring/clusters.go@@ -84,20 +84,38 @@ func buildClusterRecords(return records}+// getTopClusterGroupsAndMetadata concurrently fetches metadata + the ordering-metric+// ranking (plus the full-scope pod-status / node-readiness keysets when filtering,+// to intersect all).func (m *module) getTopClusterGroupsAndMetadata(ctx context.Context,orgID valuer.UUID,req *inframonitoringtypes.PostableClusters,-) ([]map[string]string, map[string]map[string]string, error) {+) ([]map[string]string, map[string]map[string]string, map[string]podStatusCounts, *qbtypes.QueryWarnData, map[string]nodeConditionCounts, error) {var (- orderByKey string- metadataMap map[string]map[string]string- allMetricGroups []rankedGroup+ orderByKey string+ metadataMap map[string]map[string]string+ allMetricGroups []rankedGroup+ statusCounts map[string]podStatusCounts+ statusWarning *qbtypes.QueryWarnData+ nodeConditionCounts map[string]nodeConditionCounts+ filter *qbtypes.Filter+ filterByPodStatus []inframonitoringtypes.PodStatus+ filterByNodeReadiness []inframonitoringtypes.NodeCondition)orderByKey = req.OrderBy.Key.Name+ // When filtering by pod status / node readiness, resolve the full-scope+ // keyset(s) concurrently (pageGroups=nil spans all groups under the user+ // filter) to intersect metadata + ranked groups below. Filters compose as AND.+ if req.Filter != nil {+ filter = &req.Filter.Filter+ filterByPodStatus = req.Filter.FilterByPodStatus+ filterByNodeReadiness = req.Filter.FilterByNodeReadiness+ }+g, gCtx := errgroup.WithContext(ctx)g.Go(func() error {@@ -106,12 +124,37 @@ func (m *module) getTopClusterGroupsAndMetadata(return err})+ if len(filterByPodStatus) != 0 {+ g.Go(func() error {+ var err error+ statusCounts, statusWarning, err = m.getPerGroupPodStatusCountsWithReqMetricChecks(gCtx, orgID, req.Start, req.End, filter, req.GroupBy, nil, filterByPodStatus)+ return err+ })+ }++ if len(filterByNodeReadiness) != 0 {+ g.Go(func() error {+ var err error+ nodeConditionCounts, err = m.getPerGroupNodeConditionCounts(gCtx, orgID, req.Start, req.End, filter, req.GroupBy, nil, filterByNodeReadiness)+ return err+ })+ }+if orderByKey == inframonitoringtypes.ClusterNameAttrKey {if err := g.Wait(); err != nil {- return nil, nil, err+ return nil, nil, nil, nil, nil, err+ }+ // Secondary filter: keep only status/readiness-matching groups. A missing+ // metric yields an empty statusCounts, so this correctly empties the result+ // (the caller also surfaces the warning). Filters compose as AND.+ if len(filterByPodStatus) != 0 {+ metadataMap = intersectMap(metadataMap, statusCounts)+ }+ if len(filterByNodeReadiness) != 0 {+ metadataMap = intersectMap(metadataMap, nodeConditionCounts)}pageGroups := inframonitoringtypes.PaginateMetadataByName(metadataMap, req.GroupBy, req.OrderBy.Direction, req.Offset, req.Limit, inframonitoringtypes.ClusterNameAttrKey)- return pageGroups, metadataMap, nil+ return pageGroups, metadataMap, statusCounts, statusWarning, nodeConditionCounts, nil}queryNamesForOrderBy := orderByToClustersQueryNames[orderByKey]@@ -157,10 +200,23 @@ func (m *module) getTopClusterGroupsAndMetadata(})if err := g.Wait(); err != nil {- return nil, nil, err+ return nil, nil, nil, nil, nil, err}- return paginateWithBackfill(allMetricGroups, metadataMap, req.GroupBy, req.Offset, req.Limit), metadataMap, nil+ // Secondary filter: intersect ranked groups + metadata with the status/readiness+ // keyset. A missing metric yields an empty keyset, correctly emptying the result+ // (the caller also surfaces the warning). Filters compose as AND.+ if len(filterByPodStatus) != 0 {+ allMetricGroups = intersectRankedGroups(allMetricGroups, statusCounts)+ metadataMap = intersectMap(metadataMap, statusCounts)+ }+ if len(filterByNodeReadiness) != 0 {+ allMetricGroups = intersectRankedGroups(allMetricGroups, nodeConditionCounts)+ metadataMap = intersectMap(metadataMap, nodeConditionCounts)+ }++ pageGroups := paginateWithBackfill(allMetricGroups, metadataMap, req.GroupBy, req.Offset, req.Limit)+ return pageGroups, metadataMap, statusCounts, statusWarning, nodeConditionCounts, nil}func (m *module) getClustersTableMetadata(ctx context.Context, orgID valuer.UUID, req *inframonitoringtypes.PostableClusters) (map[string]map[string]string, error) {@@ -170,5 +226,9 @@ func (m *module) getClustersTableMetadata(ctx context.Context, orgID valuer.UUIDnonGroupByAttrs = append(nonGroupByAttrs, key)}}- return m.getMetadata(ctx, orgID, clustersTableMetricNamesList, req.GroupBy, nonGroupByAttrs, req.Filter, req.Start, req.End)+ var filter *qbtypes.Filter+ if req.Filter != nil {+ filter = &req.Filter.Filter+ }+ return m.getMetadata(ctx, orgID, clustersTableMetricNamesList, req.GroupBy, nonGroupByAttrs, filter, req.Start, req.End)}diff --git a/pkg/modules/inframonitoring/implinframonitoring/containers.go b/pkg/modules/inframonitoring/implinframonitoring/containers.goindex e3bcd67d25..087073cc14 100644--- a/pkg/modules/inframonitoring/implinframonitoring/containers.go+++ b/pkg/modules/inframonitoring/implinframonitoring/containers.go@@ -139,20 +139,34 @@ func buildContainerRecords(return records}+// getTopContainerGroupsAndMetadata concurrently fetches metadata + the ordering-metric+// ranking (plus the full-scope container-status keyset when filtering, to intersect both).func (m *module) getTopContainerGroupsAndMetadata(ctx context.Context,orgID valuer.UUID,req *inframonitoringtypes.PostableContainers,-) ([]map[string]string, map[string]map[string]string, error) {+) ([]map[string]string, map[string]map[string]string, map[string]containerStatusCounts, *qbtypes.QueryWarnData, error) {var (- orderByKey string- metadataMap map[string]map[string]string- allMetricGroups []rankedGroup+ orderByKey string+ metadataMap map[string]map[string]string+ allMetricGroups []rankedGroup+ statusCounts map[string]containerStatusCounts+ statusWarning *qbtypes.QueryWarnData+ filter *qbtypes.Filter+ filterByContainerStatus []inframonitoringtypes.ContainerStatus)orderByKey = req.OrderBy.Key.Name+ // When filtering by container status, resolve the full-scope status keyset+ // concurrently (pageGroups=nil spans all groups under the user filter) so it+ // can intersect metadata + ranked groups below.+ if req.Filter != nil {+ filter = &req.Filter.Filter+ filterByContainerStatus = req.Filter.FilterByContainerStatus+ }+g, gCtx := errgroup.WithContext(ctx)g.Go(func() error {@@ -161,12 +175,26 @@ func (m *module) getTopContainerGroupsAndMetadata(return err})+ if len(filterByContainerStatus) != 0 {+ g.Go(func() error {+ var err error+ statusCounts, statusWarning, err = m.getPerGroupContainerStatusCountsWithReqMetricChecks(gCtx, orgID, req.Start, req.End, filter, req.GroupBy, nil, filterByContainerStatus)+ return err+ })+ }+if orderByKey == inframonitoringtypes.ContainerNameAttrKey {if err := g.Wait(); err != nil {- return nil, nil, err+ return nil, nil, nil, nil, err+ }+ // Secondary filter: keep only status-matching groups. A missing metric+ // yields an empty statusCounts, so this correctly empties the result+ // (the caller also surfaces the warning).+ if len(filterByContainerStatus) != 0 {+ metadataMap = intersectMap(metadataMap, statusCounts)}pageGroups := inframonitoringtypes.PaginateMetadataByName(metadataMap, req.GroupBy, req.OrderBy.Direction, req.Offset, req.Limit, inframonitoringtypes.ContainerNameAttrKey)- return pageGroups, metadataMap, nil+ return pageGroups, metadataMap, statusCounts, statusWarning, nil}queryNamesForOrderBy := orderByToContainersQueryNames[orderByKey]@@ -212,10 +240,19 @@ func (m *module) getTopContainerGroupsAndMetadata(})if err := g.Wait(); err != nil {- return nil, nil, err+ return nil, nil, nil, nil, err}- return paginateWithBackfill(allMetricGroups, metadataMap, req.GroupBy, req.Offset, req.Limit), metadataMap, nil+ // Secondary filter: intersect ranked groups + metadata with the status keyset.+ // A missing metric yields an empty statusCounts, correctly emptying the result+ // (the caller also surfaces the warning).+ if len(filterByContainerStatus) != 0 {+ allMetricGroups = intersectRankedGroups(allMetricGroups, statusCounts)+ metadataMap = intersectMap(metadataMap, statusCounts)+ }++ pageGroups := paginateWithBackfill(allMetricGroups, metadataMap, req.GroupBy, req.Offset, req.Limit)+ return pageGroups, metadataMap, statusCounts, statusWarning, nil}func (m *module) getContainersTableMetadata(ctx context.Context, orgID valuer.UUID, req *inframonitoringtypes.PostableContainers) (map[string]map[string]string, error) {@@ -225,7 +262,11 @@ func (m *module) getContainersTableMetadata(ctx context.Context, orgID valuer.UUnonGroupByAttrs = append(nonGroupByAttrs, key)}}- return m.getMetadata(ctx, orgID, containersTableMetricNamesList, req.GroupBy, nonGroupByAttrs, req.Filter, req.Start, req.End)+ var filter *qbtypes.Filter+ if req.Filter != nil {+ filter = &req.Filter.Filter+ }+ return m.getMetadata(ctx, orgID, containersTableMetricNamesList, req.GroupBy, nonGroupByAttrs, filter, req.Start, req.End)}// getPerGroupContainerStatusCountsWithReqMetricChecks gates@@ -241,6 +282,7 @@ func (m *module) getPerGroupContainerStatusCountsWithReqMetricChecks(filter *qbtypes.Filter,groupBy []qbtypes.GroupByKey,pageGroups []map[string]string,+ filterByContainerStatus []inframonitoringtypes.ContainerStatus,) (map[string]containerStatusCounts, *qbtypes.QueryWarnData, error) {present, err := m.getMetricsExistence(ctx, containerStatusMetricNamesList)if err != nil {@@ -266,13 +308,28 @@ func (m *module) getPerGroupContainerStatusCountsWithReqMetricChecks(return map[string]containerStatusCounts{}, warning, nil}- counts, err := m.getPerGroupContainerStatusCounts(ctx, orgID, start, end, filter, groupBy, pageGroups)+ counts, err := m.getPerGroupContainerStatusCounts(ctx, orgID, start, end, filter, groupBy, pageGroups, filterByContainerStatus)if err != nil {return nil, nil, err}return counts, nil, nil}+// applyContainerStatusFilter adds the display-status push-down (lower(display_status)+// IN (...)) to the outer count builder. valuer lowercases the wire value while+// display_status is kubectl-cased, so we compare lower() on both. No-op when the+// requested set is empty.+func applyContainerStatusFilter(cb *sqlbuilder.SelectBuilder, filterByContainerStatus []inframonitoringtypes.ContainerStatus) {+ if len(filterByContainerStatus) == 0 {+ return+ }+ vals := make([]string, len(filterByContainerStatus))+ for i, c := range filterByContainerStatus {+ vals[i] = c.StringValue()+ }+ cb.Where(cb.In("lower(display_status)", sqlbuilder.List(vals)))+}+// getPerGroupContainerStatusCounts computes per-group counts of distinct// containers bucketed by their latest kubectl-style display status in window.// Caller must ensure the required metrics exist@@ -297,8 +354,11 @@ func (m *module) getPerGroupContainerStatusCounts(filter *qbtypes.Filter,groupBy []qbtypes.GroupByKey,pageGroups []map[string]string,+ filterByContainerStatus []inframonitoringtypes.ContainerStatus,) (map[string]containerStatusCounts, error) {- if len(pageGroups) == 0 || len(groupBy) == 0 {+ // Empty pageGroups means "span all under user filter", allowed only in+ // full-scope mode (filtering by status). Otherwise it's an empty page.+ if len(groupBy) == 0 || (len(pageGroups) == 0 && len(filterByContainerStatus) == 0) {return map[string]containerStatusCounts{}, nil}@@ -482,11 +542,15 @@ func (m *module) getPerGroupContainerStatusCounts(countGroupBy = append(countGroupBy, col)}countSelectCols = append(countSelectCols, statusCountCols...)- countSQL := fmt.Sprintf(- "SELECT %s FROM container_status GROUP BY %s",- strings.Join(countSelectCols, ", "),- strings.Join(countGroupBy, ", "),- )++ // Outer count query. Built with sqlbuilder so the status push-down uses a+ // proper IN (keep only containers whose display status is in the requested set).+ countBuilder := sqlbuilder.NewSelectBuilder()+ countBuilder.Select(countSelectCols...)+ countBuilder.From("container_status")+ applyContainerStatusFilter(countBuilder, filterByContainerStatus)+ countBuilder.GroupBy(countGroupBy...)+ countSQL, countArgs := countBuilder.BuildWithFlavor(sqlbuilder.ClickHouse)// Combine CTEs + outer. Arg order mirrors CTE declaration order.cteFragments := []string{@@ -499,7 +563,7 @@ func (m *module) getPerGroupContainerStatusCounts(finalSQL := querybuilder.CombineCTEs(cteFragments) + countSQLfinalArgs := querybuilder.PrependArgs([][]any{stateFpsArgs, containerStateArgs, reasonFpsArgs, reasonInnerArgs,- }, nil)+ }, countArgs)rows, err := m.telemetryStore.ClickhouseDB().Query(ctx, finalSQL, finalArgs...)if err != nil {diff --git a/pkg/modules/inframonitoring/implinframonitoring/containers_test.go b/pkg/modules/inframonitoring/implinframonitoring/containers_test.gonew file mode 100644index 0000000000..dc11ab7a70--- /dev/null+++ b/pkg/modules/inframonitoring/implinframonitoring/containers_test.go@@ -0,0 +1,59 @@+package implinframonitoring++import (+ "strings"+ "testing"++ "github.com/SigNoz/signoz/pkg/types/inframonitoringtypes"+ "github.com/huandu/go-sqlbuilder"+ "github.com/stretchr/testify/assert"+)++func TestApplyContainerStatusFilter(t *testing.T) {+ tests := []struct {+ name string+ statuses []inframonitoringtypes.ContainerStatus+ wantWhere bool+ wantArgs []any+ }{+ {+ name: "empty set yields no clause",+ statuses: nil,+ wantWhere: false,+ wantArgs: nil,+ },+ {+ name: "single status pushes lowercased arg via IN",+ statuses: []inframonitoringtypes.ContainerStatus{inframonitoringtypes.ContainerStatusRunning},+ wantWhere: true,+ wantArgs: []any{"running"},+ },+ {+ name: "multiple statuses push lowercased args via IN",+ statuses: []inframonitoringtypes.ContainerStatus{+ inframonitoringtypes.ContainerStatusRunning,+ inframonitoringtypes.ContainerStatusCrashLoopBackOff,+ },+ wantWhere: true,+ wantArgs: []any{"running", "crashloopbackoff"},+ },+ }++ for _, tt := range tests {+ t.Run(tt.name, func(t *testing.T) {+ cb := sqlbuilder.NewSelectBuilder()+ cb.Select("pod_uid")+ cb.From("container_status")+ applyContainerStatusFilter(cb, tt.statuses)+ sql, args := cb.BuildWithFlavor(sqlbuilder.ClickHouse)++ hasWhere := strings.Contains(sql, "lower(display_status) IN (")+ assert.Equal(t, tt.wantWhere, hasWhere)+ if len(tt.wantArgs) == 0 {+ assert.Empty(t, args)+ } else {+ assert.Equal(t, tt.wantArgs, args)+ }+ })+ }+}diff --git a/pkg/modules/inframonitoring/implinframonitoring/daemonsets.go b/pkg/modules/inframonitoring/implinframonitoring/daemonsets.goindex 73a520a726..86f668bb97 100644--- a/pkg/modules/inframonitoring/implinframonitoring/daemonsets.go+++ b/pkg/modules/inframonitoring/implinframonitoring/daemonsets.go@@ -90,20 +90,34 @@ func buildDaemonSetRecords(return records}+// getTopDaemonSetGroupsAndMetadata concurrently fetches metadata + the ordering-metric+// ranking (plus the full-scope pod-status keyset when filtering, to intersect both).func (m *module) getTopDaemonSetGroupsAndMetadata(ctx context.Context,orgID valuer.UUID,req *inframonitoringtypes.PostableDaemonSets,-) ([]map[string]string, map[string]map[string]string, error) {+) ([]map[string]string, map[string]map[string]string, map[string]podStatusCounts, *qbtypes.QueryWarnData, error) {var (- orderByKey string- metadataMap map[string]map[string]string- allMetricGroups []rankedGroup+ orderByKey string+ metadataMap map[string]map[string]string+ allMetricGroups []rankedGroup+ statusCounts map[string]podStatusCounts+ statusWarning *qbtypes.QueryWarnData+ filter *qbtypes.Filter+ filterByPodStatus []inframonitoringtypes.PodStatus)orderByKey = req.OrderBy.Key.Name+ // When filtering by pod status, resolve the full-scope status keyset+ // concurrently (pageGroups=nil spans all groups under the user filter) so it+ // can intersect metadata + ranked groups below.+ if req.Filter != nil {+ filter = &req.Filter.Filter+ filterByPodStatus = req.Filter.FilterByPodStatus+ }+g, gCtx := errgroup.WithContext(ctx)g.Go(func() error {@@ -112,12 +126,26 @@ func (m *module) getTopDaemonSetGroupsAndMetadata(return err})+ if len(filterByPodStatus) != 0 {+ g.Go(func() error {+ var err error+ statusCounts, statusWarning, err = m.getPerGroupPodStatusCountsWithReqMetricChecks(gCtx, orgID, req.Start, req.End, filter, req.GroupBy, nil, filterByPodStatus)+ return err+ })+ }+if orderByKey == inframonitoringtypes.DaemonSetNameAttrKey {if err := g.Wait(); err != nil {- return nil, nil, err+ return nil, nil, nil, nil, err+ }+ // Secondary filter: keep only status-matching groups. A missing metric+ // yields an empty statusCounts, so this correctly empties the result+ // (the caller also surfaces the warning).+ if len(filterByPodStatus) != 0 {+ metadataMap = intersectMap(metadataMap, statusCounts)}pageGroups := inframonitoringtypes.PaginateMetadataByName(metadataMap, req.GroupBy, req.OrderBy.Direction, req.Offset, req.Limit, inframonitoringtypes.DaemonSetNameAttrKey)- return pageGroups, metadataMap, nil+ return pageGroups, metadataMap, statusCounts, statusWarning, nil}queryNamesForOrderBy := orderByToDaemonSetsQueryNames[orderByKey]@@ -163,10 +191,19 @@ func (m *module) getTopDaemonSetGroupsAndMetadata(})if err := g.Wait(); err != nil {- return nil, nil, err+ return nil, nil, nil, nil, err+ }++ // Secondary filter: intersect ranked groups + metadata with the status keyset.+ // A missing metric yields an empty statusCounts, correctly emptying the result+ // (the caller also surfaces the warning).+ if len(filterByPodStatus) != 0 {+ allMetricGroups = intersectRankedGroups(allMetricGroups, statusCounts)+ metadataMap = intersectMap(metadataMap, statusCounts)}- return paginateWithBackfill(allMetricGroups, metadataMap, req.GroupBy, req.Offset, req.Limit), metadataMap, nil+ pageGroups := paginateWithBackfill(allMetricGroups, metadataMap, req.GroupBy, req.Offset, req.Limit)+ return pageGroups, metadataMap, statusCounts, statusWarning, nil}func (m *module) getDaemonSetsTableMetadata(ctx context.Context, orgID valuer.UUID, req *inframonitoringtypes.PostableDaemonSets) (map[string]map[string]string, error) {@@ -176,5 +213,9 @@ func (m *module) getDaemonSetsTableMetadata(ctx context.Context, orgID valuer.UUnonGroupByAttrs = append(nonGroupByAttrs, key)}}- return m.getMetadata(ctx, orgID, daemonSetsTableMetricNamesList, req.GroupBy, nonGroupByAttrs, req.Filter, req.Start, req.End)+ var filter *qbtypes.Filter+ if req.Filter != nil {+ filter = &req.Filter.Filter+ }+ return m.getMetadata(ctx, orgID, daemonSetsTableMetricNamesList, req.GroupBy, nonGroupByAttrs, filter, req.Start, req.End)}diff --git a/pkg/modules/inframonitoring/implinframonitoring/deployments.go b/pkg/modules/inframonitoring/implinframonitoring/deployments.goindex 7204d897e3..f3e84fc944 100644--- a/pkg/modules/inframonitoring/implinframonitoring/deployments.go+++ b/pkg/modules/inframonitoring/implinframonitoring/deployments.go@@ -82,20 +82,34 @@ func buildDeploymentRecords(return records}+// getTopDeploymentGroupsAndMetadata concurrently fetches metadata + the ordering-metric+// ranking (plus the full-scope pod-status keyset when filtering, to intersect both).func (m *module) getTopDeploymentGroupsAndMetadata(ctx context.Context,orgID valuer.UUID,req *inframonitoringtypes.PostableDeployments,-) ([]map[string]string, map[string]map[string]string, error) {+) ([]map[string]string, map[string]map[string]string, map[string]podStatusCounts, *qbtypes.QueryWarnData, error) {var (- orderByKey string- metadataMap map[string]map[string]string- allMetricGroups []rankedGroup+ orderByKey string+ metadataMap map[string]map[string]string+ allMetricGroups []rankedGroup+ statusCounts map[string]podStatusCounts+ statusWarning *qbtypes.QueryWarnData+ filter *qbtypes.Filter+ filterByPodStatus []inframonitoringtypes.PodStatus)orderByKey = req.OrderBy.Key.Name+ // When filtering by pod status, resolve the full-scope status keyset+ // concurrently (pageGroups=nil spans all groups under the user filter) so it+ // can intersect metadata + ranked groups below.+ if req.Filter != nil {+ filter = &req.Filter.Filter+ filterByPodStatus = req.Filter.FilterByPodStatus+ }+g, gCtx := errgroup.WithContext(ctx)g.Go(func() error {@@ -104,12 +118,26 @@ func (m *module) getTopDeploymentGroupsAndMetadata(return err})+ if len(filterByPodStatus) != 0 {+ g.Go(func() error {+ var err error+ statusCounts, statusWarning, err = m.getPerGroupPodStatusCountsWithReqMetricChecks(gCtx, orgID, req.Start, req.End, filter, req.GroupBy, nil, filterByPodStatus)+ return err+ })+ }+if orderByKey == inframonitoringtypes.DeploymentNameAttrKey {if err := g.Wait(); err != nil {- return nil, nil, err+ return nil, nil, nil, nil, err+ }+ // Secondary filter: keep only status-matching groups. A missing metric+ // yields an empty statusCounts, so this correctly empties the result+ // (the caller also surfaces the warning).+ if len(filterByPodStatus) != 0 {+ metadataMap = intersectMap(metadataMap, statusCounts)}pageGroups := inframonitoringtypes.PaginateMetadataByName(metadataMap, req.GroupBy, req.OrderBy.Direction, req.Offset, req.Limit, inframonitoringtypes.DeploymentNameAttrKey)- return pageGroups, metadataMap, nil+ return pageGroups, metadataMap, statusCounts, statusWarning, nil}queryNamesForOrderBy := orderByToDeploymentsQueryNames[orderByKey]@@ -155,10 +183,19 @@ func (m *module) getTopDeploymentGroupsAndMetadata(})if err := g.Wait(); err != nil {- return nil, nil, err+ return nil, nil, nil, nil, err+ }++ // Secondary filter: intersect ranked groups + metadata with the status keyset.+ // A missing metric yields an empty statusCounts, correctly emptying the result+ // (the caller also surfaces the warning).+ if len(filterByPodStatus) != 0 {+ allMetricGroups = intersectRankedGroups(allMetricGroups, statusCounts)+ metadataMap = intersectMap(metadataMap, statusCounts)}- return paginateWithBackfill(allMetricGroups, metadataMap, req.GroupBy, req.Offset, req.Limit), metadataMap, nil+ pageGroups := paginateWithBackfill(allMetricGroups, metadataMap, req.GroupBy, req.Offset, req.Limit)+ return pageGroups, metadataMap, statusCounts, statusWarning, nil}func (m *module) getDeploymentsTableMetadata(ctx context.Context, orgID valuer.UUID, req *inframonitoringtypes.PostableDeployments) (map[string]map[string]string, error) {@@ -168,5 +205,9 @@ func (m *module) getDeploymentsTableMetadata(ctx context.Context, orgID valuer.UnonGroupByAttrs = append(nonGroupByAttrs, key)}}- return m.getMetadata(ctx, orgID, deploymentsTableMetricNamesList, req.GroupBy, nonGroupByAttrs, req.Filter, req.Start, req.End)+ var filter *qbtypes.Filter+ if req.Filter != nil {+ filter = &req.Filter.Filter+ }+ return m.getMetadata(ctx, orgID, deploymentsTableMetricNamesList, req.GroupBy, nonGroupByAttrs, filter, req.Start, req.End)}diff --git a/pkg/modules/inframonitoring/implinframonitoring/helpers.go b/pkg/modules/inframonitoring/implinframonitoring/helpers.goindex 90d5b111b5..72d8930586 100644--- a/pkg/modules/inframonitoring/implinframonitoring/helpers.go+++ b/pkg/modules/inframonitoring/implinframonitoring/helpers.go@@ -63,6 +63,33 @@ func compositeKeyFromLabels(labels map[string]string, groupBy []qbtypes.GroupByKreturn compositeKeyFromList(parts)}+// intersectMap returns the entries of m whose key is present in keep (a new+// map). keep's value type is irrelevant — only its keys are read — so a+// per-group counts map (already filtered by the SQL push-down) can be passed+// directly. Used to trim metadataMap to the status-matching groups.+func intersectMap[V any, K any](m map[string]V, keep map[string]K) map[string]V {+ out := make(map[string]V, len(m))+ for k, v := range m {+ if _, ok := keep[k]; ok {+ out[k] = v+ }+ }+ return out+}++// intersectRankedGroups returns the ranked groups whose compositeKey is present+// in keep, preserving order. Keeps status-unmatched groups out of the ranked+// page slots.