Compare commits

..

3 Commits

Author SHA1 Message Date
nityanandagohain
3fee26e65b fix: skip resource filter if it exceeds a threshold 2026-06-01 17:30:55 +05:30
Nityananda Gohain
bfc50ee9c3 chore: use distinct fingerprint in resource filter (#11521)
* chore: use distinct fingerprint in resource filter

* fix: use group by instead of distinct
2026-06-01 09:47:56 +00:00
Tushar Vats
4b08ba1330 fix: qb warnings (#11518) 2026-06-01 09:43:22 +00:00
29 changed files with 1543 additions and 327 deletions

View File

@@ -52,6 +52,7 @@ jobs:
- rootuser
- serviceaccount
- querier_json_body
- querier_skip_resource_fingerprint
- ttl
sqlstore-provider:
- postgres

View File

@@ -870,6 +870,14 @@ components:
- timestampMillis
- data
type: object
CloudintegrationtypesAssets:
properties:
dashboards:
items:
$ref: '#/components/schemas/CloudintegrationtypesDashboard'
nullable: true
type: array
type: object
CloudintegrationtypesAzureAccountConfig:
properties:
deploymentRegion:
@@ -1017,6 +1025,17 @@ components:
- ingestionUrl
- ingestionKey
type: object
CloudintegrationtypesDashboard:
properties:
definition:
$ref: '#/components/schemas/DashboardtypesStorableDashboardData'
description:
type: string
id:
type: string
title:
type: string
type: object
CloudintegrationtypesDataCollected:
properties:
logs:
@@ -1190,7 +1209,7 @@ components:
CloudintegrationtypesService:
properties:
assets:
$ref: '#/components/schemas/CloudintegrationtypesServiceAssets'
$ref: '#/components/schemas/CloudintegrationtypesAssets'
cloudIntegrationService:
$ref: '#/components/schemas/CloudintegrationtypesCloudIntegrationService'
dataCollected:
@@ -1203,6 +1222,8 @@ components:
type: string
supportedSignals:
$ref: '#/components/schemas/CloudintegrationtypesSupportedSignals'
telemetryCollectionStrategy:
$ref: '#/components/schemas/CloudintegrationtypesTelemetryCollectionStrategy'
title:
type: string
required:
@@ -1213,17 +1234,9 @@ components:
- assets
- supportedSignals
- dataCollected
- telemetryCollectionStrategy
- cloudIntegrationService
type: object
CloudintegrationtypesServiceAssets:
properties:
dashboards:
items:
$ref: '#/components/schemas/CloudintegrationtypesServiceDashboard'
type: array
required:
- dashboards
type: object
CloudintegrationtypesServiceConfig:
properties:
aws:
@@ -1231,15 +1244,6 @@ components:
azure:
$ref: '#/components/schemas/CloudintegrationtypesAzureServiceConfig'
type: object
CloudintegrationtypesServiceDashboard:
properties:
description:
type: string
integrationDashboard:
$ref: '#/components/schemas/CloudintegrationtypesStorableIntegrationDashboard'
title:
type: string
type: object
CloudintegrationtypesServiceID:
enum:
- alb
@@ -1274,23 +1278,6 @@ components:
- icon
- enabled
type: object
CloudintegrationtypesStorableIntegrationDashboard:
properties:
createdAt:
format: date-time
type: string
dashboardId:
type: string
id:
type: string
provider:
type: string
slug:
type: string
updatedAt:
format: date-time
type: string
type: object
CloudintegrationtypesSupportedSignals:
properties:
logs:
@@ -1298,6 +1285,13 @@ components:
metrics:
type: boolean
type: object
CloudintegrationtypesTelemetryCollectionStrategy:
properties:
aws:
$ref: '#/components/schemas/CloudintegrationtypesAWSTelemetryCollectionStrategy'
azure:
$ref: '#/components/schemas/CloudintegrationtypesAzureTelemetryCollectionStrategy'
type: object
CloudintegrationtypesUpdatableAccount:
properties:
config:

View File

@@ -355,32 +355,26 @@ func (module *module) GetService(ctx context.Context, orgID valuer.UUID, service
var integrationService *cloudintegrationtypes.CloudIntegrationService
if cloudIntegrationID.IsZero() {
return cloudintegrationtypes.NewService(provider, serviceDefinition, nil, nil), nil
if !cloudIntegrationID.IsZero() {
storedService, err := module.store.GetServiceByServiceID(ctx, cloudIntegrationID, serviceID)
if err != nil && !errors.Ast(err, errors.TypeNotFound) {
return nil, err
}
if storedService != nil {
serviceConfig, err := cloudintegrationtypes.NewServiceConfigFromJSON(provider, storedService.Config)
if err != nil {
return nil, err
}
integrationService = cloudintegrationtypes.NewCloudIntegrationServiceFromStorable(storedService, serviceConfig)
}
if err := module.enrichDashboardIDs(ctx, orgID, provider, serviceID, serviceDefinition); err != nil {
return nil, err
}
}
storedService, err := module.store.GetServiceByServiceID(ctx, cloudIntegrationID, serviceID)
if err != nil && !errors.Ast(err, errors.TypeNotFound) {
return nil, err
}
if storedService == nil {
return cloudintegrationtypes.NewService(provider, serviceDefinition, nil, nil), nil
}
serviceConfig, err := cloudintegrationtypes.NewServiceConfigFromJSON(provider, storedService.Config)
if err != nil {
return nil, err
}
integrationService = cloudintegrationtypes.NewCloudIntegrationServiceFromStorable(storedService, serviceConfig)
slugPrefix := cloudintegrationtypes.CloudIntegrationDashboardSlugPrefix(provider, serviceID)
integrationDashboards, err := module.store.ListIntegrationDashboardsBySlugPrefix(ctx, orgID, cloudintegrationtypes.IntegrationDashboardProviderCloudIntegration, slugPrefix)
if err != nil {
return nil, err
}
return cloudintegrationtypes.NewService(provider, serviceDefinition, integrationService, integrationDashboards), nil
return cloudintegrationtypes.NewService(*serviceDefinition, integrationService), nil
}
func (module *module) CreateService(ctx context.Context, orgID valuer.UUID, createdBy string, creator valuer.UUID, service *cloudintegrationtypes.CloudIntegrationService, provider cloudintegrationtypes.CloudProviderType) error {
@@ -589,3 +583,20 @@ func (module *module) deprovisionDashboards(ctx context.Context, orgID valuer.UU
}
return nil
}
// enrichDashboardIDs replaces the raw dashboard name in each Dashboard.ID with the provisioned UUID.
// TODO: remove this hack and send idiomatic response to client.
func (module *module) enrichDashboardIDs(ctx context.Context, orgID valuer.UUID, provider cloudintegrationtypes.CloudProviderType, serviceID cloudintegrationtypes.ServiceID, serviceDefinition *cloudintegrationtypes.ServiceDefinition) error {
for i, d := range serviceDefinition.Assets.Dashboards {
slug := cloudintegrationtypes.CloudIntegrationDashboardSlug(provider, serviceID, d.ID)
row, err := module.store.GetIntegrationDashboardBySlug(ctx, orgID, cloudintegrationtypes.IntegrationDashboardProviderCloudIntegration, slug)
if err != nil {
if errors.Ast(err, errors.TypeNotFound) {
continue
}
return err
}
serviceDefinition.Assets.Dashboards[i].ID = row.DashboardID
}
return nil
}

View File

@@ -440,3 +440,4 @@ func (handler *handler) AgentCheckIn(rw http.ResponseWriter, r *http.Request) {
render.Success(rw, http.StatusOK, cloudintegrationtypes.NewGettableAgentCheckIn(provider, resp))
}

View File

@@ -679,21 +679,7 @@ func (bc *bucketCache) mergeAndDeduplicateBuckets(existing, fresh []*qbtypes.Cac
// deduplicateWarnings removes duplicate warnings.
func (bc *bucketCache) deduplicateWarnings(warnings []string) []string {
if len(warnings) == 0 {
return nil
}
seen := make(map[string]bool, len(warnings))
unique := make([]string, 0, len(warnings)) // Pre-allocate capacity
for _, warning := range warnings {
if !seen[warning] {
seen[warning] = true
unique = append(unique, warning)
}
}
return unique
return dedupeWarnings(warnings)
}
// trimResultToFluxBoundary trims the result to exclude data points beyond the flux boundary.