+func intersectRankedGroups[K any](groups []rankedGroup, keep map[string]K) []rankedGroup {+ out := make([]rankedGroup, 0, len(groups))+ for _, g := range groups {+ if _, ok := keep[g.compositeKey]; ok {+ out = append(out, g)+ }+ }+ return out+}+// parseAndSortGroups extracts group label maps from a ScalarData response and// sorts them by the ranking query's aggregation value.func parseAndSortGroups(diff --git a/pkg/modules/inframonitoring/implinframonitoring/helpers_test.go b/pkg/modules/inframonitoring/implinframonitoring/helpers_test.goindex 6f70f3fef9..95fe2c8add 100644--- a/pkg/modules/inframonitoring/implinframonitoring/helpers_test.go+++ b/pkg/modules/inframonitoring/implinframonitoring/helpers_test.go@@ -5,6 +5,7 @@ import (qbtypes "github.com/SigNoz/signoz/pkg/types/querybuildertypes/querybuildertypesv5""github.com/SigNoz/signoz/pkg/types/telemetrytypes"+ "github.com/stretchr/testify/assert")func groupByKey(name string) qbtypes.GroupByKey {@@ -88,10 +89,7 @@ func TestIsKeyInGroupByAttrs(t *testing.T) {for _, tt := range tests {t.Run(tt.name, func(t *testing.T) {got := isKeyInGroupByAttrs(tt.groupByAttrs, tt.key)- if got != tt.expectedFound {- t.Errorf("isKeyInGroupByAttrs(%v, %q) = %v, want %v",- tt.groupByAttrs, tt.key, got, tt.expectedFound)- }+ assert.Equal(t, tt.expectedFound, got)})}}@@ -156,10 +154,7 @@ func TestMergeFilterExpressions(t *testing.T) {for _, tt := range tests {t.Run(tt.name, func(t *testing.T) {got := mergeFilterExpressions(tt.queryFilterExpr, tt.reqFilterExpr)- if got != tt.expected {- t.Errorf("mergeFilterExpressions(%q, %q) = %q, want %q",- tt.queryFilterExpr, tt.reqFilterExpr, got, tt.expected)- }+ assert.Equal(t, tt.expected, got)})}}@@ -205,10 +200,7 @@ func TestCompositeKeyFromList(t *testing.T) {for _, tt := range tests {t.Run(tt.name, func(t *testing.T) {got := compositeKeyFromList(tt.parts)- if got != tt.expected {- t.Errorf("compositeKeyFromList(%v) = %q, want %q",- tt.parts, got, tt.expected)- }+ assert.Equal(t, tt.expected, got)})}}@@ -376,10 +368,81 @@ func TestCompositeKeyFromLabels(t *testing.T) {for _, tt := range tests {t.Run(tt.name, func(t *testing.T) {got := compositeKeyFromLabels(tt.labels, tt.groupBy)- if got != tt.expected {- t.Errorf("compositeKeyFromLabels(%v, %v) = %q, want %q",- tt.labels, tt.groupBy, got, tt.expected)+ assert.Equal(t, tt.expected, got)+ })+ }+}++func TestIntersectMap(t *testing.T) {+ tests := []struct {+ name string+ m map[string]int+ keep map[string]podStatusCounts+ expected map[string]int+ }{+ {+ name: "keep subset",+ m: map[string]int{"a": 1, "b": 2, "c": 3},+ keep: map[string]podStatusCounts{"a": {}, "c": {}},+ expected: map[string]int{"a": 1, "c": 3},+ },+ {+ name: "empty keep drops everything",+ m: map[string]int{"a": 1, "b": 2},+ keep: map[string]podStatusCounts{},+ expected: map[string]int{},+ },+ {+ name: "keep key absent from m is ignored",+ m: map[string]int{"a": 1},+ keep: map[string]podStatusCounts{"a": {}, "z": {}},+ expected: map[string]int{"a": 1},+ },+ }++ for _, tt := range tests {+ t.Run(tt.name, func(t *testing.T) {+ got := intersectMap(tt.m, tt.keep)+ assert.Equal(t, tt.expected, got)+ })+ }+}++func TestIntersectRankedGroups(t *testing.T) {+ groups := []rankedGroup{+ {compositeKey: "a", value: 3},+ {compositeKey: "b", value: 2},+ {compositeKey: "c", value: 1},+ }++ tests := []struct {+ name string+ groups []rankedGroup+ keep map[string]podStatusCounts+ expected []string // compositeKeys in order+ }{+ {+ name: "preserves order, drops non-matching",+ groups: groups,+ keep: map[string]podStatusCounts{"a": {}, "c": {}},+ expected: []string{"a", "c"},+ },+ {+ name: "empty keep drops all",+ groups: groups,+ keep: map[string]podStatusCounts{},+ expected: []string{},+ },+ }++ for _, tt := range tests {+ t.Run(tt.name, func(t *testing.T) {+ got := intersectRankedGroups(tt.groups, tt.keep)+ gotKeys := make([]string, 0, len(got))+ for _, g := range got {+ gotKeys = append(gotKeys, g.compositeKey)}+ assert.Equal(t, tt.expected, gotKeys)})}}diff --git a/pkg/modules/inframonitoring/implinframonitoring/jobs.go b/pkg/modules/inframonitoring/implinframonitoring/jobs.goindex e18c822ab0..0720170c70 100644--- a/pkg/modules/inframonitoring/implinframonitoring/jobs.go+++ b/pkg/modules/inframonitoring/implinframonitoring/jobs.go@@ -90,20 +90,34 @@ func buildJobRecords(return records}+// getTopJobGroupsAndMetadata concurrently fetches metadata + the ordering-metric+// ranking (plus the full-scope pod-status keyset when filtering, to intersect both).func (m *module) getTopJobGroupsAndMetadata(ctx context.Context,orgID valuer.UUID,req *inframonitoringtypes.PostableJobs,-) ([]map[string]string, map[string]map[string]string, error) {+) ([]map[string]string, map[string]map[string]string, map[string]podStatusCounts, *qbtypes.QueryWarnData, error) {var (- orderByKey string- metadataMap map[string]map[string]string- allMetricGroups []rankedGroup+ orderByKey string+ metadataMap map[string]map[string]string+ allMetricGroups []rankedGroup+ statusCounts map[string]podStatusCounts+ statusWarning *qbtypes.QueryWarnData+ filter *qbtypes.Filter+ filterByPodStatus []inframonitoringtypes.PodStatus)orderByKey = req.OrderBy.Key.Name+ // When filtering by pod status, resolve the full-scope status keyset+ // concurrently (pageGroups=nil spans all groups under the user filter) so it+ // can intersect metadata + ranked groups below.+ if req.Filter != nil {+ filter = &req.Filter.Filter+ filterByPodStatus = req.Filter.FilterByPodStatus+ }+g, gCtx := errgroup.WithContext(ctx)g.Go(func() error {@@ -112,12 +126,26 @@ func (m *module) getTopJobGroupsAndMetadata(return err})+ if len(filterByPodStatus) != 0 {+ g.Go(func() error {+ var err error+ statusCounts, statusWarning, err = m.getPerGroupPodStatusCountsWithReqMetricChecks(gCtx, orgID, req.Start, req.End, filter, req.GroupBy, nil, filterByPodStatus)+ return err+ })+ }+if orderByKey == inframonitoringtypes.JobNameAttrKey {if err := g.Wait(); err != nil {- return nil, nil, err+ return nil, nil, nil, nil, err+ }+ // Secondary filter: keep only status-matching groups. A missing metric+ // yields an empty statusCounts, so this correctly empties the result+ // (the caller also surfaces the warning).+ if len(filterByPodStatus) != 0 {+ metadataMap = intersectMap(metadataMap, statusCounts)}pageGroups := inframonitoringtypes.PaginateMetadataByName(metadataMap, req.GroupBy, req.OrderBy.Direction, req.Offset, req.Limit, inframonitoringtypes.JobNameAttrKey)- return pageGroups, metadataMap, nil+ return pageGroups, metadataMap, statusCounts, statusWarning, nil}queryNamesForOrderBy := orderByToJobsQueryNames[orderByKey]@@ -163,10 +191,19 @@ func (m *module) getTopJobGroupsAndMetadata(})if err := g.Wait(); err != nil {- return nil, nil, err+ return nil, nil, nil, nil, err+ }++ // Secondary filter: intersect ranked groups + metadata with the status keyset.+ // A missing metric yields an empty statusCounts, correctly emptying the result+ // (the caller also surfaces the warning).+ if len(filterByPodStatus) != 0 {+ allMetricGroups = intersectRankedGroups(allMetricGroups, statusCounts)+ metadataMap = intersectMap(metadataMap, statusCounts)}- return paginateWithBackfill(allMetricGroups, metadataMap, req.GroupBy, req.Offset, req.Limit), metadataMap, nil+ pageGroups := paginateWithBackfill(allMetricGroups, metadataMap, req.GroupBy, req.Offset, req.Limit)+ return pageGroups, metadataMap, statusCounts, statusWarning, nil}func (m *module) getJobsTableMetadata(ctx context.Context, orgID valuer.UUID, req *inframonitoringtypes.PostableJobs) (map[string]map[string]string, error) {@@ -176,5 +213,9 @@ func (m *module) getJobsTableMetadata(ctx context.Context, orgID valuer.UUID, renonGroupByAttrs = append(nonGroupByAttrs, key)}}- return m.getMetadata(ctx, orgID, jobsTableMetricNamesList, req.GroupBy, nonGroupByAttrs, req.Filter, req.Start, req.End)+ var filter *qbtypes.Filter+ if req.Filter != nil {+ filter = &req.Filter.Filter+ }+ return m.getMetadata(ctx, orgID, jobsTableMetricNamesList, req.GroupBy, nonGroupByAttrs, filter, req.Start, req.End)}diff --git a/pkg/modules/inframonitoring/implinframonitoring/module.go b/pkg/modules/inframonitoring/implinframonitoring/module.goindex 9ca2b9ff68..d00d8daf19 100644--- a/pkg/modules/inframonitoring/implinframonitoring/module.go+++ b/pkg/modules/inframonitoring/implinframonitoring/module.go@@ -286,11 +286,36 @@ func (m *module) ListPods(ctx context.Context, orgID valuer.UUID, req *inframonireturn resp, nil}- pageGroups, metadataMap, err := m.getTopPodGroupsAndMetadata(ctx, orgID, req)+ var (+ filterExpr string+ podFilter *qbtypes.Filter+ filterByPodStatus []inframonitoringtypes.PodStatus+ queryResp *qbtypes.QueryRangeResponse+ restartCounts map[string]int64+ )++ if req.Filter != nil {+ filterExpr = req.Filter.Expression+ podFilter = &req.Filter.Filter+ filterByPodStatus = req.Filter.FilterByPodStatus+ }++ // getTopPodGroupsAndMetadata fetches metadata + ranking (+ full-scope pod+ // status when filtering) concurrently, intersecting metadata/ranked groups+ // against the status keyset. It returns the keyset + its warning.+ pageGroups, metadataMap, statusCounts, statusWarning, err := m.getTopPodGroupsAndMetadata(ctx, orgID, req)if err != nil {return nil, err}+ // Required metric missing while filtering: surface the warning + empty result.+ if len(filterByPodStatus) != 0 && statusWarning != nil {+ resp.Warning = statusWarning+ resp.Records = []inframonitoringtypes.PodRecord{}+ resp.Total = 0+ return resp, nil+ }+resp.Total = len(metadataMap)if len(pageGroups) == 0 {@@ -298,20 +323,8 @@ func (m *module) ListPods(ctx context.Context, orgID valuer.UUID, req *inframonireturn resp, nil}- filterExpr := ""- if req.Filter != nil {- filterExpr = req.Filter.Expression- }-fullQueryReq := buildFullQueryRequest(req.Start, req.End, filterExpr, req.GroupBy, pageGroups, m.newPodsTableListQuery())- var (- queryResp *qbtypes.QueryRangeResponse- statusCounts map[string]podStatusCounts- statusWarning *qbtypes.QueryWarnData- restartCounts map[string]int64- )-g, gCtx := errgroup.WithContext(ctx)g.Go(func() error {@@ -321,14 +334,18 @@ func (m *module) ListPods(ctx context.Context, orgID valuer.UUID, req *inframoni})g.Go(func() error {var err error- statusCounts, statusWarning, err = m.getPerGroupPodStatusCountsWithReqMetricChecks(gCtx, orgID, req.Start, req.End, req.Filter, req.GroupBy, pageGroups)- return err- })- g.Go(func() error {- var err error- restartCounts, err = m.getPerGroupPodRestartCounts(gCtx, orgID, req.Start, req.End, req.Filter, req.GroupBy, pageGroups)+ restartCounts, err = m.getPerGroupPodRestartCounts(gCtx, orgID, req.Start, req.End, podFilter, req.GroupBy, pageGroups)return err})+ // When filtering, statusCounts already holds the full-scope map (a superset+ // of the page); otherwise compute it page-scoped here.+ if len(filterByPodStatus) == 0 {+ g.Go(func() error {+ var err error+ statusCounts, statusWarning, err = m.getPerGroupPodStatusCountsWithReqMetricChecks(gCtx, orgID, req.Start, req.End, podFilter, req.GroupBy, pageGroups, nil)+ return err+ })+ }if err := g.Wait(); err != nil {return nil, err@@ -379,11 +396,37 @@ func (m *module) ListContainers(ctx context.Context, orgID valuer.UUID, req *infreturn resp, nil}- pageGroups, metadataMap, err := m.getTopContainerGroupsAndMetadata(ctx, orgID, req)+ var (+ filterExpr string+ containerFilter *qbtypes.Filter+ filterByContainerStatus []inframonitoringtypes.ContainerStatus+ queryResp *qbtypes.QueryRangeResponse+ restartCounts map[string]int64+ readyCounts map[string]containerReadyCounts+ )++ if req.Filter != nil {+ filterExpr = req.Filter.Expression+ containerFilter = &req.Filter.Filter+ filterByContainerStatus = req.Filter.FilterByContainerStatus+ }++ // getTopContainerGroupsAndMetadata fetches metadata + ranking (+ full-scope+ // container status when filtering) concurrently, intersecting metadata/ranked+ // groups against the status keyset. It returns the keyset + its warning.+ pageGroups, metadataMap, statusCounts, statusWarning, err := m.getTopContainerGroupsAndMetadata(ctx, orgID, req)if err != nil {return nil, err}+ // Required metric missing while filtering: surface the warning + empty result.+ if len(filterByContainerStatus) != 0 && statusWarning != nil {+ resp.Warning = statusWarning+ resp.Records = []inframonitoringtypes.ContainerRecord{}+ resp.Total = 0+ return resp, nil+ }+resp.Total = len(metadataMap)if len(pageGroups) == 0 {@@ -391,21 +434,8 @@ func (m *module) ListContainers(ctx context.Context, orgID valuer.UUID, req *infreturn resp, nil}- filterExpr := ""- if req.Filter != nil {- filterExpr = req.Filter.Expression- }-fullQueryReq := buildFullQueryRequest(req.Start, req.End, filterExpr, req.GroupBy, pageGroups, m.newContainersTableListQuery())- var (- queryResp *qbtypes.QueryRangeResponse- statusCounts map[string]containerStatusCounts- statusWarning *qbtypes.QueryWarnData- restartCounts map[string]int64- readyCounts map[string]containerReadyCounts- )-g, gCtx := errgroup.WithContext(ctx)g.Go(func() error {@@ -415,19 +445,23 @@ func (m *module) ListContainers(ctx context.Context, orgID valuer.UUID, req *inf})g.Go(func() error {var err error- statusCounts, statusWarning, err = m.getPerGroupContainerStatusCountsWithReqMetricChecks(gCtx, orgID, req.Start, req.End, req.Filter, req.GroupBy, pageGroups)+ restartCounts, err = m.getPerGroupContainerRestartCounts(gCtx, orgID, req.Start, req.End, containerFilter, req.GroupBy, pageGroups)return err})g.Go(func() error {var err error- restartCounts, err = m.getPerGroupContainerRestartCounts(gCtx, orgID, req.Start, req.End, req.Filter, req.GroupBy, pageGroups)- return err- })- g.Go(func() error {- var err error- readyCounts, err = m.getPerGroupContainerReadyCounts(gCtx, orgID, req.Start, req.End, req.Filter, req.GroupBy, pageGroups)+ readyCounts, err = m.getPerGroupContainerReadyCounts(gCtx, orgID, req.Start, req.End, containerFilter, req.GroupBy, pageGroups)return err})+ // When filtering, statusCounts already holds the full-scope map (a superset+ // of the page); otherwise compute it page-scoped here.+ if len(filterByContainerStatus) == 0 {+ g.Go(func() error {+ var err error+ statusCounts, statusWarning, err = m.getPerGroupContainerStatusCountsWithReqMetricChecks(gCtx, orgID, req.Start, req.End, containerFilter, req.GroupBy, pageGroups, nil)+ return err+ })+ }if err := g.Wait(); err != nil {return nil, err@@ -478,11 +512,37 @@ func (m *module) ListNodes(ctx context.Context, orgID valuer.UUID, req *inframonreturn resp, nil}- pageGroups, metadataMap, err := m.getTopNodeGroupsAndMetadata(ctx, orgID, req)+ var (+ filterExpr string+ nodeFilter *qbtypes.Filter+ filterByPodStatus []inframonitoringtypes.PodStatus+ filterByNodeReadiness []inframonitoringtypes.NodeCondition+ queryResp *qbtypes.QueryRangeResponse+ )++ if req.Filter != nil {+ filterExpr = req.Filter.Expression+ nodeFilter = &req.Filter.Filter+ filterByPodStatus = req.Filter.FilterByPodStatus+ filterByNodeReadiness = req.Filter.FilterByNodeReadiness+ }++ // getTopNodeGroupsAndMetadata fetches metadata + ranking (+ full-scope pod+ // status / node readiness when filtering) concurrently, intersecting+ // metadata/ranked groups against the keysets. It returns the keysets + warning.+ pageGroups, metadataMap, podStatusCounts, podStatusWarning, nodeConditionCounts, err := m.getTopNodeGroupsAndMetadata(ctx, orgID, req)if err != nil {return nil, err}+ // Required metric missing while filtering: surface the warning + empty result.+ if len(filterByPodStatus) != 0 && podStatusWarning != nil {+ resp.Warning = podStatusWarning+ resp.Records = []inframonitoringtypes.NodeRecord{}+ resp.Total = 0+ return resp, nil+ }+resp.Total = len(metadataMap)if len(pageGroups) == 0 {@@ -490,20 +550,8 @@ func (m *module) ListNodes(ctx context.Context, orgID valuer.UUID, req *inframonreturn resp, nil}- filterExpr := ""- if req.Filter != nil {- filterExpr = req.Filter.Expression- }-fullQueryReq := buildFullQueryRequest(req.Start, req.End, filterExpr, req.GroupBy, pageGroups, m.newNodesTableListQuery())- var (- queryResp *qbtypes.QueryRangeResponse- nodeConditionCounts map[string]nodeConditionCounts- podStatusCounts map[string]podStatusCounts- podStatusWarning *qbtypes.QueryWarnData- )-g, gCtx := errgroup.WithContext(ctx)g.Go(func() error {@@ -511,16 +559,24 @@ func (m *module) ListNodes(ctx context.Context, orgID valuer.UUID, req *inframonqueryResp, err = m.querier.QueryRange(gCtx, orgID, fullQueryReq)return err})- g.Go(func() error {- var err error- nodeConditionCounts, err = m.getPerGroupNodeConditionCounts(gCtx, orgID, req.Start, req.End, req.Filter, req.GroupBy, pageGroups)- return err- })- g.Go(func() error {- var err error- podStatusCounts, podStatusWarning, err = m.getPerGroupPodStatusCountsWithReqMetricChecks(gCtx, orgID, req.Start, req.End, req.Filter, req.GroupBy, pageGroups)- return err- })+ // When filtering by readiness, nodeConditionCounts already holds the full-scope+ // map (a superset of the page); otherwise compute it page-scoped here.+ if len(filterByNodeReadiness) == 0 {+ g.Go(func() error {+ var err error+ nodeConditionCounts, err = m.getPerGroupNodeConditionCounts(gCtx, orgID, req.Start, req.End, nodeFilter, req.GroupBy, pageGroups, nil)+ return err+ })+ }+ // When filtering by pod status, podStatusCounts already holds the full-scope+ // map; otherwise compute it page-scoped here.+ if len(filterByPodStatus) == 0 {+ g.Go(func() error {+ var err error+ podStatusCounts, podStatusWarning, err = m.getPerGroupPodStatusCountsWithReqMetricChecks(gCtx, orgID, req.Start, req.End, nodeFilter, req.GroupBy, pageGroups, nil)+ return err+ })+ }if err := g.Wait(); err != nil {return nil, err@@ -571,11 +627,36 @@ func (m *module) ListNamespaces(ctx context.Context, orgID valuer.UUID, req *infreturn resp, nil}- pageGroups, metadataMap, err := m.getTopNamespaceGroupsAndMetadata(ctx, orgID, req)+ var (+ filterExpr string+ namespaceFilter *qbtypes.Filter+ filterByPodStatus []inframonitoringtypes.PodStatus+ queryResp *qbtypes.QueryRangeResponse+ resourceCounts map[string]map[string]int64+ )++ if req.Filter != nil {+ filterExpr = req.Filter.Expression+ namespaceFilter = &req.Filter.Filter+ filterByPodStatus = req.Filter.FilterByPodStatus+ }++ // getTopNamespaceGroupsAndMetadata fetches metadata + ranking (+ full-scope pod+ // status when filtering) concurrently, intersecting metadata/ranked groups+ // against the status keyset. It returns the keyset + its warning.+ pageGroups, metadataMap, podStatusCounts, podStatusWarning, err := m.getTopNamespaceGroupsAndMetadata(ctx, orgID, req)if err != nil {return nil, err}+ // Required metric missing while filtering: surface the warning + empty result.+ if len(filterByPodStatus) != 0 && podStatusWarning != nil {+ resp.Warning = podStatusWarning+ resp.Records = []inframonitoringtypes.NamespaceRecord{}+ resp.Total = 0+ return resp, nil+ }+resp.Total = len(metadataMap)if len(pageGroups) == 0 {@@ -583,20 +664,8 @@ func (m *module) ListNamespaces(ctx context.Context, orgID valuer.UUID, req *infreturn resp, nil}- filterExpr := ""- if req.Filter != nil {- filterExpr = req.Filter.Expression- }-fullQueryReq := buildFullQueryRequest(req.Start, req.End, filterExpr, req.GroupBy, pageGroups, m.newNamespacesTableListQuery())- var (- queryResp *qbtypes.QueryRangeResponse- podStatusCounts map[string]podStatusCounts- podStatusWarning *qbtypes.QueryWarnData- resourceCounts map[string]map[string]int64- )-g, gCtx := errgroup.WithContext(ctx)g.Go(func() error {@@ -606,14 +675,18 @@ func (m *module) ListNamespaces(ctx context.Context, orgID valuer.UUID, req *inf})g.Go(func() error {var err error- podStatusCounts, podStatusWarning, err = m.getPerGroupPodStatusCountsWithReqMetricChecks(gCtx, orgID, req.Start, req.End, req.Filter, req.GroupBy, pageGroups)- return err- })- g.Go(func() error {- var err error- resourceCounts, err = m.getPerGroupDistinctCounts(gCtx, orgID, req.Start, req.End, req.Filter, req.GroupBy, pageGroups, namespaceCountAttrKeys, namespacesMetricNamesListForCounts)+ resourceCounts, err = m.getPerGroupDistinctCounts(gCtx, orgID, req.Start, req.End, namespaceFilter, req.GroupBy, pageGroups, namespaceCountAttrKeys, namespacesMetricNamesListForCounts)return err})+ // When filtering, podStatusCounts already holds the full-scope map (a superset+ // of the page); otherwise compute it page-scoped here.+ if len(filterByPodStatus) == 0 {+ g.Go(func() error {+ var err error+ podStatusCounts, podStatusWarning, err = m.getPerGroupPodStatusCountsWithReqMetricChecks(gCtx, orgID, req.Start, req.End, namespaceFilter, req.GroupBy, pageGroups, nil)+ return err+ })+ }if err := g.Wait(); err != nil {return nil, err@@ -663,11 +736,39 @@ func (m *module) ListClusters(ctx context.Context, orgID valuer.UUID, req *infrareturn resp, nil}- pageGroups, metadataMap, err := m.getTopClusterGroupsAndMetadata(ctx, orgID, req)+ var (+ filterExpr string+ clusterFilter *qbtypes.Filter+ filterByPodStatus []inframonitoringtypes.PodStatus+ filterByNodeReadiness []inframonitoringtypes.NodeCondition+ queryResp *qbtypes.QueryRangeResponse+ nodeConditionCountsMap map[string]nodeConditionCounts+ resourceCounts map[string]map[string]int64+ )++ if req.Filter != nil {+ filterExpr = req.Filter.Expression+ clusterFilter = &req.Filter.Filter+ filterByPodStatus = req.Filter.FilterByPodStatus+ filterByNodeReadiness = req.Filter.FilterByNodeReadiness+ }++ // getTopClusterGroupsAndMetadata fetches metadata + ranking (+ full-scope pod+ // status / node readiness when filtering) concurrently, intersecting+ // metadata/ranked groups against the keysets. It returns the keysets + warning.+ pageGroups, metadataMap, podStatusCounts, podStatusWarning, nodeConditionCountsMap, err := m.getTopClusterGroupsAndMetadata(ctx, orgID, req)if err != nil {return nil, err}+ // Required metric missing while filtering: surface the warning + empty result.+ if len(filterByPodStatus) != 0 && podStatusWarning != nil {+ resp.Warning = podStatusWarning+ resp.Records = []inframonitoringtypes.ClusterRecord{}+ resp.Total = 0+ return resp, nil+ }+resp.Total = len(metadataMap)if len(pageGroups) == 0 {@@ -675,23 +776,8 @@ func (m *module) ListClusters(ctx context.Context, orgID valuer.UUID, req *infrareturn resp, nil}- filterExpr := ""- if req.Filter != nil {- filterExpr = req.Filter.Expression- }-fullQueryReq := buildFullQueryRequest(req.Start, req.End, filterExpr, req.GroupBy, pageGroups, m.newClustersTableListQuery())- // With default groupBy [k8s.cluster.name], counts are bucketed per cluster;- // with a custom groupBy, they aggregate across clusters in that group.- var (- queryResp *qbtypes.QueryRangeResponse- nodeConditionCountsMap map[string]nodeConditionCounts- podStatusCounts map[string]podStatusCounts- podStatusWarning *qbtypes.QueryWarnData- resourceCounts map[string]map[string]int64- )-g, gCtx := errgroup.WithContext(ctx)g.Go(func() error {@@ -699,21 +785,29 @@ func (m *module) ListClusters(ctx context.Context, orgID valuer.UUID, req *infraqueryResp, err = m.querier.QueryRange(gCtx, orgID, fullQueryReq)return err})+ // When filtering by readiness, nodeConditionCountsMap already holds the+ // full-scope map (a superset of the page); otherwise compute it page-scoped here.+ if len(filterByNodeReadiness) == 0 {+ g.Go(func() error {+ var err error+ nodeConditionCountsMap, err = m.getPerGroupNodeConditionCounts(gCtx, orgID, req.Start, req.End, clusterFilter, req.GroupBy, pageGroups, nil)+ return err+ })+ }g.Go(func() error {var err error- nodeConditionCountsMap, err = m.getPerGroupNodeConditionCounts(gCtx, orgID, req.Start, req.End, req.Filter, req.GroupBy, pageGroups)- return err- })- g.Go(func() error {- var err error- podStatusCounts, podStatusWarning, err = m.getPerGroupPodStatusCountsWithReqMetricChecks(gCtx, orgID, req.Start, req.End, req.Filter, req.GroupBy, pageGroups)- return err- })- g.Go(func() error {- var err error- resourceCounts, err = m.getPerGroupDistinctCounts(gCtx, orgID, req.Start, req.End, req.Filter, req.GroupBy, pageGroups, clusterCountAttrKeys, clusterMetricNamesListForCounts)+ resourceCounts, err = m.getPerGroupDistinctCounts(gCtx, orgID, req.Start, req.End, clusterFilter, req.GroupBy, pageGroups, clusterCountAttrKeys, clusterMetricNamesListForCounts)return err})+ // When filtering by pod status, podStatusCounts already holds the full-scope+ // map; otherwise compute it page-scoped here.+ if len(filterByPodStatus) == 0 {+ g.Go(func() error {+ var err error+ podStatusCounts, podStatusWarning, err = m.getPerGroupPodStatusCountsWithReqMetricChecks(gCtx, orgID, req.Start, req.End, clusterFilter, req.GroupBy, pageGroups, nil)+ return err+ })+ }if err := g.Wait(); err != nil {return nil, err@@ -827,7 +921,7 @@ func (m *module) ListDeployments(ctx context.Context, orgID valuer.UUID, req *in// Bake the deployments base filter into req.Filter so all downstream helpers pick it up.if req.Filter == nil {- req.Filter = &qbtypes.Filter{}+ req.Filter = &inframonitoringtypes.DeploymentFilter{}}req.Filter.Expression = mergeFilterExpressions(deploymentsBaseFilterExpr, req.Filter.Expression)@@ -842,11 +936,35 @@ func (m *module) ListDeployments(ctx context.Context, orgID valuer.UUID, req *inreturn resp, nil}- pageGroups, metadataMap, err := m.getTopDeploymentGroupsAndMetadata(ctx, orgID, req)+ var (+ filterExpr string+ deploymentFilter *qbtypes.Filter+ filterByPodStatus []inframonitoringtypes.PodStatus+ queryResp *qbtypes.QueryRangeResponse+ )++ if req.Filter != nil {+ filterExpr = req.Filter.Expression+ deploymentFilter = &req.Filter.Filter+ filterByPodStatus = req.Filter.FilterByPodStatus+ }++ // getTopDeploymentGroupsAndMetadata fetches metadata + ranking (+ full-scope pod+ // status when filtering) concurrently, intersecting metadata/ranked groups+ // against the status keyset. It returns the keyset + its warning.+ pageGroups, metadataMap, podStatusCounts, podStatusWarning, err := m.getTopDeploymentGroupsAndMetadata(ctx, orgID, req)if err != nil {return nil, err}+ // Required metric missing while filtering: surface the warning + empty result.+ if len(filterByPodStatus) != 0 && podStatusWarning != nil {+ resp.Warning = podStatusWarning+ resp.Records = []inframonitoringtypes.DeploymentRecord{}+ resp.Total = 0+ return resp, nil+ }+resp.Total = len(metadataMap)if len(pageGroups) == 0 {@@ -854,19 +972,8 @@ func (m *module) ListDeployments(ctx context.Context, orgID valuer.UUID, req *inreturn resp, nil}- filterExpr := ""- if req.Filter != nil {- filterExpr = req.Filter.Expression- }-fullQueryReq := buildFullQueryRequest(req.Start, req.End, filterExpr, req.GroupBy, pageGroups, m.newDeploymentsTableListQuery())- var (- queryResp *qbtypes.QueryRangeResponse- podStatusCounts map[string]podStatusCounts- podStatusWarning *qbtypes.QueryWarnData- )-g, gCtx := errgroup.WithContext(ctx)g.Go(func() error {@@ -874,11 +981,15 @@ func (m *module) ListDeployments(ctx context.Context, orgID valuer.UUID, req *inqueryResp, err = m.querier.QueryRange(gCtx, orgID, fullQueryReq)return err})- g.Go(func() error {- var err error- podStatusCounts, podStatusWarning, err = m.getPerGroupPodStatusCountsWithReqMetricChecks(gCtx, orgID, req.Start, req.End, req.Filter, req.GroupBy, pageGroups)- return err- })+ // When filtering, podStatusCounts already holds the full-scope map (a superset+ // of the page); otherwise compute it page-scoped here.