View File

@@ -7,6 +7,12 @@ import (
"github.com/SigNoz/signoz/pkg/factory"
)
type SkipResourceFingerprint struct {
Enabled bool `yaml:"enabled" mapstructure:"enabled"`
// If count of fingerprint is above threshold, skip the fingerprint subquery and filter on main table instead.
Threshold uint64 `yaml:"threshold" mapstructure:"threshold"`
}
// Config represents the configuration for the querier.
type Config struct {
// CacheTTL is the TTL for cached query results
@@ -15,6 +21,8 @@ type Config struct {
FluxInterval time.Duration `yaml:"flux_interval" mapstructure:"flux_interval"`
// MaxConcurrentQueries is the maximum number of concurrent queries for missing ranges
MaxConcurrentQueries int `yaml:"max_concurrent_queries" mapstructure:"max_concurrent_queries"`
// SkipResourceFingerprint configures when the resource fingerprint subquery is skipped in favor of main-table filtering.
SkipResourceFingerprint SkipResourceFingerprint `yaml:"skip_resource_fingerprint" mapstructure:"skip_resource_fingerprint"`
}
// NewConfigFactory creates a new config factory for querier.
@@ -28,6 +36,10 @@ func newConfig() factory.Config {
CacheTTL: 168 * time.Hour,
FluxInterval: 5 * time.Minute,
MaxConcurrentQueries: 4,
SkipResourceFingerprint: SkipResourceFingerprint{
Enabled: false,
Threshold: 100000,
},
}
}
@@ -42,6 +54,9 @@ func (c Config) Validate() error {
if c.MaxConcurrentQueries <= 0 {
return errors.NewInvalidInputf(errors.CodeInvalidInput, "max_concurrent_queries must be positive, got %v", c.MaxConcurrentQueries)
}
if c.SkipResourceFingerprint.Enabled && c.SkipResourceFingerprint.Threshold == 0 {
return errors.NewInvalidInputf(errors.CodeInvalidInput, "skip_resource_fingerprint.threshold must be > 0 when enabled")
}
return nil
}

View File

@@ -642,6 +642,12 @@ func (q *querier) run(
},
}
// Warnings can arrive duplicated: the bucket cache returns the cached
// portion's warnings alongside an identical warning emitted by every
// freshly-executed missing range (see mergeResults), and distinct queries
// can surface the same warning. Collapse exact duplicates before building
// the response.
warnings = dedupeWarnings(warnings)
if len(warnings) != 0 {
warns := make([]qbtypes.QueryWarnDataAdditional, len(warnings))
for i, warning := range warnings {
@@ -1125,6 +1131,8 @@ func (q *querier) adjustStepInterval(queries []qbtypes.QueryEnvelope, start, end
case qbtypes.QueryBuilderQuery[qbtypes.MetricAggregation]:
if qe.GetSource() == telemetrytypes.SourceMeter {
clampStep(qe, meterRecommended, meterMin, &warnings)
// we don't want to return warnings for meter metrics.
warnings = nil
} else {
clampStep(qe, metricRecommended, metricMin, &warnings)
}
@@ -1140,3 +1148,21 @@ func (q *querier) adjustStepInterval(queries []qbtypes.QueryEnvelope, start, end
}
return warnings
}
// dedupeWarnings removes exact-duplicate warning messages while preserving the
// order of first occurrence. Returns nil for an empty input. Warning counts are
// tiny (a handful per request), so a linear scan beats the allocation and
// hashing overhead of a map.
func dedupeWarnings(warnings []string) []string {
if len(warnings) == 0 {
return nil
}
unique := make([]string, 0, len(warnings))
// N^2 is faster than map-based deduping for small warning counts, and it preserves order of first occurrence without extra bookkeeping.
for _, warning := range warnings {
if !slices.Contains(unique, warning) {
unique = append(unique, warning)
}
}
return unique
}