+ if len(filterByPodStatus) == 0 {+ g.Go(func() error {+ var err error+ podStatusCounts, podStatusWarning, err = m.getPerGroupPodStatusCountsWithReqMetricChecks(gCtx, orgID, req.Start, req.End, deploymentFilter, req.GroupBy, pageGroups, nil)+ return err+ })+ }if err := g.Wait(); err != nil {return nil, err@@ -919,7 +1030,7 @@ func (m *module) ListStatefulSets(ctx context.Context, orgID valuer.UUID, req *i// Bake the workload base filter into req.Filter so all downstream helpers pick it up.if req.Filter == nil {- req.Filter = &qbtypes.Filter{}+ req.Filter = &inframonitoringtypes.StatefulSetFilter{}}req.Filter.Expression = mergeFilterExpressions(statefulSetsBaseFilterExpr, req.Filter.Expression)@@ -934,11 +1045,35 @@ func (m *module) ListStatefulSets(ctx context.Context, orgID valuer.UUID, req *ireturn resp, nil}- pageGroups, metadataMap, err := m.getTopStatefulSetGroupsAndMetadata(ctx, orgID, req)+ var (+ filterExpr string+ statefulSetFilter *qbtypes.Filter+ filterByPodStatus []inframonitoringtypes.PodStatus+ queryResp *qbtypes.QueryRangeResponse+ )++ if req.Filter != nil {+ filterExpr = req.Filter.Expression+ statefulSetFilter = &req.Filter.Filter+ filterByPodStatus = req.Filter.FilterByPodStatus+ }++ // getTopStatefulSetGroupsAndMetadata fetches metadata + ranking (+ full-scope pod+ // status when filtering) concurrently, intersecting metadata/ranked groups+ // against the status keyset. It returns the keyset + its warning.+ pageGroups, metadataMap, podStatusCounts, podStatusWarning, err := m.getTopStatefulSetGroupsAndMetadata(ctx, orgID, req)if err != nil {return nil, err}+ // Required metric missing while filtering: surface the warning + empty result.+ if len(filterByPodStatus) != 0 && podStatusWarning != nil {+ resp.Warning = podStatusWarning+ resp.Records = []inframonitoringtypes.StatefulSetRecord{}+ resp.Total = 0+ return resp, nil+ }+resp.Total = len(metadataMap)if len(pageGroups) == 0 {@@ -946,21 +1081,8 @@ func (m *module) ListStatefulSets(ctx context.Context, orgID valuer.UUID, req *ireturn resp, nil}- filterExpr := ""- if req.Filter != nil {- filterExpr = req.Filter.Expression- }-fullQueryReq := buildFullQueryRequest(req.Start, req.End, filterExpr, req.GroupBy, pageGroups, m.newStatefulSetsTableListQuery())- // Pods owned by a StatefulSet carry k8s.statefulset.name as a resource attribute,- // so default-groupBy gives per-statefulset status counts automatically.- var (- queryResp *qbtypes.QueryRangeResponse- podStatusCounts map[string]podStatusCounts- podStatusWarning *qbtypes.QueryWarnData- )-g, gCtx := errgroup.WithContext(ctx)g.Go(func() error {@@ -968,11 +1090,15 @@ func (m *module) ListStatefulSets(ctx context.Context, orgID valuer.UUID, req *iqueryResp, err = m.querier.QueryRange(gCtx, orgID, fullQueryReq)return err})- g.Go(func() error {- var err error- podStatusCounts, podStatusWarning, err = m.getPerGroupPodStatusCountsWithReqMetricChecks(gCtx, orgID, req.Start, req.End, req.Filter, req.GroupBy, pageGroups)- return err- })+ // When filtering, podStatusCounts already holds the full-scope map (a superset+ // of the page); otherwise compute it page-scoped here.+ if len(filterByPodStatus) == 0 {+ g.Go(func() error {+ var err error+ podStatusCounts, podStatusWarning, err = m.getPerGroupPodStatusCountsWithReqMetricChecks(gCtx, orgID, req.Start, req.End, statefulSetFilter, req.GroupBy, pageGroups, nil)+ return err+ })+ }if err := g.Wait(); err != nil {return nil, err@@ -1013,7 +1139,7 @@ func (m *module) ListJobs(ctx context.Context, orgID valuer.UUID, req *inframoni// Bake the jobs base filter into req.Filter so all downstream helpers pick it up.if req.Filter == nil {- req.Filter = &qbtypes.Filter{}+ req.Filter = &inframonitoringtypes.JobFilter{}}req.Filter.Expression = mergeFilterExpressions(jobsBaseFilterExpr, req.Filter.Expression)@@ -1028,11 +1154,35 @@ func (m *module) ListJobs(ctx context.Context, orgID valuer.UUID, req *inframonireturn resp, nil}- pageGroups, metadataMap, err := m.getTopJobGroupsAndMetadata(ctx, orgID, req)+ var (+ filterExpr string+ jobFilter *qbtypes.Filter+ filterByPodStatus []inframonitoringtypes.PodStatus+ queryResp *qbtypes.QueryRangeResponse+ )++ if req.Filter != nil {+ filterExpr = req.Filter.Expression+ jobFilter = &req.Filter.Filter+ filterByPodStatus = req.Filter.FilterByPodStatus+ }++ // getTopJobGroupsAndMetadata fetches metadata + ranking (+ full-scope pod+ // status when filtering) concurrently, intersecting metadata/ranked groups+ // against the status keyset. It returns the keyset + its warning.+ pageGroups, metadataMap, podStatusCounts, podStatusWarning, err := m.getTopJobGroupsAndMetadata(ctx, orgID, req)if err != nil {return nil, err}+ // Required metric missing while filtering: surface the warning + empty result.+ if len(filterByPodStatus) != 0 && podStatusWarning != nil {+ resp.Warning = podStatusWarning+ resp.Records = []inframonitoringtypes.JobRecord{}+ resp.Total = 0+ return resp, nil+ }+resp.Total = len(metadataMap)if len(pageGroups) == 0 {@@ -1040,21 +1190,8 @@ func (m *module) ListJobs(ctx context.Context, orgID valuer.UUID, req *inframonireturn resp, nil}- filterExpr := ""- if req.Filter != nil {- filterExpr = req.Filter.Expression- }-fullQueryReq := buildFullQueryRequest(req.Start, req.End, filterExpr, req.GroupBy, pageGroups, m.newJobsTableListQuery())- // Pods owned by a Job carry k8s.job.name as a resource attribute, so default-groupBy- // gives per-job status counts automatically.- var (- queryResp *qbtypes.QueryRangeResponse- podStatusCounts map[string]podStatusCounts- podStatusWarning *qbtypes.QueryWarnData- )-g, gCtx := errgroup.WithContext(ctx)g.Go(func() error {@@ -1062,11 +1199,15 @@ func (m *module) ListJobs(ctx context.Context, orgID valuer.UUID, req *inframoniqueryResp, err = m.querier.QueryRange(gCtx, orgID, fullQueryReq)return err})- g.Go(func() error {- var err error- podStatusCounts, podStatusWarning, err = m.getPerGroupPodStatusCountsWithReqMetricChecks(gCtx, orgID, req.Start, req.End, req.Filter, req.GroupBy, pageGroups)- return err- })+ // When filtering, podStatusCounts already holds the full-scope map (a superset+ // of the page); otherwise compute it page-scoped here.+ if len(filterByPodStatus) == 0 {+ g.Go(func() error {+ var err error+ podStatusCounts, podStatusWarning, err = m.getPerGroupPodStatusCountsWithReqMetricChecks(gCtx, orgID, req.Start, req.End, jobFilter, req.GroupBy, pageGroups, nil)+ return err+ })+ }if err := g.Wait(); err != nil {return nil, err@@ -1107,7 +1248,7 @@ func (m *module) ListDaemonSets(ctx context.Context, orgID valuer.UUID, req *inf// Bake the workload base filter into req.Filter so all downstream helpers pick it up.if req.Filter == nil {- req.Filter = &qbtypes.Filter{}+ req.Filter = &inframonitoringtypes.DaemonSetFilter{}}req.Filter.Expression = mergeFilterExpressions(daemonSetsBaseFilterExpr, req.Filter.Expression)@@ -1122,11 +1263,35 @@ func (m *module) ListDaemonSets(ctx context.Context, orgID valuer.UUID, req *infreturn resp, nil}- pageGroups, metadataMap, err := m.getTopDaemonSetGroupsAndMetadata(ctx, orgID, req)+ var (+ filterExpr string+ daemonSetFilter *qbtypes.Filter+ filterByPodStatus []inframonitoringtypes.PodStatus+ queryResp *qbtypes.QueryRangeResponse+ )++ if req.Filter != nil {+ filterExpr = req.Filter.Expression+ daemonSetFilter = &req.Filter.Filter+ filterByPodStatus = req.Filter.FilterByPodStatus+ }++ // getTopDaemonSetGroupsAndMetadata fetches metadata + ranking (+ full-scope pod+ // status when filtering) concurrently, intersecting metadata/ranked groups+ // against the status keyset. It returns the keyset + its warning.+ pageGroups, metadataMap, podStatusCounts, podStatusWarning, err := m.getTopDaemonSetGroupsAndMetadata(ctx, orgID, req)if err != nil {return nil, err}+ // Required metric missing while filtering: surface the warning + empty result.+ if len(filterByPodStatus) != 0 && podStatusWarning != nil {+ resp.Warning = podStatusWarning+ resp.Records = []inframonitoringtypes.DaemonSetRecord{}+ resp.Total = 0+ return resp, nil+ }+resp.Total = len(metadataMap)if len(pageGroups) == 0 {@@ -1134,21 +1299,8 @@ func (m *module) ListDaemonSets(ctx context.Context, orgID valuer.UUID, req *infreturn resp, nil}- filterExpr := ""- if req.Filter != nil {- filterExpr = req.Filter.Expression- }-fullQueryReq := buildFullQueryRequest(req.Start, req.End, filterExpr, req.GroupBy, pageGroups, m.newDaemonSetsTableListQuery())- // Pods owned by a DaemonSet carry k8s.daemonset.name as a resource attribute,- // so default-groupBy gives per-daemonset status counts automatically.- var (- queryResp *qbtypes.QueryRangeResponse- podStatusCounts map[string]podStatusCounts- podStatusWarning *qbtypes.QueryWarnData- )-g, gCtx := errgroup.WithContext(ctx)g.Go(func() error {@@ -1156,11 +1308,15 @@ func (m *module) ListDaemonSets(ctx context.Context, orgID valuer.UUID, req *infqueryResp, err = m.querier.QueryRange(gCtx, orgID, fullQueryReq)return err})- g.Go(func() error {- var err error- podStatusCounts, podStatusWarning, err = m.getPerGroupPodStatusCountsWithReqMetricChecks(gCtx, orgID, req.Start, req.End, req.Filter, req.GroupBy, pageGroups)- return err- })+ // When filtering, podStatusCounts already holds the full-scope map (a superset+ // of the page); otherwise compute it page-scoped here.+ if len(filterByPodStatus) == 0 {+ g.Go(func() error {+ var err error+ podStatusCounts, podStatusWarning, err = m.getPerGroupPodStatusCountsWithReqMetricChecks(gCtx, orgID, req.Start, req.End, daemonSetFilter, req.GroupBy, pageGroups, nil)+ return err+ })+ }if err := g.Wait(); err != nil {return nil, errdiff --git a/pkg/modules/inframonitoring/implinframonitoring/namespaces.go b/pkg/modules/inframonitoring/implinframonitoring/namespaces.goindex d6d442d88b..ab9aba5cfa 100644--- a/pkg/modules/inframonitoring/implinframonitoring/namespaces.go+++ b/pkg/modules/inframonitoring/implinframonitoring/namespaces.go@@ -65,20 +65,34 @@ func buildNamespaceRecords(return records}+// getTopNamespaceGroupsAndMetadata concurrently fetches metadata + the ordering-metric+// ranking (plus the full-scope pod-status keyset when filtering, to intersect both).func (m *module) getTopNamespaceGroupsAndMetadata(ctx context.Context,orgID valuer.UUID,req *inframonitoringtypes.PostableNamespaces,-) ([]map[string]string, map[string]map[string]string, error) {+) ([]map[string]string, map[string]map[string]string, map[string]podStatusCounts, *qbtypes.QueryWarnData, error) {var (- orderByKey string- metadataMap map[string]map[string]string- allMetricGroups []rankedGroup+ orderByKey string+ metadataMap map[string]map[string]string+ allMetricGroups []rankedGroup+ statusCounts map[string]podStatusCounts+ statusWarning *qbtypes.QueryWarnData+ filter *qbtypes.Filter+ filterByPodStatus []inframonitoringtypes.PodStatus)orderByKey = req.OrderBy.Key.Name+ // When filtering by pod status, resolve the full-scope status keyset+ // concurrently (pageGroups=nil spans all groups under the user filter) so it+ // can intersect metadata + ranked groups below.+ if req.Filter != nil {+ filter = &req.Filter.Filter+ filterByPodStatus = req.Filter.FilterByPodStatus+ }+g, gCtx := errgroup.WithContext(ctx)g.Go(func() error {@@ -87,12 +101,26 @@ func (m *module) getTopNamespaceGroupsAndMetadata(return err})+ if len(filterByPodStatus) != 0 {+ g.Go(func() error {+ var err error+ statusCounts, statusWarning, err = m.getPerGroupPodStatusCountsWithReqMetricChecks(gCtx, orgID, req.Start, req.End, filter, req.GroupBy, nil, filterByPodStatus)+ return err+ })+ }+if orderByKey == inframonitoringtypes.NamespaceNameAttrKey {if err := g.Wait(); err != nil {- return nil, nil, err+ return nil, nil, nil, nil, err+ }+ // Secondary filter: keep only status-matching groups. A missing metric+ // yields an empty statusCounts, so this correctly empties the result+ // (the caller also surfaces the warning).+ if len(filterByPodStatus) != 0 {+ metadataMap = intersectMap(metadataMap, statusCounts)}pageGroups := inframonitoringtypes.PaginateMetadataByName(metadataMap, req.GroupBy, req.OrderBy.Direction, req.Offset, req.Limit, inframonitoringtypes.NamespaceNameAttrKey)- return pageGroups, metadataMap, nil+ return pageGroups, metadataMap, statusCounts, statusWarning, nil}queryNamesForOrderBy := orderByToNamespacesQueryNames[orderByKey]@@ -138,10 +166,19 @@ func (m *module) getTopNamespaceGroupsAndMetadata(})if err := g.Wait(); err != nil {- return nil, nil, err+ return nil, nil, nil, nil, err+ }++ // Secondary filter: intersect ranked groups + metadata with the status keyset.+ // A missing metric yields an empty statusCounts, correctly emptying the result+ // (the caller also surfaces the warning).+ if len(filterByPodStatus) != 0 {+ allMetricGroups = intersectRankedGroups(allMetricGroups, statusCounts)+ metadataMap = intersectMap(metadataMap, statusCounts)}- return paginateWithBackfill(allMetricGroups, metadataMap, req.GroupBy, req.Offset, req.Limit), metadataMap, nil+ pageGroups := paginateWithBackfill(allMetricGroups, metadataMap, req.GroupBy, req.Offset, req.Limit)+ return pageGroups, metadataMap, statusCounts, statusWarning, nil}func (m *module) getNamespacesTableMetadata(ctx context.Context, orgID valuer.UUID, req *inframonitoringtypes.PostableNamespaces) (map[string]map[string]string, error) {@@ -151,5 +188,9 @@ func (m *module) getNamespacesTableMetadata(ctx context.Context, orgID valuer.UUnonGroupByAttrs = append(nonGroupByAttrs, key)}}- return m.getMetadata(ctx, orgID, namespacesTableMetricNamesList, req.GroupBy, nonGroupByAttrs, req.Filter, req.Start, req.End)+ var filter *qbtypes.Filter+ if req.Filter != nil {+ filter = &req.Filter.Filter+ }+ return m.getMetadata(ctx, orgID, namespacesTableMetricNamesList, req.GroupBy, nonGroupByAttrs, filter, req.Start, req.End)}diff --git a/pkg/modules/inframonitoring/implinframonitoring/nodes.go b/pkg/modules/inframonitoring/implinframonitoring/nodes.goindex db0363495b..d06a86e74f 100644--- a/pkg/modules/inframonitoring/implinframonitoring/nodes.go+++ b/pkg/modules/inframonitoring/implinframonitoring/nodes.go@@ -4,7 +4,6 @@ import ("context""fmt""slices"- "strings""github.com/SigNoz/signoz/pkg/querybuilder""github.com/SigNoz/signoz/pkg/telemetryschema/metricstelemetryschema"@@ -92,20 +91,38 @@ func buildNodeRecords(return records}+// getTopNodeGroupsAndMetadata concurrently fetches metadata + the ordering-metric+// ranking (plus the full-scope pod-status / node-readiness keysets when filtering,+// to intersect all).func (m *module) getTopNodeGroupsAndMetadata(ctx context.Context,orgID valuer.UUID,req *inframonitoringtypes.PostableNodes,-) ([]map[string]string, map[string]map[string]string, error) {+) ([]map[string]string, map[string]map[string]string, map[string]podStatusCounts, *qbtypes.QueryWarnData, map[string]nodeConditionCounts, error) {var (- orderByKey string- metadataMap map[string]map[string]string- allMetricGroups []rankedGroup+ orderByKey string+ metadataMap map[string]map[string]string+ allMetricGroups []rankedGroup+ statusCounts map[string]podStatusCounts+ statusWarning *qbtypes.QueryWarnData+ nodeConditionCounts map[string]nodeConditionCounts+ filter *qbtypes.Filter+ filterByPodStatus []inframonitoringtypes.PodStatus+ filterByNodeReadiness []inframonitoringtypes.NodeCondition)orderByKey = req.OrderBy.Key.Name+ // When filtering by pod status / node readiness, resolve the full-scope+ // keyset(s) concurrently (pageGroups=nil spans all groups under the user+ // filter) to intersect metadata + ranked groups below. Filters compose as AND.+ if req.Filter != nil {+ filter = &req.Filter.Filter+ filterByPodStatus = req.Filter.FilterByPodStatus+ filterByNodeReadiness = req.Filter.FilterByNodeReadiness+ }+g, gCtx := errgroup.WithContext(ctx)g.Go(func() error {@@ -114,12 +131,37 @@ func (m *module) getTopNodeGroupsAndMetadata(return err})+ if len(filterByPodStatus) != 0 {+ g.Go(func() error {+ var err error+ statusCounts, statusWarning, err = m.getPerGroupPodStatusCountsWithReqMetricChecks(gCtx, orgID, req.Start, req.End, filter, req.GroupBy, nil, filterByPodStatus)+ return err+ })+ }++ if len(filterByNodeReadiness) != 0 {+ g.Go(func() error {+ var err error+ nodeConditionCounts, err = m.getPerGroupNodeConditionCounts(gCtx, orgID, req.Start, req.End, filter, req.GroupBy, nil, filterByNodeReadiness)+ return err+ })+ }+if orderByKey == inframonitoringtypes.NodeNameAttrKey {if err := g.Wait(); err != nil {- return nil, nil, err+ return nil, nil, nil, nil, nil, err+ }+ // Secondary filter: keep only status/readiness-matching groups. A missing+ // metric yields an empty statusCounts, so this correctly empties the result+ // (the caller also surfaces the warning). Filters compose as AND.+ if len(filterByPodStatus) != 0 {+ metadataMap = intersectMap(metadataMap, statusCounts)+ }+ if len(filterByNodeReadiness) != 0 {+ metadataMap = intersectMap(metadataMap, nodeConditionCounts)}pageGroups := inframonitoringtypes.PaginateMetadataByName(metadataMap, req.GroupBy, req.OrderBy.Direction, req.Offset, req.Limit, inframonitoringtypes.NodeNameAttrKey)- return pageGroups, metadataMap, nil+ return pageGroups, metadataMap, statusCounts, statusWarning, nodeConditionCounts, nil}queryNamesForOrderBy := orderByToNodesQueryNames[orderByKey]@@ -165,10 +207,23 @@ func (m *module) getTopNodeGroupsAndMetadata(})if err := g.Wait(); err != nil {- return nil, nil, err+ return nil, nil, nil, nil, nil, err}- return paginateWithBackfill(allMetricGroups, metadataMap, req.GroupBy, req.Offset, req.Limit), metadataMap, nil+ // Secondary filter: intersect ranked groups + metadata with the status/readiness+ // keyset. A missing metric yields an empty keyset, correctly emptying the result+ // (the caller also surfaces the warning). Filters compose as AND.+ if len(filterByPodStatus) != 0 {+ allMetricGroups = intersectRankedGroups(allMetricGroups, statusCounts)+ metadataMap = intersectMap(metadataMap, statusCounts)+ }+ if len(filterByNodeReadiness) != 0 {+ allMetricGroups = intersectRankedGroups(allMetricGroups, nodeConditionCounts)+ metadataMap = intersectMap(metadataMap, nodeConditionCounts)+ }++ pageGroups := paginateWithBackfill(allMetricGroups, metadataMap, req.GroupBy, req.Offset, req.Limit)+ return pageGroups, metadataMap, statusCounts, statusWarning, nodeConditionCounts, nil}func (m *module) getNodesTableMetadata(ctx context.Context, orgID valuer.UUID, req *inframonitoringtypes.PostableNodes) (map[string]map[string]string, error) {@@ -178,7 +233,11 @@ func (m *module) getNodesTableMetadata(ctx context.Context, orgID valuer.UUID, rnonGroupByAttrs = append(nonGroupByAttrs, key)}}- return m.getMetadata(ctx, orgID, nodesTableMetricNamesList, req.GroupBy, nonGroupByAttrs, req.Filter, req.Start, req.End)+ var filter *qbtypes.Filter+ if req.Filter != nil {+ filter = &req.Filter.Filter+ }+ return m.getMetadata(ctx, orgID, nodesTableMetricNamesList, req.GroupBy, nonGroupByAttrs, filter, req.Start, req.End)}// getPerGroupNodeConditionCounts computes per-group node counts bucketed by each@@ -192,6 +251,24 @@ func (m *module) getNodesTableMetadata(ctx context.Context, orgID valuer.UUID, r// countNodesPerCondition: per-group uniqExactIf into ready/not_ready buckets.//// Groups absent from the result map have implicit zero counts (caller default).+// applyNodeReadinessFilter adds the readiness push-down (condition_value IN (...))+// to the outer count builder. condition_value is numeric (1=Ready, 0=NotReady), so+// we map each requested enum to its int. No-op when the requested set is empty.+func applyNodeReadinessFilter(cb *sqlbuilder.SelectBuilder, filterByNodeReadiness []inframonitoringtypes.NodeCondition) {+ if len(filterByNodeReadiness) == 0 {+ return+ }+ nums := make([]int, len(filterByNodeReadiness))+ for i, c := range filterByNodeReadiness {+ v := inframonitoringtypes.NodeConditionNumNotReady+ if c == inframonitoringtypes.NodeConditionReady {+ v = inframonitoringtypes.NodeConditionNumReady+ }+ nums[i] = v+ }+ cb.Where(cb.In("condition_value", sqlbuilder.List(nums)))+}+func (m *module) getPerGroupNodeConditionCounts(ctx context.Context,orgID valuer.UUID,@@ -199,8 +276,11 @@ func (m *module) getPerGroupNodeConditionCounts(filter *qbtypes.Filter,groupBy []qbtypes.GroupByKey,pageGroups []map[string]string,+ filterByNodeReadiness []inframonitoringtypes.NodeCondition,) (map[string]nodeConditionCounts, error) {- if len(pageGroups) == 0 || len(groupBy) == 0 {+ // Empty pageGroups means "span all under user filter", allowed only in+ // full-scope mode (filtering by readiness). Otherwise it's an empty page.+ if len(groupBy) == 0 || (len(pageGroups) == 0 && len(filterByNodeReadiness) == 0) {return map[string]nodeConditionCounts{}, nil}@@ -288,11 +368,14 @@ func (m *module) getPerGroupNodeConditionCounts(fmt.Sprintf("uniqExactIf(node_name, condition_value = %d) AS ready_count", inframonitoringtypes.NodeConditionNumReady),fmt.Sprintf("uniqExactIf(node_name, condition_value = %d) AS not_ready_count", inframonitoringtypes.NodeConditionNumNotReady),)- countNodesPerConditionSQL := fmt.Sprintf(- "SELECT %s FROM latest_condition_per_node GROUP BY %s",- strings.Join(countNodesPerConditionSelectCols, ", "),- strings.Join(countNodesPerConditionGroupBy, ", "),- )+ // Outer count query. Built with sqlbuilder so the readiness push-down uses a+ // proper IN (keep only nodes whose readiness is in the requested set).+ countBuilder := sqlbuilder.NewSelectBuilder()+ countBuilder.Select(countNodesPerConditionSelectCols...)+ countBuilder.From("latest_condition_per_node")+ applyNodeReadinessFilter(countBuilder, filterByNodeReadiness)+ countBuilder.GroupBy(countNodesPerConditionGroupBy...)+ countNodesPerConditionSQL, countArgs := countBuilder.BuildWithFlavor(sqlbuilder.ClickHouse)// Combine CTEs + outer.cteFragments := []string{@@ -300,7 +383,7 @@ func (m *module) getPerGroupNodeConditionCounts(fmt.Sprintf("latest_condition_per_node AS (%s)", latestConditionPerNodeSQL),}finalSQL := querybuilder.CombineCTEs(cteFragments) + countNodesPerConditionSQL- finalArgs := querybuilder.PrependArgs([][]any{timeSeriesFPsArgs, latestConditionPerNodeArgs}, nil)+ finalArgs := querybuilder.PrependArgs([][]any{timeSeriesFPsArgs, latestConditionPerNodeArgs}, countArgs)rows, err := m.telemetryStore.ClickhouseDB().Query(ctx, finalSQL, finalArgs...)if err != nil {diff --git a/pkg/modules/inframonitoring/implinframonitoring/nodes_test.go b/pkg/modules/inframonitoring/implinframonitoring/nodes_test.gonew file mode 100644index 0000000000..295c311252--- /dev/null+++ b/pkg/modules/inframonitoring/implinframonitoring/nodes_test.go@@ -0,0 +1,65 @@+package implinframonitoring++import (+ "strings"+ "testing"++ "github.com/SigNoz/signoz/pkg/types/inframonitoringtypes"+ "github.com/huandu/go-sqlbuilder"+ "github.com/stretchr/testify/assert"+)++func TestApplyNodeReadinessFilter(t *testing.T) {+ tests := []struct {+ name string+ readiness []inframonitoringtypes.NodeCondition+ wantWhere bool+ wantArgs []any+ }{+ {+ name: "empty set yields no clause",+ readiness: nil,+ wantWhere: false,+ wantArgs: nil,+ },+ {+ name: "ready maps to 1 via IN",+ readiness: []inframonitoringtypes.NodeCondition{inframonitoringtypes.NodeConditionReady},+ wantWhere: true,+ wantArgs: []any{inframonitoringtypes.NodeConditionNumReady},+ },+ {+ name: "not_ready maps to 0 via IN",+ readiness: []inframonitoringtypes.NodeCondition{inframonitoringtypes.NodeConditionNotReady},+ wantWhere: true,+ wantArgs: []any{inframonitoringtypes.NodeConditionNumNotReady},+ },+ {+ name: "multiple conditions map to their ints via IN",+ readiness: []inframonitoringtypes.NodeCondition{+ inframonitoringtypes.NodeConditionReady,+ inframonitoringtypes.NodeConditionNotReady,+ },+ wantWhere: true,+ wantArgs: []any{inframonitoringtypes.NodeConditionNumReady, inframonitoringtypes.NodeConditionNumNotReady},+ },+ }++ for _, tt := range tests {+ t.Run(tt.name, func(t *testing.T) {+ cb := sqlbuilder.NewSelectBuilder()+ cb.Select("node_name")+ cb.From("latest_condition_per_node")+ applyNodeReadinessFilter(cb, tt.readiness)+ sql, args := cb.BuildWithFlavor(sqlbuilder.ClickHouse)++ hasWhere := strings.Contains(sql, "condition_value IN (")+ assert.Equal(t, tt.wantWhere, hasWhere)+ if len(tt.wantArgs) == 0 {+ assert.Empty(t, args)+ } else {+ assert.Equal(t, tt.wantArgs, args)+ }+ })+ }+}diff --git a/pkg/modules/inframonitoring/implinframonitoring/pods.go b/pkg/modules/inframonitoring/implinframonitoring/pods.goindex e905fe46d5..18037bc84c 100644--- a/pkg/modules/inframonitoring/implinframonitoring/pods.go+++ b/pkg/modules/inframonitoring/implinframonitoring/pods.go@@ -146,24 +146,34 @@ func buildPodRecords(return records}-// getTopPodGroupsAndMetadata fetches the group metadata and the ordering-metric-// ranking concurrently, then pages the ranked groups, backfilling from metadata-// when the page extends past the metric-ranked groups. Returns the page of-// groups and the metadata map (needed by the caller for Total and records).+// getTopPodGroupsAndMetadata concurrently fetches metadata + the ordering-metric+// ranking (plus the full-scope pod-status keyset when filtering, to intersect both).func (m *module) getTopPodGroupsAndMetadata(ctx context.Context,orgID valuer.UUID,req *inframonitoringtypes.PostablePods,-) ([]map[string]string, map[string]map[string]string, error) {+) ([]map[string]string, map[string]map[string]string, map[string]podStatusCounts, *qbtypes.QueryWarnData, error) {var (- orderByKey string- metadataMap map[string]map[string]string- allMetricGroups []rankedGroup+ orderByKey string+ metadataMap map[string]map[string]string+ allMetricGroups []rankedGroup+ statusCounts map[string]podStatusCounts+ statusWarning *qbtypes.QueryWarnData+ filter *qbtypes.Filter+ filterByPodStatus []inframonitoringtypes.PodStatus)orderByKey = req.OrderBy.Key.Name+ // When filtering by pod status, resolve the full-scope status keyset+ // concurrently (pageGroups=nil spans all groups under the user filter) so it+ // can intersect metadata + ranked groups below.