View File

@@ -88,6 +88,8 @@ func newProvider(
traceAggExprRewriter,
telemetryStore,
flagger,
cfg.SkipResourceFingerprint.Enabled,
cfg.SkipResourceFingerprint.Threshold,
)
// Create trace operator statement builder
@@ -121,6 +123,9 @@ func newProvider(
telemetrylogs.DefaultFullTextColumn,
telemetrylogs.GetBodyJSONKey,
flagger,
telemetryStore,
cfg.SkipResourceFingerprint.Enabled,
cfg.SkipResourceFingerprint.Threshold,
)
// Create audit statement builder

View File

@@ -89,6 +89,9 @@ func prepareQuerierForLogs(t *testing.T, telemetryStore telemetrystore.Telemetry
telemetrylogs.DefaultFullTextColumn,
telemetrylogs.GetBodyJSONKey,
fl,
nil,
false,
100000,
)
return querier.New(
@@ -134,6 +137,8 @@ func prepareQuerierForTraces(t *testing.T, telemetryStore telemetrystore.Telemet
traceAggExprRewriter,
telemetryStore,
fl,
false,
100000,
)
return querier.New(

View File

@@ -127,7 +127,7 @@ func TestStatementBuilder(t *testing.T) {
Limit: 100,
},
expected: qbtypes.Statement{
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_audit.distributed_logs_resource WHERE ((simpleJSONExtractString(labels, 'signoz.audit.resource.kind') = ? AND labels LIKE ? AND labels LIKE ?) AND (simpleJSONExtractString(labels, 'signoz.audit.resource.id') = ? AND labels LIKE ? AND labels LIKE ?)) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?) SELECT timestamp, id, trace_id, span_id, trace_flags, severity_text, severity_number, scope_name, scope_version, body, event_name, attributes_string, attributes_number, attributes_bool, resource, scope_string FROM signoz_audit.distributed_logs WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND ts_bucket_start >= ? AND timestamp < ? AND ts_bucket_start <= ? LIMIT ?",
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_audit.distributed_logs_resource WHERE ((simpleJSONExtractString(labels, 'signoz.audit.resource.kind') = ? AND labels LIKE ? AND labels LIKE ?) AND (simpleJSONExtractString(labels, 'signoz.audit.resource.id') = ? AND labels LIKE ? AND labels LIKE ?)) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint) SELECT timestamp, id, trace_id, span_id, trace_flags, severity_text, severity_number, scope_name, scope_version, body, event_name, attributes_string, attributes_number, attributes_bool, resource, scope_string FROM signoz_audit.distributed_logs WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND ts_bucket_start >= ? AND timestamp < ? AND ts_bucket_start <= ? LIMIT ?",
Args: []any{"dashboard", "%signoz.audit.resource.kind%", "%signoz.audit.resource.kind\":\"dashboard%", "019b-5678-efgh-9012", "%signoz.audit.resource.id%", "%signoz.audit.resource.id\":\"019b-5678-efgh-9012%", uint64(1747945619), uint64(1747983448), "1747947419000000000", uint64(1747945619), "1747983448000000000", uint64(1747983448), 100},
},
},
@@ -144,7 +144,7 @@ func TestStatementBuilder(t *testing.T) {
Limit: 100,
},
expected: qbtypes.Statement{
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_audit.distributed_logs_resource WHERE (simpleJSONExtractString(labels, 'signoz.audit.resource.kind') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?) SELECT timestamp, id, trace_id, span_id, trace_flags, severity_text, severity_number, scope_name, scope_version, body, event_name, attributes_string, attributes_number, attributes_bool, resource, scope_string FROM signoz_audit.distributed_logs WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND (`attribute_string_signoz$$audit$$action` = ? AND `attribute_string_signoz$$audit$$action_exists` = ?) AND timestamp >= ? AND ts_bucket_start >= ? AND timestamp < ? AND ts_bucket_start <= ? LIMIT ?",
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_audit.distributed_logs_resource WHERE (simpleJSONExtractString(labels, 'signoz.audit.resource.kind') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint) SELECT timestamp, id, trace_id, span_id, trace_flags, severity_text, severity_number, scope_name, scope_version, body, event_name, attributes_string, attributes_number, attributes_bool, resource, scope_string FROM signoz_audit.distributed_logs WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND (`attribute_string_signoz$$audit$$action` = ? AND `attribute_string_signoz$$audit$$action_exists` = ?) AND timestamp >= ? AND ts_bucket_start >= ? AND timestamp < ? AND ts_bucket_start <= ? LIMIT ?",
Args: []any{"dashboard", "%signoz.audit.resource.kind%", "%signoz.audit.resource.kind\":\"dashboard%", uint64(1747945619), uint64(1747983448), "delete", true, "1747947419000000000", uint64(1747945619), "1747983448000000000", uint64(1747983448), 100},
},
},