+ if req.Filter != nil {+ filter = &req.Filter.Filter+ filterByPodStatus = req.Filter.FilterByPodStatus+ }+g, gCtx := errgroup.WithContext(ctx)g.Go(func() error {@@ -172,12 +182,26 @@ func (m *module) getTopPodGroupsAndMetadata(return err})+ if len(filterByPodStatus) != 0 {+ g.Go(func() error {+ var err error+ statusCounts, statusWarning, err = m.getPerGroupPodStatusCountsWithReqMetricChecks(gCtx, orgID, req.Start, req.End, filter, req.GroupBy, nil, filterByPodStatus)+ return err+ })+ }+if orderByKey == inframonitoringtypes.PodNameAttrKey {if err := g.Wait(); err != nil {- return nil, nil, err+ return nil, nil, nil, nil, err+ }+ // Secondary filter: keep only status-matching groups. A missing metric+ // yields an empty statusCounts, so this correctly empties the result+ // (the caller also surfaces the warning).+ if len(filterByPodStatus) != 0 {+ metadataMap = intersectMap(metadataMap, statusCounts)}pageGroups := inframonitoringtypes.PaginateMetadataByName(metadataMap, req.GroupBy, req.OrderBy.Direction, req.Offset, req.Limit, inframonitoringtypes.PodNameAttrKey)- return pageGroups, metadataMap, nil+ return pageGroups, metadataMap, statusCounts, statusWarning, nil}queryNamesForOrderBy := orderByToPodsQueryNames[orderByKey]@@ -223,10 +247,19 @@ func (m *module) getTopPodGroupsAndMetadata(})if err := g.Wait(); err != nil {- return nil, nil, err+ return nil, nil, nil, nil, err}- return paginateWithBackfill(allMetricGroups, metadataMap, req.GroupBy, req.Offset, req.Limit), metadataMap, nil+ // Secondary filter: intersect ranked groups + metadata with the status keyset.+ // A missing metric yields an empty statusCounts, correctly emptying the result+ // (the caller also surfaces the warning).+ if len(filterByPodStatus) != 0 {+ allMetricGroups = intersectRankedGroups(allMetricGroups, statusCounts)+ metadataMap = intersectMap(metadataMap, statusCounts)+ }++ pageGroups := paginateWithBackfill(allMetricGroups, metadataMap, req.GroupBy, req.Offset, req.Limit)+ return pageGroups, metadataMap, statusCounts, statusWarning, nil}func (m *module) getPodsTableMetadata(ctx context.Context, orgID valuer.UUID, req *inframonitoringtypes.PostablePods) (map[string]map[string]string, error) {@@ -236,7 +269,11 @@ func (m *module) getPodsTableMetadata(ctx context.Context, orgID valuer.UUID, renonGroupByAttrs = append(nonGroupByAttrs, key)}}- return m.getMetadata(ctx, orgID, podsTableMetricNamesList, req.GroupBy, nonGroupByAttrs, req.Filter, req.Start, req.End)+ var filter *qbtypes.Filter+ if req.Filter != nil {+ filter = &req.Filter.Filter+ }+ return m.getMetadata(ctx, orgID, podsTableMetricNamesList, req.GroupBy, nonGroupByAttrs, filter, req.Start, req.End)}// getPerGroupPodStatusCountsWithReqMetricChecks gates getPerGroupPodStatusCounts@@ -251,6 +288,7 @@ func (m *module) getPerGroupPodStatusCountsWithReqMetricChecks(filter *qbtypes.Filter,groupBy []qbtypes.GroupByKey,pageGroups []map[string]string,+ filterByPodStatus []inframonitoringtypes.PodStatus,) (map[string]podStatusCounts, *qbtypes.QueryWarnData, error) {present, err := m.getMetricsExistence(ctx, podStatusMetricNamesList)if err != nil {@@ -276,13 +314,28 @@ func (m *module) getPerGroupPodStatusCountsWithReqMetricChecks(return map[string]podStatusCounts{}, warning, nil}- counts, err := m.getPerGroupPodStatusCounts(ctx, orgID, start, end, filter, groupBy, pageGroups)+ counts, err := m.getPerGroupPodStatusCounts(ctx, orgID, start, end, filter, groupBy, pageGroups, filterByPodStatus)if err != nil {return nil, nil, err}return counts, nil, nil}+// applyPodStatusFilter adds the display-status push-down (lower(display_status)+// IN (...)) to the outer count builder. valuer lowercases the wire value while+// display_status is kubectl-cased, so we compare lower() on both. No-op when the+// requested set is empty.+func applyPodStatusFilter(cb *sqlbuilder.SelectBuilder, filterByPodStatus []inframonitoringtypes.PodStatus) {+ if len(filterByPodStatus) == 0 {+ return+ }+ vals := make([]string, len(filterByPodStatus))+ for i, s := range filterByPodStatus {+ vals[i] = s.StringValue()+ }+ cb.Where(cb.In("lower(display_status)", sqlbuilder.List(vals)))+}+// getPerGroupPodStatusCounts computes per-group pod counts bucketed by each// pod's latest kubectl-style display status in the requested window. Caller// must ensure the required metrics exist (getPerGroupPodStatusCountsWithReqMetricChecks).@@ -303,13 +356,20 @@ func (m *module) getPerGroupPodStatusCounts(filter *qbtypes.Filter,groupBy []qbtypes.GroupByKey,pageGroups []map[string]string,+ filterByPodStatus []inframonitoringtypes.PodStatus,) (map[string]podStatusCounts, error) {- if len(pageGroups) == 0 || len(groupBy) == 0 {+ // return early if no group by or (no pagegroups provided plus no filterBystatus given for a full scan)+ if len(groupBy) == 0 || (len(pageGroups) == 0 && len(filterByPodStatus) == 0) {return map[string]podStatusCounts{}, nil}+ var (+ filterClause *sqlbuilder.WhereClause+ err error+ userFilterExpr string+ )+// Merge user filter with page-groups IN clauses.- userFilterExpr := ""if filter != nil {userFilterExpr = filter.Expression}@@ -322,10 +382,7 @@ func (m *module) getPerGroupPodStatusCounts(// CTEs, and buildFilterClause hits the metadata store + parses the// expression, so we don't want to repeat it per CTE. AddWhereClause only// reads the clause, so the same instance is safe to attach to each builder.- var (- filterClause *sqlbuilder.WhereClause- err error- )+if mergedFilterExpr != "" {filterClause, err = m.buildFilterClause(ctx, orgID, &qbtypes.Filter{Expression: mergedFilterExpr}, start, end)if err != nil {@@ -540,11 +597,15 @@ func (m *module) getPerGroupPodStatusCounts(countGroupBy = append(countGroupBy, col)}countSelectCols = append(countSelectCols, statusCountCols...)- countSQL := fmt.Sprintf(- "SELECT %s FROM pod_status GROUP BY %s",- strings.Join(countSelectCols, ", "),- strings.Join(countGroupBy, ", "),- )++ // Outer count query. Built with sqlbuilder so the status push-down uses a+ // proper IN (keep only pods whose display status is in the requested set).+ countBuilder := sqlbuilder.NewSelectBuilder()+ countBuilder.Select(countSelectCols...)+ countBuilder.From("pod_status")+ applyPodStatusFilter(countBuilder, filterByPodStatus)+ countBuilder.GroupBy(countGroupBy...)+ countSQL, countArgs := countBuilder.BuildWithFlavor(sqlbuilder.ClickHouse)// Combine CTEs + outer. Arg order mirrors CTE declaration order.cteFragments := []string{@@ -561,7 +622,7 @@ func (m *module) getPerGroupPodStatusCounts(phaseFpsArgs, phasePerPodArgs,podReasonFpsArgs, podReasonPerPodArgs,containerReasonFpsArgs, containerInnerArgs,- }, nil)+ }, countArgs)rows, err := m.telemetryStore.ClickhouseDB().Query(ctx, finalSQL, finalArgs...)if err != nil {diff --git a/pkg/modules/inframonitoring/implinframonitoring/pods_test.go b/pkg/modules/inframonitoring/implinframonitoring/pods_test.gonew file mode 100644index 0000000000..74e57a86e8--- /dev/null+++ b/pkg/modules/inframonitoring/implinframonitoring/pods_test.go@@ -0,0 +1,59 @@+package implinframonitoring++import (+ "strings"+ "testing"++ "github.com/SigNoz/signoz/pkg/types/inframonitoringtypes"+ "github.com/huandu/go-sqlbuilder"+ "github.com/stretchr/testify/assert"+)++func TestApplyPodStatusFilter(t *testing.T) {+ tests := []struct {+ name string+ statuses []inframonitoringtypes.PodStatus+ wantWhere bool+ wantArgs []any+ }{+ {+ name: "empty set yields no clause",+ statuses: nil,+ wantWhere: false,+ wantArgs: nil,+ },+ {+ name: "single status pushes lowercased arg via IN",+ statuses: []inframonitoringtypes.PodStatus{inframonitoringtypes.PodStatusRunning},+ wantWhere: true,+ wantArgs: []any{"running"},+ },+ {+ name: "multiple statuses push lowercased args via IN",+ statuses: []inframonitoringtypes.PodStatus{+ inframonitoringtypes.PodStatusRunning,+ inframonitoringtypes.PodStatusCrashLoopBackOff,+ },+ wantWhere: true,+ wantArgs: []any{"running", "crashloopbackoff"},+ },+ }++ for _, tt := range tests {+ t.Run(tt.name, func(t *testing.T) {+ cb := sqlbuilder.NewSelectBuilder()+ cb.Select("pod_uid")+ cb.From("pod_status")+ applyPodStatusFilter(cb, tt.statuses)+ sql, args := cb.BuildWithFlavor(sqlbuilder.ClickHouse)++ hasWhere := strings.Contains(sql, "lower(display_status) IN (")+ assert.Equal(t, tt.wantWhere, hasWhere)+ if len(tt.wantArgs) == 0 {+ assert.Empty(t, args)+ } else {+ assert.Equal(t, tt.wantArgs, args)+ }+ })+ }+}diff --git a/pkg/modules/inframonitoring/implinframonitoring/statefulsets.go b/pkg/modules/inframonitoring/implinframonitoring/statefulsets.goindex b1b49f92f2..c184c92312 100644--- a/pkg/modules/inframonitoring/implinframonitoring/statefulsets.go+++ b/pkg/modules/inframonitoring/implinframonitoring/statefulsets.go@@ -82,20 +82,34 @@ func buildStatefulSetRecords(return records}+// getTopStatefulSetGroupsAndMetadata concurrently fetches metadata + the ordering-metric+// ranking (plus the full-scope pod-status keyset when filtering, to intersect both).func (m *module) getTopStatefulSetGroupsAndMetadata(ctx context.Context,orgID valuer.UUID,req *inframonitoringtypes.PostableStatefulSets,-) ([]map[string]string, map[string]map[string]string, error) {+) ([]map[string]string, map[string]map[string]string, map[string]podStatusCounts, *qbtypes.QueryWarnData, error) {var (- orderByKey string- metadataMap map[string]map[string]string- allMetricGroups []rankedGroup+ orderByKey string+ metadataMap map[string]map[string]string+ allMetricGroups []rankedGroup+ statusCounts map[string]podStatusCounts+ statusWarning *qbtypes.QueryWarnData+ filter *qbtypes.Filter+ filterByPodStatus []inframonitoringtypes.PodStatus)orderByKey = req.OrderBy.Key.Name+ // When filtering by pod status, resolve the full-scope status keyset+ // concurrently (pageGroups=nil spans all groups under the user filter) so it+ // can intersect metadata + ranked groups below.+ if req.Filter != nil {+ filter = &req.Filter.Filter+ filterByPodStatus = req.Filter.FilterByPodStatus+ }+g, gCtx := errgroup.WithContext(ctx)g.Go(func() error {@@ -104,12 +118,26 @@ func (m *module) getTopStatefulSetGroupsAndMetadata(return err})+ if len(filterByPodStatus) != 0 {+ g.Go(func() error {+ var err error+ statusCounts, statusWarning, err = m.getPerGroupPodStatusCountsWithReqMetricChecks(gCtx, orgID, req.Start, req.End, filter, req.GroupBy, nil, filterByPodStatus)+ return err+ })+ }+if orderByKey == inframonitoringtypes.StatefulSetNameAttrKey {if err := g.Wait(); err != nil {- return nil, nil, err+ return nil, nil, nil, nil, err+ }+ // Secondary filter: keep only status-matching groups. A missing metric+ // yields an empty statusCounts, so this correctly empties the result+ // (the caller also surfaces the warning).+ if len(filterByPodStatus) != 0 {+ metadataMap = intersectMap(metadataMap, statusCounts)}pageGroups := inframonitoringtypes.PaginateMetadataByName(metadataMap, req.GroupBy, req.OrderBy.Direction, req.Offset, req.Limit, inframonitoringtypes.StatefulSetNameAttrKey)- return pageGroups, metadataMap, nil+ return pageGroups, metadataMap, statusCounts, statusWarning, nil}queryNamesForOrderBy := orderByToStatefulSetsQueryNames[orderByKey]@@ -155,10 +183,19 @@ func (m *module) getTopStatefulSetGroupsAndMetadata(})if err := g.Wait(); err != nil {- return nil, nil, err+ return nil, nil, nil, nil, err+ }++ // Secondary filter: intersect ranked groups + metadata with the status keyset.+ // A missing metric yields an empty statusCounts, correctly emptying the result+ // (the caller also surfaces the warning).+ if len(filterByPodStatus) != 0 {+ allMetricGroups = intersectRankedGroups(allMetricGroups, statusCounts)+ metadataMap = intersectMap(metadataMap, statusCounts)}- return paginateWithBackfill(allMetricGroups, metadataMap, req.GroupBy, req.Offset, req.Limit), metadataMap, nil+ pageGroups := paginateWithBackfill(allMetricGroups, metadataMap, req.GroupBy, req.Offset, req.Limit)+ return pageGroups, metadataMap, statusCounts, statusWarning, nil}func (m *module) getStatefulSetsTableMetadata(ctx context.Context, orgID valuer.UUID, req *inframonitoringtypes.PostableStatefulSets) (map[string]map[string]string, error) {@@ -168,5 +205,9 @@ func (m *module) getStatefulSetsTableMetadata(ctx context.Context, orgID valuer.nonGroupByAttrs = append(nonGroupByAttrs, key)}}- return m.getMetadata(ctx, orgID, statefulSetsTableMetricNamesList, req.GroupBy, nonGroupByAttrs, req.Filter, req.Start, req.End)+ var filter *qbtypes.Filter+ if req.Filter != nil {+ filter = &req.Filter.Filter+ }+ return m.getMetadata(ctx, orgID, statefulSetsTableMetricNamesList, req.GroupBy, nonGroupByAttrs, filter, req.Start, req.End)}diff --git a/pkg/types/inframonitoringtypes/clusters.go b/pkg/types/inframonitoringtypes/clusters.goindex 7cbf6529df..e638de5912 100644--- a/pkg/types/inframonitoringtypes/clusters.go+++ b/pkg/types/inframonitoringtypes/clusters.go@@ -42,13 +42,22 @@ type ClusterRecord struct {type PostableClusters struct {Start int64 `json:"start" required:"true"`End int64 `json:"end" required:"true"`- Filter *qbtypes.Filter `json:"filter"`+ Filter *ClusterFilter `json:"filter"`GroupBy []qbtypes.GroupByKey `json:"groupBy"`OrderBy *qbtypes.OrderBy `json:"orderBy"`Offset int `json:"offset"`Limit int `json:"limit" required:"true"`}+// ClusterFilter is the attribute filter plus optional secondary filters on the+// derived pod display status(es) (see PodStatus; matches any listed, OR) and node+// readiness (see NodeCondition; matches any listed, OR). Empty FilterByPodStatus / FilterByNodeReadiness = off.+type ClusterFilter struct {+ qbtypes.Filter `json:",inline"`+ FilterByPodStatus []PodStatus `json:"filterByPodStatus"`+ FilterByNodeReadiness []NodeCondition `json:"filterByNodeReadiness"`+}+// Validate ensures PostableClusters contains acceptable values.func (req *PostableClusters) Validate() error {if req == nil {@@ -88,6 +97,19 @@ func (req *PostableClusters) Validate() error {return errors.NewInvalidInputf(errors.CodeInvalidInput, "offset cannot be negative")}+ if req.Filter != nil {+ for _, s := range req.Filter.FilterByPodStatus {+ if !s.IsFilterable() {+ return errors.NewInvalidInputf(errors.CodeInvalidInput, "invalid filter by pod status: %s", s)+ }+ }+ for _, c := range req.Filter.FilterByNodeReadiness {+ if !c.IsFilterable() {+ return errors.NewInvalidInputf(errors.CodeInvalidInput, "invalid filter by node readiness: %s", c)+ }+ }+ }+if req.OrderBy != nil {if !slices.Contains(ClustersValidOrderByKeys, req.OrderBy.Key.Name) {return errors.NewInvalidInputf(errors.CodeInvalidInput, "invalid order by key: %s", req.OrderBy.Key.Name)diff --git a/pkg/types/inframonitoringtypes/containers.go b/pkg/types/inframonitoringtypes/containers.goindex 09436ea699..d6a528b364 100644--- a/pkg/types/inframonitoringtypes/containers.go+++ b/pkg/types/inframonitoringtypes/containers.go@@ -75,13 +75,21 @@ type ContainerRecord struct {type PostableContainers struct {Start int64 `json:"start" required:"true"`End int64 `json:"end" required:"true"`- Filter *qbtypes.Filter `json:"filter"`+ Filter *ContainerFilter `json:"filter"`GroupBy []qbtypes.GroupByKey `json:"groupBy"`OrderBy *qbtypes.OrderBy `json:"orderBy"`Offset int `json:"offset"`Limit int `json:"limit" required:"true"`}+// ContainerFilter is the attribute filter plus an optional secondary filter on+// the derived container display status(es) (see ContainerStatus; matches any+// listed, OR). Empty FilterByContainerStatus = off.+type ContainerFilter struct {+ qbtypes.Filter `json:",inline"`+ FilterByContainerStatus []ContainerStatus `json:"filterByContainerStatus"`+}+// Validate ensures PostableContainers contains acceptable values.func (req *PostableContainers) Validate() error {if req == nil {@@ -121,6 +129,14 @@ func (req *PostableContainers) Validate() error {return errors.NewInvalidInputf(errors.CodeInvalidInput, "offset cannot be negative")}+ if req.Filter != nil {+ for _, c := range req.Filter.FilterByContainerStatus {+ if !c.IsFilterable() {+ return errors.NewInvalidInputf(errors.CodeInvalidInput, "invalid filter by container status: %s", c)+ }+ }+ }+if req.OrderBy != nil {if !slices.Contains(ContainersValidOrderByKeys, req.OrderBy.Key.Name) {return errors.NewInvalidInputf(errors.CodeInvalidInput, "invalid order by key: %s", req.OrderBy.Key.Name)diff --git a/pkg/types/inframonitoringtypes/containers_constants.go b/pkg/types/inframonitoringtypes/containers_constants.goindex f89c0a6504..0c34156d66 100644--- a/pkg/types/inframonitoringtypes/containers_constants.go+++ b/pkg/types/inframonitoringtypes/containers_constants.go@@ -1,6 +1,10 @@package inframonitoringtypes-import "github.com/SigNoz/signoz/pkg/valuer"+import (+ "slices"++ "github.com/SigNoz/signoz/pkg/valuer"+)// ContainerStatus is the kubectl-style display status of a container, derived// from k8s.container.status.state (base) + k8s.container.status.reason (overlay).@@ -31,6 +35,12 @@ var (ContainerStatusNoData = ContainerStatus{valuer.NewString("no_data")})+// IsFilterable reports whether c is a concrete, user-filterable+// container status: any Enum() member except the no_data sentinel.+func (c ContainerStatus) IsFilterable() bool {+ return c != ContainerStatusNoData && slices.Contains((ContainerStatus{}).Enum(), any(c))+}+func (ContainerStatus) Enum() []any {return []any{ContainerStatusRunning,diff --git a/pkg/types/inframonitoringtypes/daemonsets.go b/pkg/types/inframonitoringtypes/daemonsets.goindex b792898785..81af9b6298 100644--- a/pkg/types/inframonitoringtypes/daemonsets.go+++ b/pkg/types/inframonitoringtypes/daemonsets.go@@ -36,13 +36,20 @@ type DaemonSetRecord struct {type PostableDaemonSets struct {Start int64 `json:"start" required:"true"`End int64 `json:"end" required:"true"`- Filter *qbtypes.Filter `json:"filter"`+ Filter *DaemonSetFilter `json:"filter"`GroupBy []qbtypes.GroupByKey `json:"groupBy"`OrderBy *qbtypes.OrderBy `json:"orderBy"`Offset int `json:"offset"`Limit int `json:"limit" required:"true"`}+// DaemonSetFilter is the attribute filter plus an optional secondary filter on the+// derived pod display status(es) (see PodStatus); matches any listed (OR). Empty = off.+type DaemonSetFilter struct {+ qbtypes.Filter `json:",inline"`+ FilterByPodStatus []PodStatus `json:"filterByPodStatus"`+}+// Validate ensures PostableDaemonSets contains acceptable values.func (req *PostableDaemonSets) Validate() error {if req == nil {@@ -82,6 +89,14 @@ func (req *PostableDaemonSets) Validate() error {return errors.NewInvalidInputf(errors.CodeInvalidInput, "offset cannot be negative")}+ if req.Filter != nil {+ for _, s := range req.Filter.FilterByPodStatus {+ if !s.IsFilterable() {+ return errors.NewInvalidInputf(errors.CodeInvalidInput, "invalid filter by pod status: %s", s)+ }+ }+ }+if req.OrderBy != nil {if !slices.Contains(DaemonSetsValidOrderByKeys, req.OrderBy.Key.Name) {return errors.NewInvalidInputf(errors.CodeInvalidInput, "invalid order by key: %s", req.OrderBy.Key.Name)diff --git a/pkg/types/inframonitoringtypes/deployments.go b/pkg/types/inframonitoringtypes/deployments.goindex 22dd56fdcc..6554fcee20 100644--- a/pkg/types/inframonitoringtypes/deployments.go+++ b/pkg/types/inframonitoringtypes/deployments.go@@ -34,13 +34,20 @@ type DeploymentRecord struct {type PostableDeployments struct {Start int64 `json:"start" required:"true"`End int64 `json:"end" required:"true"`- Filter *qbtypes.Filter `json:"filter"`+ Filter *DeploymentFilter `json:"filter"`GroupBy []qbtypes.GroupByKey `json:"groupBy"`OrderBy *qbtypes.OrderBy `json:"orderBy"`Offset int `json:"offset"`Limit int `json:"limit" required:"true"`}+// DeploymentFilter is the attribute filter plus an optional secondary filter on the+// derived pod display status(es) (see PodStatus); matches any listed (OR). Empty = off.+type DeploymentFilter struct {+ qbtypes.Filter `json:",inline"`+ FilterByPodStatus []PodStatus `json:"filterByPodStatus"`+}+// Validate ensures PostableDeployments contains acceptable values.func (req *PostableDeployments) Validate() error {if req == nil {@@ -80,6 +87,14 @@ func (req *PostableDeployments) Validate() error {return errors.NewInvalidInputf(errors.CodeInvalidInput, "offset cannot be negative")}+ if req.Filter != nil {+ for _, s := range req.Filter.FilterByPodStatus {+ if !s.IsFilterable() {+ return errors.NewInvalidInputf(errors.CodeInvalidInput, "invalid filter by pod status: %s", s)+ }+ }+ }+if req.OrderBy != nil {if !slices.Contains(DeploymentsValidOrderByKeys, req.OrderBy.Key.Name) {return errors.NewInvalidInputf(errors.CodeInvalidInput, "invalid order by key: %s", req.OrderBy.Key.Name)diff --git a/pkg/types/inframonitoringtypes/jobs.go b/pkg/types/inframonitoringtypes/jobs.goindex 63db9b2d5c..8c20e9cefe 100644--- a/pkg/types/inframonitoringtypes/jobs.go+++ b/pkg/types/inframonitoringtypes/jobs.go@@ -36,13 +36,20 @@ type JobRecord struct {type PostableJobs struct {Start int64 `json:"start" required:"true"`End int64 `json:"end" required:"true"`- Filter *qbtypes.Filter `json:"filter"`+ Filter *JobFilter `json:"filter"`GroupBy []qbtypes.GroupByKey `json:"groupBy"`OrderBy *qbtypes.OrderBy `json:"orderBy"`Offset int `json:"offset"`Limit int `json:"limit" required:"true"`}+// JobFilter is the attribute filter plus an optional secondary filter on the+// derived pod display status(es) (see PodStatus); matches any listed (OR). Empty = off.+type JobFilter struct {+ qbtypes.Filter `json:",inline"`+ FilterByPodStatus []PodStatus `json:"filterByPodStatus"`+}+// Validate ensures PostableJobs contains acceptable values.func (req *PostableJobs) Validate() error {if req == nil {@@ -82,6 +89,14 @@ func (req *PostableJobs) Validate() error {return errors.NewInvalidInputf(errors.CodeInvalidInput, "offset cannot be negative")}+ if req.Filter != nil {+ for _, s := range req.Filter.FilterByPodStatus {+ if !s.IsFilterable() {+ return errors.NewInvalidInputf(errors.CodeInvalidInput, "invalid filter by pod status: %s", s)+ }+ }+ }+if req.OrderBy != nil {if !slices.Contains(JobsValidOrderByKeys, req.OrderBy.Key.Name) {return errors.NewInvalidInputf(errors.CodeInvalidInput, "invalid order by key: %s", req.OrderBy.Key.Name)diff --git a/pkg/types/inframonitoringtypes/namespaces.go b/pkg/types/inframonitoringtypes/namespaces.goindex a032363311..d01a0156c6 100644--- a/pkg/types/inframonitoringtypes/namespaces.go+++ b/pkg/types/inframonitoringtypes/namespaces.go@@ -34,13 +34,20 @@ type NamespaceRecord struct {type PostableNamespaces struct {Start int64 `json:"start" required:"true"`End int64 `json:"end" required:"true"`- Filter *qbtypes.Filter `json:"filter"`+ Filter *NamespaceFilter `json:"filter"`GroupBy []qbtypes.GroupByKey `json:"groupBy"`OrderBy *qbtypes.OrderBy `json:"orderBy"`Offset int `json:"offset"`Limit int `json:"limit" required:"true"`}+// NamespaceFilter is the attribute filter plus an optional secondary filter on the+// derived pod display status(es) (see PodStatus); matches any listed (OR). Empty = off.+type NamespaceFilter struct {+ qbtypes.Filter `json:",inline"`+ FilterByPodStatus []PodStatus `json:"filterByPodStatus"`+}+// Validate ensures PostableNamespaces contains acceptable values.func (req *PostableNamespaces) Validate() error {if req == nil {@@ -80,6 +87,14 @@ func (req *PostableNamespaces) Validate() error {return errors.NewInvalidInputf(errors.CodeInvalidInput, "offset cannot be negative")}+ if req.Filter != nil {+ for _, s := range req.Filter.FilterByPodStatus {+ if !s.IsFilterable() {+ return errors.NewInvalidInputf(errors.CodeInvalidInput, "invalid filter by pod status: %s", s)+ }+ }+ }+if req.OrderBy != nil {if !slices.Contains(NamespacesValidOrderByKeys, req.OrderBy.Key.Name) {return errors.NewInvalidInputf(errors.CodeInvalidInput, "invalid order by key: %s", req.OrderBy.Key.Name)diff --git a/pkg/types/inframonitoringtypes/nodes.go b/pkg/types/inframonitoringtypes/nodes.goindex 60b66644ce..f58e47356c 100644--- a/pkg/types/inframonitoringtypes/nodes.go+++ b/pkg/types/inframonitoringtypes/nodes.go@@ -39,13 +39,22 @@ type NodeRecord struct {type PostableNodes struct {Start int64 `json:"start" required:"true"`End int64 `json:"end" required:"true"`- Filter *qbtypes.Filter `json:"filter"`+ Filter *NodeFilter `json:"filter"`GroupBy []qbtypes.GroupByKey `json:"groupBy"`OrderBy *qbtypes.OrderBy `json:"orderBy"`Offset int `json:"offset"`Limit int `json:"limit" required:"true"`}+// NodeFilter is the attribute filter plus an optional secondary filter on the+// derived pod display status(es) (see PodStatus; matches any listed, OR) and node+// readiness (see NodeCondition; matches any listed, OR). Empty FilterByPodStatus / FilterByNodeReadiness = off.