View File

@@ -1205,6 +1205,9 @@ func buildJSONTestStatementBuilder(t *testing.T, addIndexes bool) (*logQueryStat
DefaultFullTextColumn,
GetBodyJSONKey,
fl,
nil,
false,
100000,
)
return statementBuilder, mockMetadataStore

View File

@@ -11,6 +11,7 @@ import (
"github.com/SigNoz/signoz/pkg/flagger"
"github.com/SigNoz/signoz/pkg/querybuilder"
"github.com/SigNoz/signoz/pkg/telemetryresourcefilter"
"github.com/SigNoz/signoz/pkg/telemetrystore"
"github.com/SigNoz/signoz/pkg/types/featuretypes"
qbtypes "github.com/SigNoz/signoz/pkg/types/querybuildertypes/querybuildertypesv5"
"github.com/SigNoz/signoz/pkg/types/telemetrytypes"
@@ -19,13 +20,14 @@ import (
)
type logQueryStatementBuilder struct {
logger *slog.Logger
metadataStore telemetrytypes.MetadataStore
fm qbtypes.FieldMapper
cb qbtypes.ConditionBuilder
resourceFilterStmtBuilder qbtypes.StatementBuilder[qbtypes.LogAggregation]
aggExprRewriter qbtypes.AggExprRewriter
fl flagger.Flagger
logger *slog.Logger
metadataStore telemetrytypes.MetadataStore
fm qbtypes.FieldMapper
cb qbtypes.ConditionBuilder
resourceFilterResolver *telemetryresourcefilter.ResourceFingerprintResolver[qbtypes.LogAggregation]
aggExprRewriter qbtypes.AggExprRewriter
fl flagger.Flagger
skipResourceFingerprintEnabled bool
fullTextColumn *telemetrytypes.TelemetryFieldKey
jsonKeyToKey qbtypes.JsonKeyToFieldFunc
@@ -42,10 +44,13 @@ func NewLogQueryStatementBuilder(
fullTextColumn *telemetrytypes.TelemetryFieldKey,
jsonKeyToKey qbtypes.JsonKeyToFieldFunc,
fl flagger.Flagger,
telemetryStore telemetrystore.TelemetryStore,
skipResourceFingerprintEnable bool,
skipResourceFingerprintThreshold uint64,
) *logQueryStatementBuilder {
logsSettings := factory.NewScopedProviderSettings(settings, "github.com/SigNoz/signoz/pkg/telemetrylogs")
resourceFilterStmtBuilder := telemetryresourcefilter.New[qbtypes.LogAggregation](
resourceFilterResolver := telemetryresourcefilter.NewResolver[qbtypes.LogAggregation](
settings,
DBName,
LogsResourceV2TableName,
@@ -55,18 +60,21 @@ func NewLogQueryStatementBuilder(
fullTextColumn,
jsonKeyToKey,
fl,
telemetryStore,
skipResourceFingerprintThreshold,
)
return &logQueryStatementBuilder{
logger: logsSettings.Logger(),
metadataStore: metadataStore,
fm: fieldMapper,
cb: conditionBuilder,
resourceFilterStmtBuilder: resourceFilterStmtBuilder,
aggExprRewriter: aggExprRewriter,
fl: fl,
fullTextColumn: fullTextColumn,
jsonKeyToKey: jsonKeyToKey,
logger: logsSettings.Logger(),
metadataStore: metadataStore,
fm: fieldMapper,
cb: conditionBuilder,
resourceFilterResolver: resourceFilterResolver,
aggExprRewriter: aggExprRewriter,
fl: fl,
skipResourceFingerprintEnabled: skipResourceFingerprintEnable,
fullTextColumn: fullTextColumn,
jsonKeyToKey: jsonKeyToKey,
}
}
@@ -271,9 +279,11 @@ func (b *logQueryStatementBuilder) buildListQuery(
bodyJSONEnabled = b.fl.BooleanOrEmpty(ctx, flagger.FeatureUseJSONBody, featuretypes.NewFlaggerEvaluationContext(valuer.UUID{}))
)
if frag, args, err := b.maybeAttachResourceFilter(ctx, sb, query, start, end, variables); err != nil {
frag, args, skipResourceFilter, err := b.maybeAttachResourceFilter(ctx, sb, query, start, end, variables)
if err != nil {
return nil, err
} else if frag != "" {
}
if frag != "" {
cteFragments = append(cteFragments, frag)
cteArgs = append(cteArgs, args)
}
@@ -315,7 +325,7 @@ func (b *logQueryStatementBuilder) buildListQuery(
sb.From(fmt.Sprintf("%s.%s", DBName, LogsV2TableName))
// Add filter conditions
preparedWhereClause, err := b.addFilterCondition(ctx, sb, start, end, query, keys, variables)
preparedWhereClause, err := b.addFilterCondition(ctx, sb, start, end, query, keys, variables, skipResourceFilter)
if err != nil {
return nil, err
@@ -373,9 +383,11 @@ func (b *logQueryStatementBuilder) buildTimeSeriesQuery(
bodyJSONEnabled = b.fl.BooleanOrEmpty(ctx, flagger.FeatureUseJSONBody, featuretypes.NewFlaggerEvaluationContext(valuer.UUID{}))
)
if frag, args, err := b.maybeAttachResourceFilter(ctx, sb, query, start, end, variables); err != nil {
frag, args, skipResourceFilter, err := b.maybeAttachResourceFilter(ctx, sb, query, start, end, variables)
if err != nil {
return nil, err
} else if frag != "" {
}
if frag != "" {
cteFragments = append(cteFragments, frag)
cteArgs = append(cteArgs, args)
}
@@ -419,7 +431,7 @@ func (b *logQueryStatementBuilder) buildTimeSeriesQuery(
// Add FROM clause
sb.From(fmt.Sprintf("%s.%s", DBName, LogsV2TableName))
preparedWhereClause, err := b.addFilterCondition(ctx, sb, start, end, query, keys, variables)
preparedWhereClause, err := b.addFilterCondition(ctx, sb, start, end, query, keys, variables, skipResourceFilter)
if err != nil {
return nil, err
@@ -531,9 +543,11 @@ func (b *logQueryStatementBuilder) buildScalarQuery(
bodyJSONEnabled = b.fl.BooleanOrEmpty(ctx, flagger.FeatureUseJSONBody, featuretypes.NewFlaggerEvaluationContext(valuer.UUID{}))
)
if frag, args, err := b.maybeAttachResourceFilter(ctx, sb, query, start, end, variables); err != nil {
frag, args, skipResourceFilter, err := b.maybeAttachResourceFilter(ctx, sb, query, start, end, variables)
if err != nil {
return nil, err
} else if frag != "" && !skipResourceCTE {
}
if frag != "" && !skipResourceCTE {
cteFragments = append(cteFragments, frag)
cteArgs = append(cteArgs, args)
}
@@ -576,7 +590,7 @@ func (b *logQueryStatementBuilder) buildScalarQuery(
sb.From(fmt.Sprintf("%s.%s", DBName, LogsV2TableName))
// Add filter conditions
preparedWhereClause, err := b.addFilterCondition(ctx, sb, start, end, query, keys, variables)
preparedWhereClause, err := b.addFilterCondition(ctx, sb, start, end, query, keys, variables, skipResourceFilter)
if err != nil {
return nil, err
@@ -640,6 +654,7 @@ func (b *logQueryStatementBuilder) addFilterCondition(
query qbtypes.QueryBuilderQuery[qbtypes.LogAggregation],
keys map[string][]*telemetrytypes.TelemetryFieldKey,
variables map[string]qbtypes.VariableItem,
skipResourceFilter bool,
) (querybuilder.PreparedWhereClause, error) {
var preparedWhereClause querybuilder.PreparedWhereClause
@@ -656,7 +671,7 @@ func (b *logQueryStatementBuilder) addFilterCondition(
ConditionBuilder: b.cb,
FieldKeys: keys,
BodyJSONEnabled: bodyJSONEnabled,
SkipResourceFilter: true,
SkipResourceFilter: skipResourceFilter,
FullTextColumn: b.fullTextColumn,
JsonKeyToKey: b.jsonKeyToKey,
Variables: variables,
@@ -707,33 +722,30 @@ func (b *logQueryStatementBuilder) maybeAttachResourceFilter(
query qbtypes.QueryBuilderQuery[qbtypes.LogAggregation],
start, end uint64,
variables map[string]qbtypes.VariableItem,
) (cteSQL string, cteArgs []any, err error) {
) (cteSQL string, cteArgs []any, skipResourceFilter bool, err error) {
stmt, err := b.buildResourceFilterCTE(ctx, query, start, end, variables)
if b.skipResourceFingerprintEnabled {
decision, err := b.resourceFilterResolver.Resolve(ctx, query, start, end, variables)
if err != nil {
return "", nil, true, err
}
switch decision {
case qbtypes.ResourceFilterResolveKindNoOp:
return "", nil, true, nil
case qbtypes.ResourceFilterResolveKindFallback:
return "", nil, false, nil
}
}
stmt, err := b.resourceFilterResolver.StatementBuilder().Build(
ctx, start, end, qbtypes.RequestTypeRaw, query, variables,
)
if err != nil {
return "", nil, err
return "", nil, true, err
}
if stmt == nil {
return "", nil, nil
return "", nil, true, nil
}
sb.Where("resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter)")
return fmt.Sprintf("__resource_filter AS (%s)", stmt.Query), stmt.Args, nil
}
func (b *logQueryStatementBuilder) buildResourceFilterCTE(
ctx context.Context,
query qbtypes.QueryBuilderQuery[qbtypes.LogAggregation],
start, end uint64,
variables map[string]qbtypes.VariableItem,
) (*qbtypes.Statement, error) {
return b.resourceFilterStmtBuilder.Build(
ctx,
start,
end,
qbtypes.RequestTypeRaw,
query,
variables,
)
return fmt.Sprintf("__resource_filter AS (%s)", stmt.Query), stmt.Args, true, nil
}