+type NodeFilter struct {+ qbtypes.Filter `json:",inline"`+ FilterByPodStatus []PodStatus `json:"filterByPodStatus"`+ FilterByNodeReadiness []NodeCondition `json:"filterByNodeReadiness"`+}+// Validate ensures PostableNodes contains acceptable values.func (req *PostableNodes) Validate() error {if req == nil {@@ -85,6 +94,19 @@ func (req *PostableNodes) Validate() error {return errors.NewInvalidInputf(errors.CodeInvalidInput, "offset cannot be negative")}+ if req.Filter != nil {+ for _, s := range req.Filter.FilterByPodStatus {+ if !s.IsFilterable() {+ return errors.NewInvalidInputf(errors.CodeInvalidInput, "invalid filter by pod status: %s", s)+ }+ }+ for _, c := range req.Filter.FilterByNodeReadiness {+ if !c.IsFilterable() {+ return errors.NewInvalidInputf(errors.CodeInvalidInput, "invalid filter by node readiness: %s", c)+ }+ }+ }+if req.OrderBy != nil {if !slices.Contains(NodesValidOrderByKeys, req.OrderBy.Key.Name) {return errors.NewInvalidInputf(errors.CodeInvalidInput, "invalid order by key: %s", req.OrderBy.Key.Name)diff --git a/pkg/types/inframonitoringtypes/nodes_constants.go b/pkg/types/inframonitoringtypes/nodes_constants.goindex 8516c2c15e..fe28ecf1ce 100644--- a/pkg/types/inframonitoringtypes/nodes_constants.go+++ b/pkg/types/inframonitoringtypes/nodes_constants.go@@ -1,6 +1,10 @@package inframonitoringtypes-import "github.com/SigNoz/signoz/pkg/valuer"+import (+ "slices"++ "github.com/SigNoz/signoz/pkg/valuer"+)type NodeCondition struct {valuer.String@@ -20,6 +24,12 @@ func (NodeCondition) Enum() []any {}}+// IsFilterable reports whether c is a concrete, user-filterable+// node readiness: any Enum() member except the no_data sentinel.+func (c NodeCondition) IsFilterable() bool {+ return c != NodeConditionNoData && slices.Contains((NodeCondition{}).Enum(), any(c))+}+// Numeric values emitted by the k8s.node.condition_ready metric// (source: OTel kubeletstats receiver).const (diff --git a/pkg/types/inframonitoringtypes/pods.go b/pkg/types/inframonitoringtypes/pods.goindex aafc1579ca..0aa26a50ff 100644--- a/pkg/types/inframonitoringtypes/pods.go+++ b/pkg/types/inframonitoringtypes/pods.go@@ -63,13 +63,20 @@ type PodRecord struct {type PostablePods struct {Start int64 `json:"start" required:"true"`End int64 `json:"end" required:"true"`- Filter *qbtypes.Filter `json:"filter"`+ Filter *PodFilter `json:"filter"`GroupBy []qbtypes.GroupByKey `json:"groupBy"`OrderBy *qbtypes.OrderBy `json:"orderBy"`Offset int `json:"offset"`Limit int `json:"limit" required:"true"`}+// PodFilter is the attribute filter plus an optional secondary filter on the+// derived pod display status(es) (see PodStatus); matches any listed (OR). Empty = off.+type PodFilter struct {+ qbtypes.Filter `json:",inline"`+ FilterByPodStatus []PodStatus `json:"filterByPodStatus"`+}+// Validate ensures PostablePods contains acceptable values.func (req *PostablePods) Validate() error {if req == nil {@@ -109,6 +116,14 @@ func (req *PostablePods) Validate() error {return errors.NewInvalidInputf(errors.CodeInvalidInput, "offset cannot be negative")}+ if req.Filter != nil {+ for _, s := range req.Filter.FilterByPodStatus {+ if !s.IsFilterable() {+ return errors.NewInvalidInputf(errors.CodeInvalidInput, "invalid filter by pod status: %s", s)+ }+ }+ }+if req.OrderBy != nil {if !slices.Contains(PodsValidOrderByKeys, req.OrderBy.Key.Name) {return errors.NewInvalidInputf(errors.CodeInvalidInput, "invalid order by key: %s", req.OrderBy.Key.Name)diff --git a/pkg/types/inframonitoringtypes/pods_constants.go b/pkg/types/inframonitoringtypes/pods_constants.goindex a4e2e78119..e088380de6 100644--- a/pkg/types/inframonitoringtypes/pods_constants.go+++ b/pkg/types/inframonitoringtypes/pods_constants.go@@ -1,6 +1,10 @@package inframonitoringtypes-import "github.com/SigNoz/signoz/pkg/valuer"+import (+ "slices"++ "github.com/SigNoz/signoz/pkg/valuer"+)// PodStatus is the kubectl-style pod display status, derived from// k8s.pod.phase + k8s.pod.status_reason + k8s.container.status.reason@@ -66,6 +70,12 @@ func (PodStatus) Enum() []any {}}+// IsFilterable reports whether s is a concrete, user-filterable pod+// status: any Enum() member except the no_data sentinel.+func (s PodStatus) IsFilterable() bool {+ return s != PodStatusNoData && slices.Contains((PodStatus{}).Enum(), any(s))+}+const PodNameAttrKey = "k8s.pod.name"const (diff --git a/pkg/types/inframonitoringtypes/statefulsets.go b/pkg/types/inframonitoringtypes/statefulsets.goindex 9ab9dcd17a..b0112272d3 100644--- a/pkg/types/inframonitoringtypes/statefulsets.go+++ b/pkg/types/inframonitoringtypes/statefulsets.go@@ -34,13 +34,20 @@ type StatefulSetRecord struct {type PostableStatefulSets struct {Start int64 `json:"start" required:"true"`End int64 `json:"end" required:"true"`- Filter *qbtypes.Filter `json:"filter"`+ Filter *StatefulSetFilter `json:"filter"`GroupBy []qbtypes.GroupByKey `json:"groupBy"`OrderBy *qbtypes.OrderBy `json:"orderBy"`Offset int `json:"offset"`Limit int `json:"limit" required:"true"`}+// StatefulSetFilter is the attribute filter plus an optional secondary filter on the+// derived pod display status(es) (see PodStatus); matches any listed (OR). Empty = off.+type StatefulSetFilter struct {+ qbtypes.Filter `json:",inline"`+ FilterByPodStatus []PodStatus `json:"filterByPodStatus"`+}+// Validate ensures PostableStatefulSets contains acceptable values.func (req *PostableStatefulSets) Validate() error {if req == nil {@@ -80,6 +87,14 @@ func (req *PostableStatefulSets) Validate() error {return errors.NewInvalidInputf(errors.CodeInvalidInput, "offset cannot be negative")}+ if req.Filter != nil {+ for _, s := range req.Filter.FilterByPodStatus {+ if !s.IsFilterable() {+ return errors.NewInvalidInputf(errors.CodeInvalidInput, "invalid filter by pod status: %s", s)+ }+ }+ }+if req.OrderBy != nil {if !slices.Contains(StatefulSetsValidOrderByKeys, req.OrderBy.Key.Name) {return errors.NewInvalidInputf(errors.CodeInvalidInput, "invalid order by key: %s", req.OrderBy.Key.Name)diff --git a/tests/integration/testdata/inframonitoring/daemonsets_non_ds_pods.jsonl b/tests/integration/testdata/inframonitoring/daemonsets_non_ds_pods.jsonlindex c0a14a1dd2..e0993f53bc 100644--- a/tests/integration/testdata/inframonitoring/daemonsets_non_ds_pods.jsonl+++ b/tests/integration/testdata/inframonitoring/daemonsets_non_ds_pods.jsonl@@ -88,3 +88,15 @@{"metric_name":"k8s.pod.phase","labels":{"k8s.pod.uid":"nd-ss-uid","k8s.pod.name":"nd-ss-pod","k8s.statefulset.name":"nd-ss","k8s.namespace.name":"ns-nd","k8s.cluster.name":"cluster-x"},"timestamp":"2025-01-10T10:00:00+00:00","value":2,"temporality":"Unspecified","type_":"Gauge","is_monotonic":false}{"metric_name":"k8s.pod.phase","labels":{"k8s.pod.uid":"nd-ss-uid","k8s.pod.name":"nd-ss-pod","k8s.statefulset.name":"nd-ss","k8s.namespace.name":"ns-nd","k8s.cluster.name":"cluster-x"},"timestamp":"2025-01-10T10:02:00+00:00","value":2,"temporality":"Unspecified","type_":"Gauge","is_monotonic":false}{"metric_name":"k8s.pod.phase","labels":{"k8s.pod.uid":"nd-ss-uid","k8s.pod.name":"nd-ss-pod","k8s.statefulset.name":"nd-ss","k8s.namespace.name":"ns-nd","k8s.cluster.name":"cluster-x"},"timestamp":"2025-01-10T10:04:00+00:00","value":2,"temporality":"Unspecified","type_":"Gauge","is_monotonic":false}+{"metric_name": "k8s.pod.status_reason", "labels": {"k8s.pod.uid": "nd-ds-p1-uid", "k8s.pod.name": "nd-ds-p1", "k8s.daemonset.name": "nd-ds", "k8s.namespace.name": "ns-nd", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:00:00+00:00", "value": 6, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.phase", "labels": {"k8s.pod.uid": "nd-ds-p1-clbo-uid", "k8s.pod.name": "nd-ds-p1-clbo", "k8s.daemonset.name": "nd-ds", "k8s.namespace.name": "ns-nd", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:00:00+00:00", "value": 2, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.status_reason", "labels": {"k8s.pod.uid": "nd-ds-p1-clbo-uid", "k8s.pod.name": "nd-ds-p1-clbo", "k8s.daemonset.name": "nd-ds", "k8s.namespace.name": "ns-nd", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:00:00+00:00", "value": 6, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.container.status.reason", "labels": {"k8s.pod.uid": "nd-ds-p1-clbo-uid", "k8s.pod.name": "nd-ds-p1-clbo", "k8s.daemonset.name": "nd-ds", "k8s.namespace.name": "ns-nd", "k8s.cluster.name": "cluster-x", "k8s.container.name": "app", "k8s.container.status.reason": "CrashLoopBackOff"}, "timestamp": "2025-01-10T10:00:00+00:00", "value": 1, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.status_reason", "labels": {"k8s.pod.uid": "nd-ds-p1-uid", "k8s.pod.name": "nd-ds-p1", "k8s.daemonset.name": "nd-ds", "k8s.namespace.name": "ns-nd", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:02:00+00:00", "value": 6, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.phase", "labels": {"k8s.pod.uid": "nd-ds-p1-clbo-uid", "k8s.pod.name": "nd-ds-p1-clbo", "k8s.daemonset.name": "nd-ds", "k8s.namespace.name": "ns-nd", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:02:00+00:00", "value": 2, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.status_reason", "labels": {"k8s.pod.uid": "nd-ds-p1-clbo-uid", "k8s.pod.name": "nd-ds-p1-clbo", "k8s.daemonset.name": "nd-ds", "k8s.namespace.name": "ns-nd", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:02:00+00:00", "value": 6, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.container.status.reason", "labels": {"k8s.pod.uid": "nd-ds-p1-clbo-uid", "k8s.pod.name": "nd-ds-p1-clbo", "k8s.daemonset.name": "nd-ds", "k8s.namespace.name": "ns-nd", "k8s.cluster.name": "cluster-x", "k8s.container.name": "app", "k8s.container.status.reason": "CrashLoopBackOff"}, "timestamp": "2025-01-10T10:02:00+00:00", "value": 1, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.status_reason", "labels": {"k8s.pod.uid": "nd-ds-p1-uid", "k8s.pod.name": "nd-ds-p1", "k8s.daemonset.name": "nd-ds", "k8s.namespace.name": "ns-nd", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:04:00+00:00", "value": 6, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.phase", "labels": {"k8s.pod.uid": "nd-ds-p1-clbo-uid", "k8s.pod.name": "nd-ds-p1-clbo", "k8s.daemonset.name": "nd-ds", "k8s.namespace.name": "ns-nd", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:04:00+00:00", "value": 2, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.status_reason", "labels": {"k8s.pod.uid": "nd-ds-p1-clbo-uid", "k8s.pod.name": "nd-ds-p1-clbo", "k8s.daemonset.name": "nd-ds", "k8s.namespace.name": "ns-nd", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:04:00+00:00", "value": 6, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.container.status.reason", "labels": {"k8s.pod.uid": "nd-ds-p1-clbo-uid", "k8s.pod.name": "nd-ds-p1-clbo", "k8s.daemonset.name": "nd-ds", "k8s.namespace.name": "ns-nd", "k8s.cluster.name": "cluster-x", "k8s.container.name": "app", "k8s.container.status.reason": "CrashLoopBackOff"}, "timestamp": "2025-01-10T10:04:00+00:00", "value": 1, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}diff --git a/tests/integration/testdata/inframonitoring/jobs_non_job_pods.jsonl b/tests/integration/testdata/inframonitoring/jobs_non_job_pods.jsonlindex 5d06c845dc..843b42d056 100644--- a/tests/integration/testdata/inframonitoring/jobs_non_job_pods.jsonl+++ b/tests/integration/testdata/inframonitoring/jobs_non_job_pods.jsonl@@ -94,3 +94,15 @@{"metric_name":"k8s.pod.phase","labels":{"k8s.pod.uid":"nj-ss-uid","k8s.pod.name":"nj-ss-pod","k8s.statefulset.name":"nj-ss","k8s.namespace.name":"ns-nj","k8s.cluster.name":"cluster-x"},"timestamp":"2025-01-10T10:00:00+00:00","value":2,"temporality":"Unspecified","type_":"Gauge","is_monotonic":false}{"metric_name":"k8s.pod.phase","labels":{"k8s.pod.uid":"nj-ss-uid","k8s.pod.name":"nj-ss-pod","k8s.statefulset.name":"nj-ss","k8s.namespace.name":"ns-nj","k8s.cluster.name":"cluster-x"},"timestamp":"2025-01-10T10:02:00+00:00","value":2,"temporality":"Unspecified","type_":"Gauge","is_monotonic":false}{"metric_name":"k8s.pod.phase","labels":{"k8s.pod.uid":"nj-ss-uid","k8s.pod.name":"nj-ss-pod","k8s.statefulset.name":"nj-ss","k8s.namespace.name":"ns-nj","k8s.cluster.name":"cluster-x"},"timestamp":"2025-01-10T10:04:00+00:00","value":2,"temporality":"Unspecified","type_":"Gauge","is_monotonic":false}+{"metric_name": "k8s.pod.status_reason", "labels": {"k8s.pod.uid": "nj-job-p1-uid", "k8s.pod.name": "nj-job-p1", "k8s.job.name": "nj-job", "k8s.namespace.name": "ns-nj", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:00:00+00:00", "value": 6, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.phase", "labels": {"k8s.pod.uid": "nj-job-p1-clbo-uid", "k8s.pod.name": "nj-job-p1-clbo", "k8s.job.name": "nj-job", "k8s.namespace.name": "ns-nj", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:00:00+00:00", "value": 2, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.status_reason", "labels": {"k8s.pod.uid": "nj-job-p1-clbo-uid", "k8s.pod.name": "nj-job-p1-clbo", "k8s.job.name": "nj-job", "k8s.namespace.name": "ns-nj", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:00:00+00:00", "value": 6, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.container.status.reason", "labels": {"k8s.pod.uid": "nj-job-p1-clbo-uid", "k8s.pod.name": "nj-job-p1-clbo", "k8s.job.name": "nj-job", "k8s.namespace.name": "ns-nj", "k8s.cluster.name": "cluster-x", "k8s.container.name": "app", "k8s.container.status.reason": "CrashLoopBackOff"}, "timestamp": "2025-01-10T10:00:00+00:00", "value": 1, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.status_reason", "labels": {"k8s.pod.uid": "nj-job-p1-uid", "k8s.pod.name": "nj-job-p1", "k8s.job.name": "nj-job", "k8s.namespace.name": "ns-nj", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:02:00+00:00", "value": 6, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.phase", "labels": {"k8s.pod.uid": "nj-job-p1-clbo-uid", "k8s.pod.name": "nj-job-p1-clbo", "k8s.job.name": "nj-job", "k8s.namespace.name": "ns-nj", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:02:00+00:00", "value": 2, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.status_reason", "labels": {"k8s.pod.uid": "nj-job-p1-clbo-uid", "k8s.pod.name": "nj-job-p1-clbo", "k8s.job.name": "nj-job", "k8s.namespace.name": "ns-nj", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:02:00+00:00", "value": 6, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.container.status.reason", "labels": {"k8s.pod.uid": "nj-job-p1-clbo-uid", "k8s.pod.name": "nj-job-p1-clbo", "k8s.job.name": "nj-job", "k8s.namespace.name": "ns-nj", "k8s.cluster.name": "cluster-x", "k8s.container.name": "app", "k8s.container.status.reason": "CrashLoopBackOff"}, "timestamp": "2025-01-10T10:02:00+00:00", "value": 1, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.status_reason", "labels": {"k8s.pod.uid": "nj-job-p1-uid", "k8s.pod.name": "nj-job-p1", "k8s.job.name": "nj-job", "k8s.namespace.name": "ns-nj", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:04:00+00:00", "value": 6, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.phase", "labels": {"k8s.pod.uid": "nj-job-p1-clbo-uid", "k8s.pod.name": "nj-job-p1-clbo", "k8s.job.name": "nj-job", "k8s.namespace.name": "ns-nj", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:04:00+00:00", "value": 2, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.status_reason", "labels": {"k8s.pod.uid": "nj-job-p1-clbo-uid", "k8s.pod.name": "nj-job-p1-clbo", "k8s.job.name": "nj-job", "k8s.namespace.name": "ns-nj", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:04:00+00:00", "value": 6, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.container.status.reason", "labels": {"k8s.pod.uid": "nj-job-p1-clbo-uid", "k8s.pod.name": "nj-job-p1-clbo", "k8s.job.name": "nj-job", "k8s.namespace.name": "ns-nj", "k8s.cluster.name": "cluster-x", "k8s.container.name": "app", "k8s.container.status.reason": "CrashLoopBackOff"}, "timestamp": "2025-01-10T10:04:00+00:00", "value": 1, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}diff --git a/tests/integration/testdata/inframonitoring/nodes_conditions.jsonl b/tests/integration/testdata/inframonitoring/nodes_conditions.jsonlindex 868b4948fb..9ea9fc3fef 100644--- a/tests/integration/testdata/inframonitoring/nodes_conditions.jsonl+++ b/tests/integration/testdata/inframonitoring/nodes_conditions.jsonl@@ -31,3 +31,48 @@{"metric_name":"k8s.pod.phase","labels":{"k8s.pod.uid":"carrier-pod-uid","k8s.pod.name":"carrier-pod","k8s.namespace.name":"carrier-ns","k8s.node.name":"carrier-phantom-host","k8s.cluster.name":"carrier-cluster"},"timestamp":"2025-01-10T10:00:00+00:00","value":2,"temporality":"Unspecified","type_":"Gauge","is_monotonic":false}{"metric_name":"k8s.pod.phase","labels":{"k8s.pod.uid":"carrier-pod-uid","k8s.pod.name":"carrier-pod","k8s.namespace.name":"carrier-ns","k8s.node.name":"carrier-phantom-host","k8s.cluster.name":"carrier-cluster"},"timestamp":"2025-01-10T10:02:00+00:00","value":2,"temporality":"Unspecified","type_":"Gauge","is_monotonic":false}{"metric_name":"k8s.pod.phase","labels":{"k8s.pod.uid":"carrier-pod-uid","k8s.pod.name":"carrier-pod","k8s.namespace.name":"carrier-ns","k8s.node.name":"carrier-phantom-host","k8s.cluster.name":"carrier-cluster"},"timestamp":"2025-01-10T10:04:00+00:00","value":2,"temporality":"Unspecified","type_":"Gauge","is_monotonic":false}+{"metric_name": "k8s.node.cpu.usage", "labels": {"k8s.node.name": "ready-n2", "k8s.node.uid": "ready-n2-uid", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:00:00+00:00", "value": 0.8, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.node.cpu.usage", "labels": {"k8s.node.name": "ready-n2", "k8s.node.uid": "ready-n2-uid", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:02:00+00:00", "value": 0.8, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.node.cpu.usage", "labels": {"k8s.node.name": "ready-n2", "k8s.node.uid": "ready-n2-uid", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:04:00+00:00", "value": 0.8, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.node.allocatable_cpu", "labels": {"k8s.node.name": "ready-n2", "k8s.node.uid": "ready-n2-uid", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:00:00+00:00", "value": 4.0, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.node.allocatable_cpu", "labels": {"k8s.node.name": "ready-n2", "k8s.node.uid": "ready-n2-uid", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:02:00+00:00", "value": 4.0, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.node.allocatable_cpu", "labels": {"k8s.node.name": "ready-n2", "k8s.node.uid": "ready-n2-uid", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:04:00+00:00", "value": 4.0, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.node.memory.working_set", "labels": {"k8s.node.name": "ready-n2", "k8s.node.uid": "ready-n2-uid", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:00:00+00:00", "value": 1000000000.0, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.node.memory.working_set", "labels": {"k8s.node.name": "ready-n2", "k8s.node.uid": "ready-n2-uid", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:02:00+00:00", "value": 1000000000.0, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.node.memory.working_set", "labels": {"k8s.node.name": "ready-n2", "k8s.node.uid": "ready-n2-uid", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:04:00+00:00", "value": 1000000000.0, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.node.allocatable_memory", "labels": {"k8s.node.name": "ready-n2", "k8s.node.uid": "ready-n2-uid", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:00:00+00:00", "value": 8000000000.0, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.node.allocatable_memory", "labels": {"k8s.node.name": "ready-n2", "k8s.node.uid": "ready-n2-uid", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:02:00+00:00", "value": 8000000000.0, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.node.allocatable_memory", "labels": {"k8s.node.name": "ready-n2", "k8s.node.uid": "ready-n2-uid", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:04:00+00:00", "value": 8000000000.0, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.node.condition_ready", "labels": {"k8s.node.name": "ready-n2", "k8s.node.uid": "ready-n2-uid", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:00:00+00:00", "value": 1, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.node.condition_ready", "labels": {"k8s.node.name": "ready-n2", "k8s.node.uid": "ready-n2-uid", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:02:00+00:00", "value": 1, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.node.condition_ready", "labels": {"k8s.node.name": "ready-n2", "k8s.node.uid": "ready-n2-uid", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:04:00+00:00", "value": 1, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.node.cpu.usage", "labels": {"k8s.node.name": "ready-n3", "k8s.node.uid": "ready-n3-uid", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:00:00+00:00", "value": 0.3, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.node.cpu.usage", "labels": {"k8s.node.name": "ready-n3", "k8s.node.uid": "ready-n3-uid", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:02:00+00:00", "value": 0.3, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.node.cpu.usage", "labels": {"k8s.node.name": "ready-n3", "k8s.node.uid": "ready-n3-uid", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:04:00+00:00", "value": 0.3, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.node.allocatable_cpu", "labels": {"k8s.node.name": "ready-n3", "k8s.node.uid": "ready-n3-uid", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:00:00+00:00", "value": 4.0, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.node.allocatable_cpu", "labels": {"k8s.node.name": "ready-n3", "k8s.node.uid": "ready-n3-uid", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:02:00+00:00", "value": 4.0, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.node.allocatable_cpu", "labels": {"k8s.node.name": "ready-n3", "k8s.node.uid": "ready-n3-uid", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:04:00+00:00", "value": 4.0, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.node.memory.working_set", "labels": {"k8s.node.name": "ready-n3", "k8s.node.uid": "ready-n3-uid", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:00:00+00:00", "value": 1000000000.0, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.node.memory.working_set", "labels": {"k8s.node.name": "ready-n3", "k8s.node.uid": "ready-n3-uid", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:02:00+00:00", "value": 1000000000.0, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.node.memory.working_set", "labels": {"k8s.node.name": "ready-n3", "k8s.node.uid": "ready-n3-uid", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:04:00+00:00", "value": 1000000000.0, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.node.allocatable_memory", "labels": {"k8s.node.name": "ready-n3", "k8s.node.uid": "ready-n3-uid", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:00:00+00:00", "value": 8000000000.0, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.node.allocatable_memory", "labels": {"k8s.node.name": "ready-n3", "k8s.node.uid": "ready-n3-uid", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:02:00+00:00", "value": 8000000000.0, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.node.allocatable_memory", "labels": {"k8s.node.name": "ready-n3", "k8s.node.uid": "ready-n3-uid", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:04:00+00:00", "value": 8000000000.0, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.node.condition_ready", "labels": {"k8s.node.name": "ready-n3", "k8s.node.uid": "ready-n3-uid", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:00:00+00:00", "value": 1, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.node.condition_ready", "labels": {"k8s.node.name": "ready-n3", "k8s.node.uid": "ready-n3-uid", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:02:00+00:00", "value": 1, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.node.condition_ready", "labels": {"k8s.node.name": "ready-n3", "k8s.node.uid": "ready-n3-uid", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:04:00+00:00", "value": 1, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.node.cpu.usage", "labels": {"k8s.node.name": "ready-n4", "k8s.node.uid": "ready-n4-uid", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:00:00+00:00", "value": 0.6, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.node.cpu.usage", "labels": {"k8s.node.name": "ready-n4", "k8s.node.uid": "ready-n4-uid", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:02:00+00:00", "value": 0.6, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.node.cpu.usage", "labels": {"k8s.node.name": "ready-n4", "k8s.node.uid": "ready-n4-uid", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:04:00+00:00", "value": 0.6, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.node.allocatable_cpu", "labels": {"k8s.node.name": "ready-n4", "k8s.node.uid": "ready-n4-uid", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:00:00+00:00", "value": 4.0, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.node.allocatable_cpu", "labels": {"k8s.node.name": "ready-n4", "k8s.node.uid": "ready-n4-uid", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:02:00+00:00", "value": 4.0, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.node.allocatable_cpu", "labels": {"k8s.node.name": "ready-n4", "k8s.node.uid": "ready-n4-uid", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:04:00+00:00", "value": 4.0, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.node.memory.working_set", "labels": {"k8s.node.name": "ready-n4", "k8s.node.uid": "ready-n4-uid", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:00:00+00:00", "value": 1000000000.0, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.node.memory.working_set", "labels": {"k8s.node.name": "ready-n4", "k8s.node.uid": "ready-n4-uid", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:02:00+00:00", "value": 1000000000.0, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.node.memory.working_set", "labels": {"k8s.node.name": "ready-n4", "k8s.node.uid": "ready-n4-uid", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:04:00+00:00", "value": 1000000000.0, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.node.allocatable_memory", "labels": {"k8s.node.name": "ready-n4", "k8s.node.uid": "ready-n4-uid", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:00:00+00:00", "value": 8000000000.0, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.node.allocatable_memory", "labels": {"k8s.node.name": "ready-n4", "k8s.node.uid": "ready-n4-uid", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:02:00+00:00", "value": 8000000000.0, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.node.allocatable_memory", "labels": {"k8s.node.name": "ready-n4", "k8s.node.uid": "ready-n4-uid", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:04:00+00:00", "value": 8000000000.0, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.node.condition_ready", "labels": {"k8s.node.name": "ready-n4", "k8s.node.uid": "ready-n4-uid", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:00:00+00:00", "value": 1, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.node.condition_ready", "labels": {"k8s.node.name": "ready-n4", "k8s.node.uid": "ready-n4-uid", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:02:00+00:00", "value": 1, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.node.condition_ready", "labels": {"k8s.node.name": "ready-n4", "k8s.node.uid": "ready-n4-uid", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:04:00+00:00", "value": 1, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}diff --git a/tests/integration/testdata/inframonitoring/pods_phases.jsonl b/tests/integration/testdata/inframonitoring/pods_phases.jsonlindex 50a2f0e82e..c3cb76923e 100644--- a/tests/integration/testdata/inframonitoring/pods_phases.jsonl+++ b/tests/integration/testdata/inframonitoring/pods_phases.jsonl@@ -145,3 +145,63 @@{"metric_name": "k8s.container.restarts", "labels": {"k8s.pod.uid": "unk-p-uid", "k8s.pod.name": "unk-p", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__", "k8s.container.name": "c-clbo"}, "timestamp": "2025-01-10T10:00:00+00:00", "value": 8, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}{"metric_name": "k8s.container.restarts", "labels": {"k8s.pod.uid": "unk-p-uid", "k8s.pod.name": "unk-p", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__", "k8s.container.name": "c-clbo"}, "timestamp": "2025-01-10T10:02:00+00:00", "value": 8, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}{"metric_name": "k8s.container.restarts", "labels": {"k8s.pod.uid": "unk-p-uid", "k8s.pod.name": "unk-p", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__", "k8s.container.name": "c-clbo"}, "timestamp": "2025-01-10T10:04:00+00:00", "value": 8, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.cpu.usage", "labels": {"k8s.pod.uid": "clbo-a-uid", "k8s.pod.name": "clbo-a", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__"}, "timestamp": "2025-01-10T10:00:00+00:00", "value": 0.7, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.cpu.usage", "labels": {"k8s.pod.uid": "clbo-a-uid", "k8s.pod.name": "clbo-a", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__"}, "timestamp": "2025-01-10T10:02:00+00:00", "value": 0.7, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.cpu.usage", "labels": {"k8s.pod.uid": "clbo-a-uid", "k8s.pod.name": "clbo-a", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__"}, "timestamp": "2025-01-10T10:04:00+00:00", "value": 0.7, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.cpu_request_utilization", "labels": {"k8s.pod.uid": "clbo-a-uid", "k8s.pod.name": "clbo-a", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__"}, "timestamp": "2025-01-10T10:00:00+00:00", "value": 0.5, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.cpu_request_utilization", "labels": {"k8s.pod.uid": "clbo-a-uid", "k8s.pod.name": "clbo-a", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__"}, "timestamp": "2025-01-10T10:02:00+00:00", "value": 0.5, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.cpu_request_utilization", "labels": {"k8s.pod.uid": "clbo-a-uid", "k8s.pod.name": "clbo-a", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__"}, "timestamp": "2025-01-10T10:04:00+00:00", "value": 0.5, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.cpu_limit_utilization", "labels": {"k8s.pod.uid": "clbo-a-uid", "k8s.pod.name": "clbo-a", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__"}, "timestamp": "2025-01-10T10:00:00+00:00", "value": 0.5, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.cpu_limit_utilization", "labels": {"k8s.pod.uid": "clbo-a-uid", "k8s.pod.name": "clbo-a", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__"}, "timestamp": "2025-01-10T10:02:00+00:00", "value": 0.5, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.cpu_limit_utilization", "labels": {"k8s.pod.uid": "clbo-a-uid", "k8s.pod.name": "clbo-a", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__"}, "timestamp": "2025-01-10T10:04:00+00:00", "value": 0.5, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.memory.working_set", "labels": {"k8s.pod.uid": "clbo-a-uid", "k8s.pod.name": "clbo-a", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__"}, "timestamp": "2025-01-10T10:00:00+00:00", "value": 100000000.0, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.memory.working_set", "labels": {"k8s.pod.uid": "clbo-a-uid", "k8s.pod.name": "clbo-a", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__"}, "timestamp": "2025-01-10T10:02:00+00:00", "value": 100000000.0, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.memory.working_set", "labels": {"k8s.pod.uid": "clbo-a-uid", "k8s.pod.name": "clbo-a", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__"}, "timestamp": "2025-01-10T10:04:00+00:00", "value": 100000000.0, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.memory_request_utilization", "labels": {"k8s.pod.uid": "clbo-a-uid", "k8s.pod.name": "clbo-a", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__"}, "timestamp": "2025-01-10T10:00:00+00:00", "value": 0.5, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.memory_request_utilization", "labels": {"k8s.pod.uid": "clbo-a-uid", "k8s.pod.name": "clbo-a", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__"}, "timestamp": "2025-01-10T10:02:00+00:00", "value": 0.5, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.memory_request_utilization", "labels": {"k8s.pod.uid": "clbo-a-uid", "k8s.pod.name": "clbo-a", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__"}, "timestamp": "2025-01-10T10:04:00+00:00", "value": 0.5, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.memory_limit_utilization", "labels": {"k8s.pod.uid": "clbo-a-uid", "k8s.pod.name": "clbo-a", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__"}, "timestamp": "2025-01-10T10:00:00+00:00", "value": 0.5, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.memory_limit_utilization", "labels": {"k8s.pod.uid": "clbo-a-uid", "k8s.pod.name": "clbo-a", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__"}, "timestamp": "2025-01-10T10:02:00+00:00", "value": 0.5, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.memory_limit_utilization", "labels": {"k8s.pod.uid": "clbo-a-uid", "k8s.pod.name": "clbo-a", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__"}, "timestamp": "2025-01-10T10:04:00+00:00", "value": 0.5, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.phase", "labels": {"k8s.pod.uid": "clbo-a-uid", "k8s.pod.name": "clbo-a", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__"}, "timestamp": "2025-01-10T10:00:00+00:00", "value": 2, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.phase", "labels": {"k8s.pod.uid": "clbo-a-uid", "k8s.pod.name": "clbo-a", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__"}, "timestamp": "2025-01-10T10:02:00+00:00", "value": 2, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.phase", "labels": {"k8s.pod.uid": "clbo-a-uid", "k8s.pod.name": "clbo-a", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__"}, "timestamp": "2025-01-10T10:04:00+00:00", "value": 2, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.status_reason", "labels": {"k8s.pod.uid": "clbo-a-uid", "k8s.pod.name": "clbo-a", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__"}, "timestamp": "2025-01-10T10:00:00+00:00", "value": 6, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.status_reason", "labels": {"k8s.pod.uid": "clbo-a-uid", "k8s.pod.name": "clbo-a", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__"}, "timestamp": "2025-01-10T10:02:00+00:00", "value": 6, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.status_reason", "labels": {"k8s.pod.uid": "clbo-a-uid", "k8s.pod.name": "clbo-a", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__"}, "timestamp": "2025-01-10T10:04:00+00:00", "value": 6, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.container.status.reason", "labels": {"k8s.pod.uid": "clbo-a-uid", "k8s.pod.name": "clbo-a", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__", "k8s.container.name": "app", "k8s.container.status.reason": "CrashLoopBackOff"}, "timestamp": "2025-01-10T10:00:00+00:00", "value": 1, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.container.status.reason", "labels": {"k8s.pod.uid": "clbo-a-uid", "k8s.pod.name": "clbo-a", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__", "k8s.container.name": "app", "k8s.container.status.reason": "CrashLoopBackOff"}, "timestamp": "2025-01-10T10:02:00+00:00", "value": 1, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.container.status.reason", "labels": {"k8s.pod.uid": "clbo-a-uid", "k8s.pod.name": "clbo-a", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__", "k8s.container.name": "app", "k8s.container.status.reason": "CrashLoopBackOff"}, "timestamp": "2025-01-10T10:04:00+00:00", "value": 1, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.container.restarts", "labels": {"k8s.pod.uid": "clbo-a-uid", "k8s.pod.name": "clbo-a", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__", "k8s.container.name": "app"}, "timestamp": "2025-01-10T10:00:00+00:00", "value": 5, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.container.restarts", "labels": {"k8s.pod.uid": "clbo-a-uid", "k8s.pod.name": "clbo-a", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__", "k8s.container.name": "app"}, "timestamp": "2025-01-10T10:02:00+00:00", "value": 5, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.container.restarts", "labels": {"k8s.pod.uid": "clbo-a-uid", "k8s.pod.name": "clbo-a", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__", "k8s.container.name": "app"}, "timestamp": "2025-01-10T10:04:00+00:00", "value": 5, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.cpu.usage", "labels": {"k8s.pod.uid": "clbo-b-uid", "k8s.pod.name": "clbo-b", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__"}, "timestamp": "2025-01-10T10:00:00+00:00", "value": 0.3, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.cpu.usage", "labels": {"k8s.pod.uid": "clbo-b-uid", "k8s.pod.name": "clbo-b", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__"}, "timestamp": "2025-01-10T10:02:00+00:00", "value": 0.3, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.cpu.usage", "labels": {"k8s.pod.uid": "clbo-b-uid", "k8s.pod.name": "clbo-b", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__"}, "timestamp": "2025-01-10T10:04:00+00:00", "value": 0.3, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.cpu_request_utilization", "labels": {"k8s.pod.uid": "clbo-b-uid", "k8s.pod.name": "clbo-b", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__"}, "timestamp": "2025-01-10T10:00:00+00:00", "value": 0.5, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.cpu_request_utilization", "labels": {"k8s.pod.uid": "clbo-b-uid", "k8s.pod.name": "clbo-b", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__"}, "timestamp": "2025-01-10T10:02:00+00:00", "value": 0.5, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.cpu_request_utilization", "labels": {"k8s.pod.uid": "clbo-b-uid", "k8s.pod.name": "clbo-b", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__"}, "timestamp": "2025-01-10T10:04:00+00:00", "value": 0.5, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.cpu_limit_utilization", "labels": {"k8s.pod.uid": "clbo-b-uid", "k8s.pod.name": "clbo-b", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__"}, "timestamp": "2025-01-10T10:00:00+00:00", "value": 0.5, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.cpu_limit_utilization", "labels": {"k8s.pod.uid": "clbo-b-uid", "k8s.pod.name": "clbo-b", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__"}, "timestamp": "2025-01-10T10:02:00+00:00", "value": 0.5, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.cpu_limit_utilization", "labels": {"k8s.pod.uid": "clbo-b-uid", "k8s.pod.name": "clbo-b", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__"}, "timestamp": "2025-01-10T10:04:00+00:00", "value": 0.5, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.memory.working_set", "labels": {"k8s.pod.uid": "clbo-b-uid", "k8s.pod.name": "clbo-b", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__"}, "timestamp": "2025-01-10T10:00:00+00:00", "value": 100000000.0, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.memory.working_set", "labels": {"k8s.pod.uid": "clbo-b-uid", "k8s.pod.name": "clbo-b", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__"}, "timestamp": "2025-01-10T10:02:00+00:00", "value": 100000000.0, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.memory.working_set", "labels": {"k8s.pod.uid": "clbo-b-uid", "k8s.pod.name": "clbo-b", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__"}, "timestamp": "2025-01-10T10:04:00+00:00", "value": 100000000.0, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.memory_request_utilization", "labels": {"k8s.pod.uid": "clbo-b-uid", "k8s.pod.name": "clbo-b", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__"}, "timestamp": "2025-01-10T10:00:00+00:00", "value": 0.5, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.memory_request_utilization", "labels": {"k8s.pod.uid": "clbo-b-uid", "k8s.pod.name": "clbo-b", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__"}, "timestamp": "2025-01-10T10:02:00+00:00", "value": 0.5, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.memory_request_utilization", "labels": {"k8s.pod.uid": "clbo-b-uid", "k8s.pod.name": "clbo-b", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__"}, "timestamp": "2025-01-10T10:04:00+00:00", "value": 0.5, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.memory_limit_utilization", "labels": {"k8s.pod.uid": "clbo-b-uid", "k8s.pod.name": "clbo-b", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__"}, "timestamp": "2025-01-10T10:00:00+00:00", "value": 0.5, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.memory_limit_utilization", "labels": {"k8s.pod.uid": "clbo-b-uid", "k8s.pod.name": "clbo-b", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__"}, "timestamp": "2025-01-10T10:02:00+00:00", "value": 0.5, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.memory_limit_utilization", "labels": {"k8s.pod.uid": "clbo-b-uid", "k8s.pod.name": "clbo-b", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__"}, "timestamp": "2025-01-10T10:04:00+00:00", "value": 0.5, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.phase", "labels": {"k8s.pod.uid": "clbo-b-uid", "k8s.pod.name": "clbo-b", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__"}, "timestamp": "2025-01-10T10:00:00+00:00", "value": 2, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.phase", "labels": {"k8s.pod.uid": "clbo-b-uid", "k8s.pod.name": "clbo-b", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__"}, "timestamp": "2025-01-10T10:02:00+00:00", "value": 2, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.phase", "labels": {"k8s.pod.uid": "clbo-b-uid", "k8s.pod.name": "clbo-b", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__"}, "timestamp": "2025-01-10T10:04:00+00:00", "value": 2, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.status_reason", "labels": {"k8s.pod.uid": "clbo-b-uid", "k8s.pod.name": "clbo-b", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__"}, "timestamp": "2025-01-10T10:00:00+00:00", "value": 6, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.status_reason", "labels": {"k8s.pod.uid": "clbo-b-uid", "k8s.pod.name": "clbo-b", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__"}, "timestamp": "2025-01-10T10:02:00+00:00", "value": 6, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.status_reason", "labels": {"k8s.pod.uid": "clbo-b-uid", "k8s.pod.name": "clbo-b", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__"}, "timestamp": "2025-01-10T10:04:00+00:00", "value": 6, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.container.status.reason", "labels": {"k8s.pod.uid": "clbo-b-uid", "k8s.pod.name": "clbo-b", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__", "k8s.container.name": "app", "k8s.container.status.reason": "CrashLoopBackOff"}, "timestamp": "2025-01-10T10:00:00+00:00", "value": 1, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.container.status.reason", "labels": {"k8s.pod.uid": "clbo-b-uid", "k8s.pod.name": "clbo-b", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__", "k8s.container.name": "app", "k8s.container.status.reason": "CrashLoopBackOff"}, "timestamp": "2025-01-10T10:02:00+00:00", "value": 1, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.container.status.reason", "labels": {"k8s.pod.uid": "clbo-b-uid", "k8s.pod.name": "clbo-b", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__", "k8s.container.name": "app", "k8s.container.status.reason": "CrashLoopBackOff"}, "timestamp": "2025-01-10T10:04:00+00:00", "value": 1, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.container.restarts", "labels": {"k8s.pod.uid": "clbo-b-uid", "k8s.pod.name": "clbo-b", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__", "k8s.container.name": "app"}, "timestamp": "2025-01-10T10:00:00+00:00", "value": 5, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.container.restarts", "labels": {"k8s.pod.uid": "clbo-b-uid", "k8s.pod.name": "clbo-b", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__", "k8s.container.name": "app"}, "timestamp": "2025-01-10T10:02:00+00:00", "value": 5, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.container.restarts", "labels": {"k8s.pod.uid": "clbo-b-uid", "k8s.pod.name": "clbo-b", "k8s.namespace.name": "ns-x", "k8s.node.name": "node-a", "k8s.deployment.name": "phase-test", "k8s.cluster.name": "cluster-x", "k8s.statefulset.name": "", "k8s.daemonset.name": "", "k8s.job.name": "", "k8s.cronjob.name": "", "k8s.pod.start_time": "__START_TIME__", "k8s.container.name": "app"}, "timestamp": "2025-01-10T10:04:00+00:00", "value": 5, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}diff --git a/tests/integration/testdata/inframonitoring/statefulsets_non_ss_pods.jsonl b/tests/integration/testdata/inframonitoring/statefulsets_non_ss_pods.jsonlindex f990d0c382..cea5568bcd 100644--- a/tests/integration/testdata/inframonitoring/statefulsets_non_ss_pods.jsonl+++ b/tests/integration/testdata/inframonitoring/statefulsets_non_ss_pods.jsonl@@ -67,3 +67,15 @@{"metric_name":"k8s.pod.phase","labels":{"k8s.pod.uid":"nd-dep-uid","k8s.pod.name":"nd-dep-pod","k8s.deployment.name":"nd-dep","k8s.namespace.name":"ns-nd","k8s.cluster.name":"cluster-x"},"timestamp":"2025-01-10T10:00:00+00:00","value":2,"temporality":"Unspecified","type_":"Gauge","is_monotonic":false}{"metric_name":"k8s.pod.phase","labels":{"k8s.pod.uid":"nd-dep-uid","k8s.pod.name":"nd-dep-pod","k8s.deployment.name":"nd-dep","k8s.namespace.name":"ns-nd","k8s.cluster.name":"cluster-x"},"timestamp":"2025-01-10T10:02:00+00:00","value":2,"temporality":"Unspecified","type_":"Gauge","is_monotonic":false}{"metric_name":"k8s.pod.phase","labels":{"k8s.pod.uid":"nd-dep-uid","k8s.pod.name":"nd-dep-pod","k8s.deployment.name":"nd-dep","k8s.namespace.name":"ns-nd","k8s.cluster.name":"cluster-x"},"timestamp":"2025-01-10T10:04:00+00:00","value":2,"temporality":"Unspecified","type_":"Gauge","is_monotonic":false}+{"metric_name": "k8s.pod.status_reason", "labels": {"k8s.pod.uid": "ns-ss-p1-uid", "k8s.pod.name": "ns-ss-p1", "k8s.statefulset.name": "ns-ss", "k8s.namespace.name": "ns-nd", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:00:00+00:00", "value": 6, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.phase", "labels": {"k8s.pod.uid": "ns-ss-p1-clbo-uid", "k8s.pod.name": "ns-ss-p1-clbo", "k8s.statefulset.name": "ns-ss", "k8s.namespace.name": "ns-nd", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:00:00+00:00", "value": 2, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.status_reason", "labels": {"k8s.pod.uid": "ns-ss-p1-clbo-uid", "k8s.pod.name": "ns-ss-p1-clbo", "k8s.statefulset.name": "ns-ss", "k8s.namespace.name": "ns-nd", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:00:00+00:00", "value": 6, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.container.status.reason", "labels": {"k8s.pod.uid": "ns-ss-p1-clbo-uid", "k8s.pod.name": "ns-ss-p1-clbo", "k8s.statefulset.name": "ns-ss", "k8s.namespace.name": "ns-nd", "k8s.cluster.name": "cluster-x", "k8s.container.name": "app", "k8s.container.status.reason": "CrashLoopBackOff"}, "timestamp": "2025-01-10T10:00:00+00:00", "value": 1, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.status_reason", "labels": {"k8s.pod.uid": "ns-ss-p1-uid", "k8s.pod.name": "ns-ss-p1", "k8s.statefulset.name": "ns-ss", "k8s.namespace.name": "ns-nd", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:02:00+00:00", "value": 6, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.phase", "labels": {"k8s.pod.uid": "ns-ss-p1-clbo-uid", "k8s.pod.name": "ns-ss-p1-clbo", "k8s.statefulset.name": "ns-ss", "k8s.namespace.name": "ns-nd", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:02:00+00:00", "value": 2, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.status_reason", "labels": {"k8s.pod.uid": "ns-ss-p1-clbo-uid", "k8s.pod.name": "ns-ss-p1-clbo", "k8s.statefulset.name": "ns-ss", "k8s.namespace.name": "ns-nd", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:02:00+00:00", "value": 6, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.container.status.reason", "labels": {"k8s.pod.uid": "ns-ss-p1-clbo-uid", "k8s.pod.name": "ns-ss-p1-clbo", "k8s.statefulset.name": "ns-ss", "k8s.namespace.name": "ns-nd", "k8s.cluster.name": "cluster-x", "k8s.container.name": "app", "k8s.container.status.reason": "CrashLoopBackOff"}, "timestamp": "2025-01-10T10:02:00+00:00", "value": 1, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.status_reason", "labels": {"k8s.pod.uid": "ns-ss-p1-uid", "k8s.pod.name": "ns-ss-p1", "k8s.statefulset.name": "ns-ss", "k8s.namespace.name": "ns-nd", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:04:00+00:00", "value": 6, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.phase", "labels": {"k8s.pod.uid": "ns-ss-p1-clbo-uid", "k8s.pod.name": "ns-ss-p1-clbo", "k8s.statefulset.name": "ns-ss", "k8s.namespace.name": "ns-nd", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:04:00+00:00", "value": 2, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.pod.status_reason", "labels": {"k8s.pod.uid": "ns-ss-p1-clbo-uid", "k8s.pod.name": "ns-ss-p1-clbo", "k8s.statefulset.name": "ns-ss", "k8s.namespace.name": "ns-nd", "k8s.cluster.name": "cluster-x"}, "timestamp": "2025-01-10T10:04:00+00:00", "value": 6, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}+{"metric_name": "k8s.container.status.reason", "labels": {"k8s.pod.uid": "ns-ss-p1-clbo-uid", "k8s.pod.name": "ns-ss-p1-clbo", "k8s.statefulset.name": "ns-ss", "k8s.namespace.name": "ns-nd", "k8s.cluster.name": "cluster-x", "k8s.container.name": "app", "k8s.container.status.reason": "CrashLoopBackOff"}, "timestamp": "2025-01-10T10:04:00+00:00", "value": 1, "temporality": "Unspecified", "type_": "Gauge", "is_monotonic": false}diff --git a/tests/integration/tests/inframonitoring/02_pods.py b/tests/integration/tests/inframonitoring/02_pods.pyindex d8ee2cd118..dae249b2fc 100644--- a/tests/integration/tests/inframonitoring/02_pods.py+++ b/tests/integration/tests/inframonitoring/02_pods.py@@ -8,7 +8,11 @@ import requestsfrom fixtures import typesfrom fixtures.auth import USER_ADMIN_EMAIL, USER_ADMIN_PASSWORDfrom fixtures.fs import get_testdata_file_path-from fixtures.inframonitoring import STATUS_BUCKETS, STATUS_TO_BUCKET+from fixtures.inframonitoring import (+ STATUS_BUCKETS,+ STATUS_TO_BUCKET,+ expected_status_counts,+)from fixtures.metrics import Metricsfrom fixtures.querier import compare_values, get_all_warnings@@ -451,6 +455,95 @@ def test_pods_pagination(assert set(seen_pods) == {f"page-p{i}" for i in range(1, K + 1)}+def test_pods_filter_pagination_and_ordering(+ signoz: types.SigNoz,+ create_user_admin: None, # pylint: disable=unused-argument+ get_token,+ insert_metrics,+) -> None:+ """filterByPodStatus composed with pagination + name ordering. The full-scope+ status keyset is resolved before slicing, so total reflects the full matched+ set and pages stay disjoint + complete on both the metric-ordering branch+ (paginateWithBackfill) and the name-ordering branch (PaginateMetadataByName).