View File

@@ -2,19 +2,36 @@ package telemetrylogs
import (
"context"
"regexp"
"testing"
"time"
cmock "github.com/SigNoz/clickhouse-go-mock"
"github.com/SigNoz/signoz/pkg/errors"
"github.com/SigNoz/signoz/pkg/flagger/flaggertest"
"github.com/SigNoz/signoz/pkg/instrumentation/instrumentationtest"
"github.com/SigNoz/signoz/pkg/querybuilder"
"github.com/SigNoz/signoz/pkg/telemetrystore"
"github.com/SigNoz/signoz/pkg/telemetrystore/telemetrystoretest"
qbtypes "github.com/SigNoz/signoz/pkg/types/querybuildertypes/querybuildertypesv5"
"github.com/SigNoz/signoz/pkg/types/telemetrytypes"
"github.com/SigNoz/signoz/pkg/types/telemetrytypes/telemetrytypestest"
"github.com/stretchr/testify/require"
)
type regexQueryMatcher struct{}
func (m *regexQueryMatcher) Match(expectedSQL, actualSQL string) error {
re, err := regexp.Compile(expectedSQL)
if err != nil {
return err
}
if !re.MatchString(actualSQL) {
return errors.NewInvalidInputf(errors.CodeInvalidInput, "expected query to match %s, got %s", expectedSQL, actualSQL)
}
return nil
}
func TestStatementBuilderTimeSeries(t *testing.T) {
// Create a test release time
releaseTime := time.Date(2024, 1, 15, 10, 0, 0, 0, time.UTC)
@@ -55,7 +72,7 @@ func TestStatementBuilderTimeSeries(t *testing.T) {
},
},
expected: qbtypes.Statement{
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_logs.distributed_logs_v2_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), __limit_cte AS (SELECT toString(multiIf(resource.`service.name`::String IS NOT NULL, resource.`service.name`::String, NULL)) AS `service.name`, countDistinct(multiIf(resource.`service.name`::String IS NOT NULL, resource.`service.name`::String, NULL)) AS __result_0 FROM signoz_logs.distributed_logs_v2 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND ts_bucket_start >= ? AND timestamp < ? AND ts_bucket_start <= ? GROUP BY `service.name` ORDER BY __result_0 DESC LIMIT ?) SELECT toStartOfInterval(fromUnixTimestamp64Nano(timestamp), INTERVAL 30 SECOND) AS ts, toString(multiIf(resource.`service.name`::String IS NOT NULL, resource.`service.name`::String, NULL)) AS `service.name`, countDistinct(multiIf(resource.`service.name`::String IS NOT NULL, resource.`service.name`::String, NULL)) AS __result_0 FROM signoz_logs.distributed_logs_v2 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND ts_bucket_start >= ? AND timestamp < ? AND ts_bucket_start <= ? AND (`service.name`) GLOBAL IN (SELECT `service.name` FROM __limit_cte) GROUP BY ts, `service.name`",