+ crashloopbackoff matches 4 pods in pods_phases.jsonl: clbo-a, clbo-b, run-p, unk-p."""+ now = datetime.now(tz=UTC).replace(microsecond=0)+ insert_metrics(+ _load_pods_metrics(+ "inframonitoring/pods_phases.jsonl",+ base_time=now - timedelta(minutes=4),+ )+ )+ token = get_token(USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD)+ start = int((now - timedelta(minutes=5)).timestamp() * 1000)+ end = int(now.timestamp() * 1000)+ matched = {"clbo-a", "clbo-b", "run-p", "unk-p"}++ # Metric-ordering branch (default cpu order): total is invariant across a paged+ # walk and the pages are disjoint + cover the full matched set.+ seen: list[str] = []+ totals: set[int] = set()+ for offset in (0, 2, 4):+ resp = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": start,+ "end": end,+ "limit": 2,+ "offset": offset,+ "filter": {"filterByPodStatus": ["crashloopbackoff"]},+ },+ timeout=5,+ )+ assert resp.status_code == HTTPStatus.OK, resp.text+ data = resp.json()["data"]+ totals.add(data["total"])+ assert len(data["records"]) == max(0, min(2, 4 - offset)), f"offset={offset}: {data['records']!r}"+ seen.extend(r["meta"]["k8s.pod.name"] for r in data["records"])+ assert totals == {4}, f"total not invariant under filter+pagination: {totals}"+ assert len(seen) == 4, f"pages overlapped: {seen}"+ assert set(seen) == matched++ # Name-ordering branch (orderBy k8s.pod.name asc, groupBy empty): the filtered+ # set is returned in name order; total is the full matched count.+ ordered = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": start,+ "end": end,+ "limit": 50,+ "filter": {"filterByPodStatus": ["crashloopbackoff"]},+ "orderBy": {"key": {"name": "k8s.pod.name"}, "direction": "asc"},+ },+ timeout=5,+ )+ assert ordered.status_code == HTTPStatus.OK, ordered.text+ odata = ordered.json()["data"]+ assert odata["total"] == 4+ assert [r["meta"]["k8s.pod.name"] for r in odata["records"]] == ["clbo-a", "clbo-b", "run-p", "unk-p"]++ # Second page of the name branch: PaginateMetadataByName slices the filtered set+ # correctly (offset past the first 2 matched -> the last 2), total unchanged.+ page2 = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": start,+ "end": end,+ "limit": 2,+ "offset": 2,+ "filter": {"filterByPodStatus": ["crashloopbackoff"]},+ "orderBy": {"key": {"name": "k8s.pod.name"}, "direction": "asc"},+ },+ timeout=5,+ )+ assert page2.status_code == HTTPStatus.OK, page2.text+ p2 = page2.json()["data"]+ assert p2["total"] == 4+ assert [r["meta"]["k8s.pod.name"] for r in p2["records"]] == ["run-p", "unk-p"]++# orderBy keys per pods_constants.go:42-48 (snake_case request keys, camelCase# response fields). k8s.pod.name sorts via the metadata-name branch# (PaginateMetadataByName) and is only allowed when groupBy is empty.@@ -553,6 +646,16 @@ def test_pods_orderby( # pylint: disable=too-many-arguments,too-many-positional"is only allowed when groupBy is empty",id="orderby_podname_with_groupby",),+ pytest.param(+ {"filter": {"filterByPodStatus": ["Bogus"]}},+ "invalid filter by pod status",+ id="filter_by_pod_status_invalid",+ ),+ pytest.param(+ {"filter": {"filterByPodStatus": ["running", "Bogus"]}},+ "invalid filter by pod status",+ id="filter_by_pod_status_invalid_member",+ ),],)def test_pods_validation_errors(@@ -658,6 +761,44 @@ def test_pods_status_list_mode(if other != bucket:assert rec["podCountsByStatus"][other] == 0, f"expected {other}=0 when status={expected_status}, got {rec['podCountsByStatus']}"+ # filterByPodStatus (secondary filter, applied after status is assigned):+ # a set containing the pod's status keeps it; a set without it filters it+ # out. Wire value is case-insensitive (valuer lowercases). Multi-select is+ # OR: [own_status, other] still keeps the pod.+ for variant in ([expected_status], [expected_status.upper()], [expected_status, "running"]):+ matched = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "limit": 50,+ "filter": {"expression": f"k8s.pod.name = '{pod_name}'", "filterByPodStatus": variant},+ },+ timeout=5,+ )+ assert matched.status_code == HTTPStatus.OK, matched.text+ mdata = matched.json()["data"]+ assert mdata["total"] == 1, f"filterByPodStatus={variant!r} should keep {pod_name}"+ assert mdata["records"][0]["podCountsByStatus"][bucket] == 1++ # None of the seeded pods resolves to Running or OOMKilled -> reliable+ # mismatches, as a single value and as an all-absent multi-select set.+ for fbps in (["running"], ["running", "oomKilled"]):+ filtered_out = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "limit": 50,+ "filter": {"expression": f"k8s.pod.name = '{pod_name}'", "filterByPodStatus": fbps},+ },+ timeout=5,+ )+ assert filtered_out.status_code == HTTPStatus.OK, filtered_out.text+ assert filtered_out.json()["data"]["total"] == 0, f"{pod_name} must be filtered out by filterByPodStatus={fbps!r}"+@pytest.mark.parametrize("pod_name,expected_restarts",@@ -838,6 +979,61 @@ def test_pods_status_grouped_mode()assert rec["podCountsByStatus"] == expected_counts+ # filterByPodStatus in grouped mode: the group is kept because >=1 pod+ # matches, and only the filtered buckets are populated (others zeroed).+ running = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "limit": 50,+ "groupBy": [{"name": "k8s.namespace.name", "fieldDataType": "string", "fieldContext": "resource"}],+ "filter": {"filterByPodStatus": ["running"]},+ },+ timeout=5,+ )+ assert running.status_code == HTTPStatus.OK, running.text+ rdata = running.json()["data"]+ assert rdata["total"] == 1+ assert rdata["records"][0]["podCountsByStatus"] == expected_status_counts(running=2)++ # Multi-select is OR: both requested buckets populated (union), others zeroed.+ multi = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "limit": 50,+ "groupBy": [{"name": "k8s.namespace.name", "fieldDataType": "string", "fieldContext": "resource"}],+ "filter": {"filterByPodStatus": ["running", "crashLoopBackOff"]},+ },+ timeout=5,+ )+ assert multi.status_code == HTTPStatus.OK, multi.text+ mdata = multi.json()["data"]+ assert mdata["total"] == 1+ assert mdata["records"][0]["podCountsByStatus"] == expected_status_counts(running=2, crashLoopBackOff=1)++ # A set fully absent from the group -> group dropped, empty page (single+ # and multi-select both).+ for fbps in (["oomKilled"], ["oomKilled", "unknown"]):+ absent = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "limit": 50,+ "groupBy": [{"name": "k8s.namespace.name", "fieldDataType": "string", "fieldContext": "resource"}],+ "filter": {"filterByPodStatus": fbps},+ },+ timeout=5,+ )+ assert absent.status_code == HTTPStatus.OK, absent.text+ assert absent.json()["data"]["total"] == 0, f"group must be dropped by filterByPodStatus={fbps!r}"+def test_pods_restarts_grouped_mode(signoz: types.SigNoz,@@ -927,3 +1123,25 @@ def test_pods_status_missing_metric_warning(for bucket in STATUS_BUCKETS:assert rec["podCountsByStatus"][bucket] == 0, f"expected {bucket}=0 when gated off, got {rec['podCountsByStatus']}"assert rec["podRestarts"] == -1++ # filterByPodStatus + missing status metric: the up-front gate surfaces the+ # warning and returns an empty page (Total 0) instead of silently filtering+ # everything out.+ filtered = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "limit": 50,+ "filter": {"expression": "k8s.pod.name = 'miss-p1'", "filterByPodStatus": ["running"]},+ },+ timeout=5,+ )+ assert filtered.status_code == HTTPStatus.OK, filtered.text+ fdata = filtered.json()["data"]+ assert fdata["total"] == 0+ assert fdata["records"] == []+ fwarn = fdata.get("warning") or {}+ fmsgs = ([fwarn["message"]] if fwarn.get("message") else []) + [w["message"] for w in fwarn.get("warnings", [])]+ assert any("Pod status could not be computed" in m for m in fmsgs), f"gate warning missing on filtered call: {fmsgs!r}"diff --git a/tests/integration/tests/inframonitoring/03_nodes.py b/tests/integration/tests/inframonitoring/03_nodes.pyindex 3b814e2a70..e05189e5ea 100644--- a/tests/integration/tests/inframonitoring/03_nodes.py+++ b/tests/integration/tests/inframonitoring/03_nodes.py@@ -8,6 +8,7 @@ import requestsfrom fixtures import typesfrom fixtures.auth import USER_ADMIN_EMAIL, USER_ADMIN_PASSWORDfrom fixtures.fs import get_testdata_file_path+from fixtures.inframonitoring import expected_status_countsfrom fixtures.metrics import Metricsfrom fixtures.querier import compare_values, get_all_warnings@@ -310,6 +311,115 @@ def test_nodes_filter_invalid(assert any(err_substr in e["message"] for e in body["error"]["errors"]), f"{err_substr!r} not surfaced: {body['error']['errors']!r}"+def test_nodes_filter_by_pod_status(+ signoz: types.SigNoz,+ create_user_admin: None, # pylint: disable=unused-argument+ get_token,+ insert_metrics,+) -> None:+ """filterByPodStatus on nodes: a node is kept when >=1 of its pods matches+ the requested display status, and podCountsByStatus reflects only that+ status (others 0); an absent status yields an empty page. Reuses+ clusters_pod_phases.jsonl (carries k8s.node.name + full status metrics):+ pp-node has running=3, crashLoopBackOff=1, error=1, evicted=1, pending=1."""+ now = datetime.now(tz=UTC).replace(microsecond=0)+ insert_metrics(+ Metrics.load_from_file(+ get_testdata_file_path("inframonitoring/clusters_pod_phases.jsonl"),+ base_time=now - timedelta(minutes=4),+ )+ )+ token = get_token(USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD)++ # Multi-select is OR -> the union of requested buckets (others zeroed).+ for fbps, expected in (+ (["running"], expected_status_counts(running=3)),+ (["CrashLoopBackOff"], expected_status_counts(crashLoopBackOff=1)),+ (["running", "CrashLoopBackOff"], expected_status_counts(running=3, crashLoopBackOff=1)),+ ):+ resp = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "limit": 50,+ "filter": {"filterByPodStatus": fbps},+ },+ timeout=5,+ )+ assert resp.status_code == HTTPStatus.OK, resp.text+ data = resp.json()["data"]+ assert data["total"] == 1+ rec = data["records"][0]+ assert rec["nodeName"] == "pp-node"+ assert rec["podCountsByStatus"] == expected++ # A set fully absent from the node -> empty page (single and multi).+ for fbps in (["completed"], ["completed", "oomKilled"]):+ absent = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "limit": 50,+ "filter": {"filterByPodStatus": fbps},+ },+ timeout=5,+ )+ assert absent.status_code == HTTPStatus.OK, absent.text+ assert absent.json()["data"]["total"] == 0, f"node must be dropped by filterByPodStatus={fbps!r}"++ # Combined filterByPodStatus + filterByNodeReadiness = AND. pp-node is Ready+ # and runs pods -> kept when both match; dropped if either side has no match.+ both_match = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "limit": 50,+ "filter": {"filterByPodStatus": ["running"], "filterByNodeReadiness": ["ready"]},+ },+ timeout=5,+ )+ assert both_match.status_code == HTTPStatus.OK, both_match.text+ bdata = both_match.json()["data"]+ assert bdata["total"] == 1+ assert bdata["records"][0]["nodeName"] == "pp-node"++ # readiness side fails (pp-node is not not_ready) -> dropped.+ readiness_fails = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "limit": 50,+ "filter": {"filterByPodStatus": ["running"], "filterByNodeReadiness": ["not_ready"]},+ },+ timeout=5,+ )+ assert readiness_fails.status_code == HTTPStatus.OK, readiness_fails.text+ assert readiness_fails.json()["data"]["total"] == 0++ # pod-status side fails (no completed pod) -> dropped.+ status_fails = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "limit": 50,+ "filter": {"filterByPodStatus": ["completed"], "filterByNodeReadiness": ["ready"]},+ },+ timeout=5,+ )+ assert status_fails.status_code == HTTPStatus.OK, status_fails.text+ assert status_fails.json()["data"]["total"] == 0++@pytest.mark.parametrize("node_name,expected_condition",[@@ -358,6 +468,56 @@ def test_nodes_condition_list_mode(else:assert rec["nodeCountsByReadiness"] == {"ready": 0, "notReady": 1}+ # filterByNodeReadiness (secondary filter): matching readiness keeps the node,+ # the opposite readiness filters it out.+ matched = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "limit": 50,+ "filter": {"expression": f"k8s.node.name = '{node_name}'", "filterByNodeReadiness": [expected_condition]},+ },+ timeout=5,+ )+ assert matched.status_code == HTTPStatus.OK, matched.text+ mdata = matched.json()["data"]+ assert mdata["total"] == 1+ assert mdata["records"][0]["nodeName"] == node_name++ opposite = "not_ready" if expected_condition == "ready" else "ready"++ # Multi-select is OR: a set containing the node's condition keeps it even+ # alongside the opposite readiness.+ or_keep = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "limit": 50,+ "filter": {"expression": f"k8s.node.name = '{node_name}'", "filterByNodeReadiness": [expected_condition, opposite]},+ },+ timeout=5,+ )+ assert or_keep.status_code == HTTPStatus.OK, or_keep.text+ assert or_keep.json()["data"]["total"] == 1, f"multi-select should keep {node_name}"++ dropped = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "limit": 50,+ "filter": {"expression": f"k8s.node.name = '{node_name}'", "filterByNodeReadiness": [opposite]},+ },+ timeout=5,+ )+ assert dropped.status_code == HTTPStatus.OK, dropped.text+ assert dropped.json()["data"]["total"] == 0+def test_nodes_condition_latest_wins(signoz: types.SigNoz,@@ -449,6 +609,60 @@ def test_nodes_condition_grouped_mode(# meta surfaces the groupBy key.assert rec["meta"].get("k8s.cluster.name") == "cluster-mixed"+ # filterByNodeReadiness in grouped mode: the group is kept (>=1 matching+ # node) and only the filtered bucket is populated.+ ready = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "limit": 50,+ "groupBy": [{"name": "k8s.cluster.name", "fieldDataType": "string", "fieldContext": "resource"}],+ "filter": {"filterByNodeReadiness": ["ready"]},+ },+ timeout=5,+ )+ assert ready.status_code == HTTPStatus.OK, ready.text+ rdata = ready.json()["data"]+ assert rdata["total"] == 1+ assert rdata["records"][0]["nodeCountsByReadiness"] == {"ready": 2, "notReady": 0}++ not_ready = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "limit": 50,+ "groupBy": [{"name": "k8s.cluster.name", "fieldDataType": "string", "fieldContext": "resource"}],+ "filter": {"filterByNodeReadiness": ["not_ready"]},+ },+ timeout=5,+ )+ assert not_ready.status_code == HTTPStatus.OK, not_ready.text+ ndata = not_ready.json()["data"]+ assert ndata["total"] == 1+ assert ndata["records"][0]["nodeCountsByReadiness"] == {"ready": 0, "notReady": 1}++ # Multi-select is OR: both buckets populated (union) for the kept group.+ both = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "limit": 50,+ "groupBy": [{"name": "k8s.cluster.name", "fieldDataType": "string", "fieldContext": "resource"}],+ "filter": {"filterByNodeReadiness": ["ready", "not_ready"]},+ },+ timeout=5,+ )+ assert both.status_code == HTTPStatus.OK, both.text+ bdata = both.json()["data"]+ assert bdata["total"] == 1+ assert bdata["records"][0]["nodeCountsByReadiness"] == {"ready": 2, "notReady": 1}+@pytest.mark.parametrize("group_key,expected",@@ -579,6 +793,95 @@ def test_nodes_pagination(assert set(seen_nodes) == {f"page-n{i}" for i in range(1, K + 1)}+def test_nodes_filter_readiness_pagination_and_ordering(+ signoz: types.SigNoz,+ create_user_admin: None, # pylint: disable=unused-argument+ get_token,+ insert_metrics,+) -> None:+ """filterByNodeReadiness composed with pagination + name ordering. The full-scope+ readiness keyset is resolved before slicing, so total reflects the full matched+ set and pages stay disjoint + complete on both the metric-ordering branch+ (paginateWithBackfill) and the name-ordering branch (PaginateMetadataByName).+ ready matches 4 nodes in nodes_conditions.jsonl: ready-n, ready-n2, ready-n3, ready-n4."""+ now = datetime.now(tz=UTC).replace(microsecond=0)+ insert_metrics(+ Metrics.load_from_file(+ get_testdata_file_path("inframonitoring/nodes_conditions.jsonl"),+ base_time=now - timedelta(minutes=4),+ )+ )+ token = get_token(USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD)+ start = int((now - timedelta(minutes=5)).timestamp() * 1000)+ end = int(now.timestamp() * 1000)+ matched = {"ready-n", "ready-n2", "ready-n3", "ready-n4"}++ # Metric-ordering branch (default cpu order): total is invariant across a paged+ # walk and the pages are disjoint + cover the full matched set.+ seen: list[str] = []+ totals: set[int] = set()+ for offset in (0, 2, 4):+ resp = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": start,+ "end": end,+ "limit": 2,+ "offset": offset,+ "filter": {"filterByNodeReadiness": ["ready"]},+ },+ timeout=5,+ )+ assert resp.status_code == HTTPStatus.OK, resp.text+ data = resp.json()["data"]+ totals.add(data["total"])+ assert len(data["records"]) == max(0, min(2, 4 - offset)), f"offset={offset}: {data['records']!r}"+ seen.extend(r["meta"]["k8s.node.name"] for r in data["records"])+ assert totals == {4}, f"total not invariant under filter+pagination: {totals}"+ assert len(seen) == 4, f"pages overlapped: {seen}"+ assert set(seen) == matched++ # Name-ordering branch (orderBy k8s.node.name asc, groupBy empty): the filtered+ # set is returned in name order; total is the full matched count.+ ordered = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": start,+ "end": end,+ "limit": 50,+ "filter": {"filterByNodeReadiness": ["ready"]},+ "orderBy": {"key": {"name": "k8s.node.name"}, "direction": "asc"},+ },+ timeout=5,+ )+ assert ordered.status_code == HTTPStatus.OK, ordered.text+ odata = ordered.json()["data"]+ assert odata["total"] == 4+ assert [r["meta"]["k8s.node.name"] for r in odata["records"]] == ["ready-n", "ready-n2", "ready-n3", "ready-n4"]++ # Second page of the name branch: PaginateMetadataByName slices the filtered set+ # correctly (offset past the first 2 matched -> the last 2), total unchanged.+ page2 = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": start,+ "end": end,+ "limit": 2,+ "offset": 2,+ "filter": {"filterByNodeReadiness": ["ready"]},+ "orderBy": {"key": {"name": "k8s.node.name"}, "direction": "asc"},+ },+ timeout=5,+ )+ assert page2.status_code == HTTPStatus.OK, page2.text+ p2 = page2.json()["data"]+ assert p2["total"] == 4+ assert [r["meta"]["k8s.node.name"] for r in p2["records"]] == ["ready-n3", "ready-n4"]++# orderBy keys per nodes_constants.go:33-37 (snake_case request keys,# camelCase response fields). k8s.node.name sorts via the metadata-name branch# (PaginateMetadataByName) and is only allowed when groupBy is empty.@@ -680,6 +983,21 @@ def test_nodes_orderby( # pylint: disable=too-many-arguments,too-many-positiona"is only allowed when groupBy is empty",id="orderby_nodename_with_groupby",),+ pytest.param(+ {"filter": {"filterByPodStatus": ["Bogus"]}},+ "invalid filter by pod status",+ id="filter_by_pod_status_invalid",+ ),+ pytest.param(+ {"filter": {"filterByNodeReadiness": ["bogus"]}},+ "invalid filter by node readiness",+ id="filter_by_node_readiness_invalid",+ ),+ pytest.param(+ {"filter": {"filterByNodeReadiness": ["notready"]}},+ "invalid filter by node readiness",+ id="filter_by_node_readiness_missing_underscore",+ ),],)def test_nodes_validation_errors(diff --git a/tests/integration/tests/inframonitoring/04_namespaces.py b/tests/integration/tests/inframonitoring/04_namespaces.pyindex 665b9ffd60..aba55237f1 100644--- a/tests/integration/tests/inframonitoring/04_namespaces.py+++ b/tests/integration/tests/inframonitoring/04_namespaces.py@@ -8,6 +8,7 @@ import requestsfrom fixtures import typesfrom fixtures.auth import USER_ADMIN_EMAIL, USER_ADMIN_PASSWORDfrom fixtures.fs import get_testdata_file_path+from fixtures.inframonitoring import expected_status_countsfrom fixtures.metrics import Metricsfrom fixtures.querier import compare_values, get_all_warnings@@ -307,6 +308,67 @@ def test_namespaces_filter_invalid(assert any(err_substr in e["message"] for e in body["error"]["errors"]), f"{err_substr!r} not surfaced: {body['error']['errors']!r}"+def test_namespaces_filter_by_pod_status(+ signoz: types.SigNoz,+ create_user_admin: None, # pylint: disable=unused-argument+ get_token,+ insert_metrics,+) -> None:+ """filterByPodStatus on namespaces: a namespace is kept when >=1 of its pods+ matches the requested display status, and podCountsByStatus reflects only+ that status (others 0); an absent status yields an empty page. Reuses+ clusters_pod_phases.jsonl (carries k8s.namespace.name + full status metrics):+ ns-x has running=3, crashLoopBackOff=1, error=1, evicted=1, pending=1."""+ now = datetime.now(tz=UTC).replace(microsecond=0)+ insert_metrics(+ Metrics.load_from_file(+ get_testdata_file_path("inframonitoring/clusters_pod_phases.jsonl"),+ base_time=now - timedelta(minutes=4),+ )+ )+ token = get_token(USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD)++ # Multi-select is OR -> the union of requested buckets (others zeroed).+ for fbps, expected in (+ (["running"], expected_status_counts(running=3)),+ (["CrashLoopBackOff"], expected_status_counts(crashLoopBackOff=1)),+ (["running", "CrashLoopBackOff"], expected_status_counts(running=3, crashLoopBackOff=1)),+ ):+ resp = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "limit": 50,+ "filter": {"filterByPodStatus": fbps},+ },+ timeout=5,+ )+ assert resp.status_code == HTTPStatus.OK, resp.text+ data = resp.json()["data"]+ assert data["total"] == 1+ rec = data["records"][0]+ assert rec["namespaceName"] == "ns-x"+ assert rec["podCountsByStatus"] == expected++ # A set fully absent from the namespace -> empty page (single and multi).+ for fbps in (["completed"], ["completed", "oomKilled"]):+ absent = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "limit": 50,+ "filter": {"filterByPodStatus": fbps},+ },+ timeout=5,+ )+ assert absent.status_code == HTTPStatus.OK, absent.text+ assert absent.json()["data"]["total"] == 0, f"namespace must be dropped by filterByPodStatus={fbps!r}"++# Float record fields compared with tolerance; everything else compared with ==._GROUPBY_FLOAT_FIELDS = {"namespaceCPU",@@ -621,6 +683,11 @@ def test_namespaces_orderby( # pylint: disable=too-many-arguments,too-many-posi"is only allowed when groupBy is empty",id="orderby_nsname_with_groupby",),+ pytest.param(+ {"filter": {"filterByPodStatus": ["Bogus"]}},+ "invalid filter by pod status",+ id="filter_by_pod_status_invalid",+ ),],)def test_namespaces_validation_errors(diff --git a/tests/integration/tests/inframonitoring/05_clusters.py b/tests/integration/tests/inframonitoring/05_clusters.pyindex d490cb5cc1..a300b6e961 100644--- a/tests/integration/tests/inframonitoring/05_clusters.py+++ b/tests/integration/tests/inframonitoring/05_clusters.py@@ -361,6 +361,51 @@ def test_clusters_node_readiness_aggregation(assert rec["clusterName"] == "rn-cluster"assert rec["nodeCountsByReadiness"] == {"ready": 3, "notReady": 2}+ # filterByNodeReadiness: cluster kept (>=1 matching node), only the filtered+ # bucket populated.+ ready = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "limit": 50,+ "filter": {"expression": "k8s.cluster.name = 'rn-cluster'", "filterByNodeReadiness": ["ready"]},+ },+ timeout=5,+ )+ assert ready.status_code == HTTPStatus.OK, ready.text+ assert ready.json()["data"]["records"][0]["nodeCountsByReadiness"] == {"ready": 3, "notReady": 0}++ not_ready = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "limit": 50,+ "filter": {"expression": "k8s.cluster.name = 'rn-cluster'", "filterByNodeReadiness": ["not_ready"]},+ },+ timeout=5,+ )+ assert not_ready.status_code == HTTPStatus.OK, not_ready.text+ assert not_ready.json()["data"]["records"][0]["nodeCountsByReadiness"] == {"ready": 0, "notReady": 2}++ # Multi-select is OR: both buckets populated (union) for the kept cluster.+ both = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "limit": 50,+ "filter": {"expression": "k8s.cluster.name = 'rn-cluster'", "filterByNodeReadiness": ["ready", "not_ready"]},+ },+ timeout=5,+ )+ assert both.status_code == HTTPStatus.OK, both.text+ assert both.json()["data"]["records"][0]["nodeCountsByReadiness"] == {"ready": 3, "notReady": 2}+def test_clusters_pod_status_aggregation(signoz: types.SigNoz,@@ -402,6 +447,95 @@ def test_clusters_pod_status_aggregation(# All status metrics present -> gate satisfied -> no status warning.assert all("Pod status could not be computed" not in w["message"] for w in get_all_warnings(response.json()))+ # filterByPodStatus: cluster kept (>=1 matching pod), only the filtered+ # buckets populated. Multi-select is OR -> the union of requested buckets.+ for fbps, expected in (+ (["running"], expected_status_counts(running=3)),+ (["CrashLoopBackOff"], expected_status_counts(crashLoopBackOff=1)),+ (["running", "CrashLoopBackOff"], expected_status_counts(running=3, crashLoopBackOff=1)),+ ):+ resp = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "limit": 50,+ "filter": {"expression": "k8s.cluster.name = 'pp-cluster'", "filterByPodStatus": fbps},+ },+ timeout=5,+ )+ assert resp.status_code == HTTPStatus.OK, resp.text+ fdata = resp.json()["data"]+ assert fdata["total"] == 1+ assert fdata["records"][0]["podCountsByStatus"] == expected++ # A set fully absent from the cluster -> empty page (single and multi).+ for fbps in (["completed"], ["completed", "oomKilled"]):+ absent = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "limit": 50,+ "filter": {"expression": "k8s.cluster.name = 'pp-cluster'", "filterByPodStatus": fbps},+ },+ timeout=5,+ )+ assert absent.status_code == HTTPStatus.OK, absent.text+ assert absent.json()["data"]["total"] == 0, f"cluster must be dropped by filterByPodStatus={fbps!r}"++ # Combined filterByPodStatus + filterByNodeReadiness = AND. pp-cluster has a+ # Ready node and Running pods -> kept when both match; dropped if either side+ # has no match.+ both_match = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "limit": 50,+ "filter": {"expression": "k8s.cluster.name = 'pp-cluster'", "filterByPodStatus": ["running"], "filterByNodeReadiness": ["ready"]},+ },+ timeout=5,+ )+ assert both_match.status_code == HTTPStatus.OK, both_match.text+ bdata = both_match.json()["data"]+ assert bdata["total"] == 1+ assert bdata["records"][0]["podCountsByStatus"] == expected_status_counts(running=3)+ assert bdata["records"][0]["nodeCountsByReadiness"] == {"ready": 1, "notReady": 0}++ # readiness side fails (no not_ready node) -> dropped.+ readiness_fails = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "limit": 50,+ "filter": {"expression": "k8s.cluster.name = 'pp-cluster'", "filterByPodStatus": ["running"], "filterByNodeReadiness": ["not_ready"]},+ },+ timeout=5,+ )+ assert readiness_fails.status_code == HTTPStatus.OK, readiness_fails.text+ assert readiness_fails.json()["data"]["total"] == 0++ # pod-status side fails (no completed pod) -> dropped.+ status_fails = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "limit": 50,+ "filter": {"expression": "k8s.cluster.name = 'pp-cluster'", "filterByPodStatus": ["completed"], "filterByNodeReadiness": ["ready"]},+ },+ timeout=5,+ )+ assert status_fails.status_code == HTTPStatus.OK, status_fails.text+ assert status_fails.json()["data"]["total"] == 0+@pytest.mark.parametrize("group_key,flt,expected",@@ -645,6 +779,21 @@ def test_clusters_orderby( # pylint: disable=too-many-arguments,too-many-positi"is only allowed when groupBy is empty",id="orderby_clustername_with_groupby",),+ pytest.param(+ {"filter": {"filterByPodStatus": ["Bogus"]}},+ "invalid filter by pod status",+ id="filter_by_pod_status_invalid",+ ),+ pytest.param(+ {"filter": {"filterByNodeReadiness": ["bogus"]}},+ "invalid filter by node readiness",+ id="filter_by_node_readiness_invalid",+ ),+ pytest.param(+ {"filter": {"filterByNodeReadiness": ["notready"]}},+ "invalid filter by node readiness",+ id="filter_by_node_readiness_missing_underscore",+ ),],)def test_clusters_validation_errors(diff --git a/tests/integration/tests/inframonitoring/07_deployments.py b/tests/integration/tests/inframonitoring/07_deployments.pyindex 40eaad3b85..69fbf446e3 100644--- a/tests/integration/tests/inframonitoring/07_deployments.py+++ b/tests/integration/tests/inframonitoring/07_deployments.py@@ -377,6 +377,45 @@ def test_deployments_pod_status_aggregation(# All status metrics present -> gate satisfied -> no status warning.assert all("Pod status could not be computed" not in w["message"] for w in get_all_warnings(response.json()))+ # filterByPodStatus: deployment kept (>=1 matching pod), only the filtered+ # buckets populated. Multi-select is OR -> the union of requested buckets.+ for fbps, expected in (+ (["running"], expected_status_counts(running=3)),+ (["CrashLoopBackOff"], expected_status_counts(crashLoopBackOff=1)),+ (["running", "CrashLoopBackOff"], expected_status_counts(running=3, crashLoopBackOff=1)),+ ):+ resp = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "limit": 50,+ "filter": {"expression": "k8s.deployment.name = 'pp-dep'", "filterByPodStatus": fbps},+ },+ timeout=5,+ )+ assert resp.status_code == HTTPStatus.OK, resp.text+ fdata = resp.json()["data"]+ assert fdata["total"] == 1+ assert fdata["records"][0]["podCountsByStatus"] == expected++ # A set fully absent from the deployment -> empty page (single and multi).+ for fbps in (["completed"], ["completed", "oomKilled"]):+ absent = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "limit": 50,+ "filter": {"expression": "k8s.deployment.name = 'pp-dep'", "filterByPodStatus": fbps},+ },+ timeout=5,+ )+ assert absent.status_code == HTTPStatus.OK, absent.text+ assert absent.json()["data"]["total"] == 0, f"deployment must be dropped by filterByPodStatus={fbps!r}"+def test_deployments_desired_available_counts(signoz: types.SigNoz,@@ -783,6 +822,11 @@ def test_deployments_orderby( # pylint: disable=too-many-arguments,too-many-pos"is only allowed when groupBy is empty",id="orderby_depname_with_groupby",),+ pytest.param(+ {"filter": {"filterByPodStatus": ["Bogus"]}},+ "invalid filter by pod status",+ id="filter_by_pod_status_invalid",+ ),],)def test_deployments_validation_errors(diff --git a/tests/integration/tests/inframonitoring/08_statefulsets.py b/tests/integration/tests/inframonitoring/08_statefulsets.pyindex 6d1c83ae65..bced763ae1 100644--- a/tests/integration/tests/inframonitoring/08_statefulsets.py+++ b/tests/integration/tests/inframonitoring/08_statefulsets.py@@ -8,6 +8,7 @@ import requestsfrom fixtures import typesfrom fixtures.auth import USER_ADMIN_EMAIL, USER_ADMIN_PASSWORDfrom fixtures.fs import get_testdata_file_path+from fixtures.inframonitoring import expected_status_countsfrom fixtures.metrics import Metricsfrom fixtures.querier import compare_values, get_all_warnings@@ -326,6 +327,77 @@ def test_statefulsets_base_filter_drops_non_statefulset_pods(# No empty-name group leaking through.assert all(r["statefulSetName"] != "" for r in data["records"])+ # filterByPodStatus: ns-ss (ns-ss-p1 Running + ns-ss-p1-clbo CrashLoopBackOff)+ # is kept when >=1 pod matches; counts reflect only the filtered status; an+ # absent status yields an empty page. Metric aggregation stays undistorted.+ unfiltered_cpu = rec["statefulSetCPU"]++ running = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "limit": 50,+ "filter": {"filterByPodStatus": ["running"]},+ },+ timeout=5,+ )+ assert running.status_code == HTTPStatus.OK, running.text+ rdata = running.json()["data"]+ assert rdata["total"] == 1+ rrec = rdata["records"][0]+ assert rrec["statefulSetName"] == "ns-ss"+ assert rrec["podCountsByStatus"] == expected_status_counts(running=1)+ assert compare_values(rrec["statefulSetCPU"], unfiltered_cpu, 1e-6), "filterByPodStatus distorted statefulSetCPU"++ clbo = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "limit": 50,+ "filter": {"filterByPodStatus": ["CrashLoopBackOff"]},+ },+ timeout=5,+ )+ assert clbo.status_code == HTTPStatus.OK, clbo.text+ cdata = clbo.json()["data"]+ assert cdata["total"] == 1+ assert cdata["records"][0]["podCountsByStatus"] == expected_status_counts(crashLoopBackOff=1)++ # Multi-select is OR: both requested buckets populated (union), others zeroed.+ multi = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "limit": 50,+ "filter": {"filterByPodStatus": ["running", "CrashLoopBackOff"]},+ },+ timeout=5,+ )+ assert multi.status_code == HTTPStatus.OK, multi.text+ assert multi.json()["data"]["records"][0]["podCountsByStatus"] == expected_status_counts(running=1, crashLoopBackOff=1)++ # A set fully absent from the group -> empty page (single and multi-select).+ for fbps in (["pending"], ["pending", "oomKilled"]):+ absent = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "limit": 50,+ "filter": {"filterByPodStatus": fbps},+ },+ timeout=5,+ )+ assert absent.status_code == HTTPStatus.OK, absent.text+ assert absent.json()["data"]["total"] == 0, f"group must be dropped by filterByPodStatus={fbps!r}"+# Float record fields compared with tolerance; everything else compared with ==._GROUPBY_FLOAT_FIELDS = {@@ -655,6 +727,11 @@ def test_statefulsets_orderby( # pylint: disable=too-many-arguments,too-many-po"is only allowed when groupBy is empty",id="orderby_ssname_with_groupby",),+ pytest.param(+ {"filter": {"filterByPodStatus": ["Bogus"]}},+ "invalid filter by pod status",+ id="filter_by_pod_status_invalid",+ ),],)def test_statefulsets_validation_errors(diff --git a/tests/integration/tests/inframonitoring/09_jobs.py b/tests/integration/tests/inframonitoring/09_jobs.pyindex 2bd8f86ced..5898830f21 100644--- a/tests/integration/tests/inframonitoring/09_jobs.py+++ b/tests/integration/tests/inframonitoring/09_jobs.py@@ -8,6 +8,7 @@ import requestsfrom fixtures import typesfrom fixtures.auth import USER_ADMIN_EMAIL, USER_ADMIN_PASSWORDfrom fixtures.fs import get_testdata_file_path+from fixtures.inframonitoring import expected_status_countsfrom fixtures.metrics import Metricsfrom fixtures.querier import compare_values, get_all_warnings@@ -458,6 +459,77 @@ def test_jobs_base_filter_drops_non_job_pods(assert rec["jobName"] == "nj-job"assert all(r["jobName"] != "" for r in data["records"])+ # filterByPodStatus: nj-job (nj-job-p1 Running + nj-job-p1-clbo CrashLoopBackOff)+ # is kept when >=1 pod matches; counts reflect only the filtered status; an+ # absent status yields an empty page. Metric aggregation stays undistorted.+ unfiltered_cpu = rec["jobCPU"]++ running = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "limit": 50,+ "filter": {"filterByPodStatus": ["running"]},+ },+ timeout=5,+ )+ assert running.status_code == HTTPStatus.OK, running.text+ rdata = running.json()["data"]+ assert rdata["total"] == 1+ rrec = rdata["records"][0]+ assert rrec["jobName"] == "nj-job"+ assert rrec["podCountsByStatus"] == expected_status_counts(running=1)+ assert compare_values(rrec["jobCPU"], unfiltered_cpu, 1e-6), "filterByPodStatus distorted jobCPU"++ clbo = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "limit": 50,+ "filter": {"filterByPodStatus": ["CrashLoopBackOff"]},+ },+ timeout=5,+ )+ assert clbo.status_code == HTTPStatus.OK, clbo.text+ cdata = clbo.json()["data"]+ assert cdata["total"] == 1+ assert cdata["records"][0]["podCountsByStatus"] == expected_status_counts(crashLoopBackOff=1)++ # Multi-select is OR: both requested buckets populated (union), others zeroed.+ multi = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "limit": 50,+ "filter": {"filterByPodStatus": ["running", "CrashLoopBackOff"]},+ },+ timeout=5,+ )+ assert multi.status_code == HTTPStatus.OK, multi.text+ assert multi.json()["data"]["records"][0]["podCountsByStatus"] == expected_status_counts(running=1, crashLoopBackOff=1)++ # A set fully absent from the group -> empty page (single and multi-select).+ for fbps in (["pending"], ["pending", "oomKilled"]):+ absent = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "limit": 50,+ "filter": {"filterByPodStatus": fbps},+ },+ timeout=5,+ )+ assert absent.status_code == HTTPStatus.OK, absent.text+ assert absent.json()["data"]["total"] == 0, f"group must be dropped by filterByPodStatus={fbps!r}"+# Float record fields compared with tolerance; everything else compared with ==._GROUPBY_FLOAT_FIELDS = {@@ -794,6 +866,11 @@ def test_jobs_orderby( # pylint: disable=too-many-arguments,too-many-positional"is only allowed when groupBy is empty",id="orderby_jobname_with_groupby",),+ pytest.param(+ {"filter": {"filterByPodStatus": ["Bogus"]}},+ "invalid filter by pod status",+ id="filter_by_pod_status_invalid",+ ),],)def test_jobs_validation_errors(diff --git a/tests/integration/tests/inframonitoring/10_daemonsets.py b/tests/integration/tests/inframonitoring/10_daemonsets.pyindex a2baa47dfe..4c13e02ab8 100644--- a/tests/integration/tests/inframonitoring/10_daemonsets.py+++ b/tests/integration/tests/inframonitoring/10_daemonsets.py@@ -8,6 +8,7 @@ import requestsfrom fixtures import typesfrom fixtures.auth import USER_ADMIN_EMAIL, USER_ADMIN_PASSWORDfrom fixtures.fs import get_testdata_file_path+from fixtures.inframonitoring import expected_status_countsfrom fixtures.metrics import Metricsfrom fixtures.querier import compare_values, get_all_warnings@@ -333,6 +334,77 @@ def test_daemonsets_base_filter_drops_non_daemonset_pods(assert rec["daemonSetName"] == "nd-ds"assert all(r["daemonSetName"] != "" for r in data["records"])+ # filterByPodStatus: nd-ds (nd-ds-p1 Running + nd-ds-p1-clbo CrashLoopBackOff)+ # is kept when >=1 pod matches; counts reflect only the filtered status; an+ # absent status yields an empty page. Metric aggregation stays undistorted.+ unfiltered_cpu = rec["daemonSetCPU"]++ running = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "limit": 50,+ "filter": {"filterByPodStatus": ["running"]},+ },+ timeout=5,+ )+ assert running.status_code == HTTPStatus.OK, running.text+ rdata = running.json()["data"]+ assert rdata["total"] == 1+ rrec = rdata["records"][0]+ assert rrec["daemonSetName"] == "nd-ds"+ assert rrec["podCountsByStatus"] == expected_status_counts(running=1)+ assert compare_values(rrec["daemonSetCPU"], unfiltered_cpu, 1e-6), "filterByPodStatus distorted daemonSetCPU"++ clbo = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "limit": 50,+ "filter": {"filterByPodStatus": ["CrashLoopBackOff"]},+ },+ timeout=5,+ )+ assert clbo.status_code == HTTPStatus.OK, clbo.text+ cdata = clbo.json()["data"]+ assert cdata["total"] == 1+ assert cdata["records"][0]["podCountsByStatus"] == expected_status_counts(crashLoopBackOff=1)++ # Multi-select is OR: both requested buckets populated (union), others zeroed.+ multi = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "limit": 50,+ "filter": {"filterByPodStatus": ["running", "CrashLoopBackOff"]},+ },+ timeout=5,+ )+ assert multi.status_code == HTTPStatus.OK, multi.text+ assert multi.json()["data"]["records"][0]["podCountsByStatus"] == expected_status_counts(running=1, crashLoopBackOff=1)++ # A set fully absent from the group -> empty page (single and multi-select).+ for fbps in (["pending"], ["pending", "oomKilled"]):+ absent = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "limit": 50,+ "filter": {"filterByPodStatus": fbps},+ },+ timeout=5,+ )+ assert absent.status_code == HTTPStatus.OK, absent.text+ assert absent.json()["data"]["total"] == 0, f"group must be dropped by filterByPodStatus={fbps!r}"+# Float record fields compared with tolerance; everything else compared with ==._GROUPBY_FLOAT_FIELDS = {@@ -663,6 +735,11 @@ def test_daemonsets_orderby( # pylint: disable=too-many-arguments,too-many-posi"is only allowed when groupBy is empty",id="orderby_dsname_with_groupby",),+ pytest.param(+ {"filter": {"filterByPodStatus": ["Bogus"]}},+ "invalid filter by pod status",+ id="filter_by_pod_status_invalid",+ ),],)def test_daemonsets_validation_errors(diff --git a/tests/integration/tests/inframonitoring/12_kube_containers.py b/tests/integration/tests/inframonitoring/12_kube_containers.pyindex fe606927b5..fa8d68a971 100644--- a/tests/integration/tests/inframonitoring/12_kube_containers.py+++ b/tests/integration/tests/inframonitoring/12_kube_containers.py@@ -120,6 +120,84 @@ def test_kube_containers_status_health_and_base_set(else:assert rec["cpu"] == -1, f"{pod}: expected cpu -1 sentinel (base-set-only), got {rec['cpu']}"+ # filterByContainerStatus (secondary filter): matching status keeps the+ # container; a mismatched status filters it out. Wire value is+ # case-insensitive (valuer lowercases): "running" == "Running".+ crun_running = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "limit": 50,+ "filter": {"expression": "k8s.pod.name = 'crun'", "filterByContainerStatus": ["running"]},+ },+ timeout=5,+ )+ assert crun_running.status_code == HTTPStatus.OK, crun_running.text+ cr = crun_running.json()["data"]+ assert cr["total"] == 1+ assert cr["records"][0]["meta"]["k8s.pod.name"] == "crun"++ cclo_clbo = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "limit": 50,+ "filter": {"expression": "k8s.pod.name = 'cclo'", "filterByContainerStatus": ["CrashLoopBackOff"]},+ },+ timeout=5,+ )+ assert cclo_clbo.status_code == HTTPStatus.OK, cclo_clbo.text+ assert cclo_clbo.json()["data"]["total"] == 1++ # crun is Running, not CrashLoopBackOff -> filtered out.+ crun_mismatch = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "limit": 50,+ "filter": {"expression": "k8s.pod.name = 'crun'", "filterByContainerStatus": ["CrashLoopBackOff"]},+ },+ timeout=5,+ )+ assert crun_mismatch.status_code == HTTPStatus.OK, crun_mismatch.text+ assert crun_mismatch.json()["data"]["total"] == 0++ # Multi-select is OR: a set containing the container's status keeps it; a set+ # with none of its statuses drops it.+ crun_or_keep = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "limit": 50,+ "filter": {"expression": "k8s.pod.name = 'crun'", "filterByContainerStatus": ["running", "CrashLoopBackOff"]},+ },+ timeout=5,+ )+ assert crun_or_keep.status_code == HTTPStatus.OK, crun_or_keep.text+ assert crun_or_keep.json()["data"]["total"] == 1++ crun_all_mismatch = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "limit": 50,+ "filter": {"expression": "k8s.pod.name = 'crun'", "filterByContainerStatus": ["CrashLoopBackOff", "OOMKilled"]},+ },+ timeout=5,+ )+ assert crun_all_mismatch.status_code == HTTPStatus.OK, crun_all_mismatch.text+ assert crun_all_mismatch.json()["data"]["total"] == 0+def test_kube_containers_status_counts_grouped_mode(signoz: types.SigNoz,@@ -165,6 +243,95 @@ def test_kube_containers_status_counts_grouped_mode(assert ns_b["oomKilled"] == 1assert by_ns["ns-b"]["containerCountsByReady"] == {"ready": 0, "notReady": 2}+ # filterByContainerStatus in grouped mode: keep only groups with a matching+ # container, only that bucket populated; a group with none is dropped.+ running = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "groupBy": [{"name": "k8s.namespace.name", "fieldDataType": "string", "fieldContext": "resource"}],+ "limit": 50,+ "filter": {"filterByContainerStatus": ["running"]},+ },+ timeout=5,+ )+ assert running.status_code == HTTPStatus.OK, running.text+ rdata = running.json()["data"]+ # ns-a has running containers, ns-b has none -> only ns-a.+ assert {r["meta"]["k8s.namespace.name"] for r in rdata["records"]} == {"ns-a"}+ assert rdata["records"][0]["containerCountsByStatus"]["running"] == 2++ oom = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "groupBy": [{"name": "k8s.namespace.name", "fieldDataType": "string", "fieldContext": "resource"}],+ "limit": 50,+ "filter": {"filterByContainerStatus": ["OOMKilled"]},+ },+ timeout=5,+ )+ assert oom.status_code == HTTPStatus.OK, oom.text+ odata = oom.json()["data"]+ assert {r["meta"]["k8s.namespace.name"] for r in odata["records"]} == {"ns-b"}+ assert odata["records"][0]["containerCountsByStatus"]["oomKilled"] == 1++ # A status absent from every group -> empty page.+ absent = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "groupBy": [{"name": "k8s.namespace.name", "fieldDataType": "string", "fieldContext": "resource"}],+ "limit": 50,+ "filter": {"filterByContainerStatus": ["Terminated"]},+ },+ timeout=5,+ )+ assert absent.status_code == HTTPStatus.OK, absent.text+ assert absent.json()["data"]["total"] == 0++ # Multi-select is OR: keeps every group with any matching container (union+ # across groups); a fully-absent set drops all.+ union = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "groupBy": [{"name": "k8s.namespace.name", "fieldDataType": "string", "fieldContext": "resource"}],+ "limit": 50,+ "filter": {"filterByContainerStatus": ["running", "OOMKilled"]},+ },+ timeout=5,+ )+ assert union.status_code == HTTPStatus.OK, union.text+ udata = union.json()["data"]+ u_by_ns = {r["meta"]["k8s.namespace.name"]: r for r in udata["records"]}+ assert set(u_by_ns) == {"ns-a", "ns-b"}+ assert u_by_ns["ns-a"]["containerCountsByStatus"]["running"] == 2+ assert u_by_ns["ns-b"]["containerCountsByStatus"]["oomKilled"] == 1++ absent_multi = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "groupBy": [{"name": "k8s.namespace.name", "fieldDataType": "string", "fieldContext": "resource"}],+ "limit": 50,+ "filter": {"filterByContainerStatus": ["Terminated", "Waiting"]},+ },+ timeout=5,+ )+ assert absent_multi.status_code == HTTPStatus.OK, absent_multi.text+ assert absent_multi.json()["data"]["total"] == 0+def test_kube_containers_status_recency(signoz: types.SigNoz,@@ -283,6 +450,31 @@ def test_kube_containers_status_warning_missing_metrics(warnings = get_all_warnings(body)assert any("status.state" in w["message"] and "status.reason" in w["message"] for w in warnings), f"status warning naming the missing metrics not surfaced: {warnings!r}"+ # filterByContainerStatus + missing status metrics: the up-front gate returns+ # the warning and an empty page (Total 0) rather than silently filtering+ # everything out.+ filtered = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": int((now - timedelta(minutes=5)).timestamp() * 1000),+ "end": int(now.timestamp() * 1000),+ "limit": 50,+ "filter": {"filterByContainerStatus": ["Running"]},+ },+ timeout=5,+ )+ assert filtered.status_code == HTTPStatus.OK, filtered.text+ fdata = filtered.json()["data"]+ assert fdata["total"] == 0+ assert fdata["records"] == []+ # The gate warning on the filtered path is surfaced at the top-level+ # warning.message (the up-front early return sets it directly); collect both+ # that and any nested warnings.+ fwarn = fdata.get("warning") or {}+ fmsgs = ([fwarn["message"]] if fwarn.get("message") else []) + [w["message"] for w in fwarn.get("warnings", [])]+ assert any("status.state" in m and "status.reason" in m for m in fmsgs), f"gate warning missing on filtered call: {fmsgs!r}"+def test_kube_containers_filter(signoz: types.SigNoz,@@ -374,6 +566,107 @@ def test_kube_containers_orderby_and_pagination(assert seen == order, f"paginated sequence {seen} != full-page order {order}"+def test_kube_containers_filter_pagination_and_ordering(+ signoz: types.SigNoz,+ create_user_admin: None, # pylint: disable=unused-argument+ get_token,+ insert_metrics,+) -> None:+ """filterByContainerStatus (multi-select) composed with pagination + name+ ordering. The full-scope status keyset is resolved before slicing, so total+ reflects the full matched set and pages stay disjoint + complete on both the+ metric-ordering branch (paginateWithBackfill -- including its metadata-only+ backfill arm, since coom has no cpu metric) and the name-ordering branch+ (PaginateMetadataByName). ['running','crashloopbackoff','oomkilled'] matches 4+ containers in kube_containers_dataset.jsonl: crun, cnr, cclo, coom.++ Name-branch order is not asserted: every container shares k8s.container.name+ ('app'), so the sort ties -- we assert the filtered set + disjoint/complete pages."""+ now = datetime.now(tz=UTC).replace(microsecond=0)+ insert_metrics(+ Metrics.load_from_file(+ get_testdata_file_path("inframonitoring/kube_containers_dataset.jsonl"),+ base_time=now - timedelta(minutes=4),+ )+ )+ token = get_token(USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD)+ start = int((now - timedelta(minutes=5)).timestamp() * 1000)+ end = int(now.timestamp() * 1000)+ statuses = ["running", "crashloopbackoff", "oomkilled"]+ matched = {"crun", "cnr", "cclo", "coom"}++ # Metric-ordering branch (default cpu order): total is invariant across a paged+ # walk and the pages are disjoint + cover the full matched set (coom arrives via+ # the metadata-only backfill arm). Non-matching containers never appear.+ seen: list[str] = []+ totals: set[int] = set()+ for offset in (0, 2, 4):+ resp = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": start,+ "end": end,+ "limit": 2,+ "offset": offset,+ "filter": {"filterByContainerStatus": statuses},+ },+ timeout=5,+ )+ assert resp.status_code == HTTPStatus.OK, resp.text+ data = resp.json()["data"]+ totals.add(data["total"])+ assert len(data["records"]) == max(0, min(2, 4 - offset)), f"offset={offset}: {data['records']!r}"+ seen.extend(r["meta"]["k8s.pod.name"] for r in data["records"])+ assert totals == {4}, f"total not invariant under filter+pagination: {totals}"+ assert len(seen) == 4, f"pages overlapped: {seen}"+ assert set(seen) == matched++ # Name-ordering branch (orderBy k8s.container.name asc, groupBy empty): the+ # filtered set is returned and total is the full matched count.+ ordered = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": start,+ "end": end,+ "limit": 50,+ "filter": {"filterByContainerStatus": statuses},+ "orderBy": {"key": {"name": "k8s.container.name"}, "direction": "asc"},+ },+ timeout=5,+ )+ assert ordered.status_code == HTTPStatus.OK, ordered.text+ odata = ordered.json()["data"]+ assert odata["total"] == 4+ assert {r["meta"]["k8s.pod.name"] for r in odata["records"]} == matched++ # Name branch paginated: PaginateMetadataByName slices the filtered set into+ # disjoint pages that together cover it, with total unchanged.+ name_seen: list[str] = []+ for offset in (0, 2):+ resp = requests.post(+ signoz.self.host_configs["8080"].get(ENDPOINT),+ headers={"authorization": f"Bearer {token}"},+ json={+ "start": start,+ "end": end,+ "limit": 2,+ "offset": offset,+ "filter": {"filterByContainerStatus": statuses},+ "orderBy": {"key": {"name": "k8s.container.name"}, "direction": "asc"},+ },+ timeout=5,+ )+ assert resp.status_code == HTTPStatus.OK, resp.text+ data = resp.json()["data"]+ assert data["total"] == 4+ assert len(data["records"]) == 2, f"offset={offset}: {data['records']!r}"+ name_seen.extend(r["meta"]["k8s.pod.name"] for r in data["records"])+ assert len(name_seen) == 4, f"name-branch pages overlapped: {name_seen}"+ assert set(name_seen) == matched++@pytest.mark.parametrize(("payload_override", "err_substr"),[@@ -398,6 +691,16 @@ def test_kube_containers_orderby_and_pagination("is only allowed when groupBy is empty",id="orderby_container_name_with_groupby",),+ pytest.param(+ {"filter": {"filterByContainerStatus": ["bogus"]}},+ "invalid filter by container status",+ id="filter_by_container_status_invalid",+ ),+ pytest.param(+ {"filter": {"filterByContainerStatus": ["no_data"]}},+ "invalid filter by container status",+ id="filter_by_container_status_no_data",+ ),],)def test_kube_containers_validation_errors(