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_logs.distributed_logs_v2_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint), __limit_cte AS (SELECT toString(multiIf(resource.`service.name`::String IS NOT NULL, resource.`service.name`::String, NULL)) AS `service.name`, countDistinct(multiIf(resource.`service.name`::String IS NOT NULL, resource.`service.name`::String, NULL)) AS __result_0 FROM signoz_logs.distributed_logs_v2 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND ts_bucket_start >= ? AND timestamp < ? AND ts_bucket_start <= ? GROUP BY `service.name` ORDER BY __result_0 DESC LIMIT ?) SELECT toStartOfInterval(fromUnixTimestamp64Nano(timestamp), INTERVAL 30 SECOND) AS ts, toString(multiIf(resource.`service.name`::String IS NOT NULL, resource.`service.name`::String, NULL)) AS `service.name`, countDistinct(multiIf(resource.`service.name`::String IS NOT NULL, resource.`service.name`::String, NULL)) AS __result_0 FROM signoz_logs.distributed_logs_v2 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND ts_bucket_start >= ? AND timestamp < ? AND ts_bucket_start <= ? AND (`service.name`) GLOBAL IN (SELECT `service.name` FROM __limit_cte) GROUP BY ts, `service.name`",
Args: []any{"cartservice", "%service.name%", "%service.name\":\"cartservice%", uint64(1705397400), uint64(1705485600), "1705399200000000000", uint64(1705397400), "1705485600000000000", uint64(1705485600), 10, "1705399200000000000", uint64(1705397400), "1705485600000000000", uint64(1705485600)},
},
expectedErr: nil,
@@ -127,7 +144,7 @@ func TestStatementBuilderTimeSeries(t *testing.T) {
},
},
expected: qbtypes.Statement{
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_logs.distributed_logs_v2_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), __limit_cte AS (SELECT toString(multiIf(resource.`service.name`::String IS NOT NULL, resource.`service.name`::String, NULL)) AS `service.name`, count() AS __result_0 FROM signoz_logs.distributed_logs_v2 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND ts_bucket_start >= ? AND timestamp < ? AND ts_bucket_start <= ? GROUP BY `service.name` ORDER BY `service.name` desc LIMIT ?) SELECT toStartOfInterval(fromUnixTimestamp64Nano(timestamp), INTERVAL 30 SECOND) AS ts, toString(multiIf(resource.`service.name`::String IS NOT NULL, resource.`service.name`::String, NULL)) AS `service.name`, count() AS __result_0 FROM signoz_logs.distributed_logs_v2 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND ts_bucket_start >= ? AND timestamp < ? AND ts_bucket_start <= ? AND (`service.name`) GLOBAL IN (SELECT `service.name` FROM __limit_cte) GROUP BY ts, `service.name` ORDER BY `service.name` desc, ts desc",
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_logs.distributed_logs_v2_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint), __limit_cte AS (SELECT toString(multiIf(resource.`service.name`::String IS NOT NULL, resource.`service.name`::String, NULL)) AS `service.name`, count() AS __result_0 FROM signoz_logs.distributed_logs_v2 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND ts_bucket_start >= ? AND timestamp < ? AND ts_bucket_start <= ? GROUP BY `service.name` ORDER BY `service.name` desc LIMIT ?) SELECT toStartOfInterval(fromUnixTimestamp64Nano(timestamp), INTERVAL 30 SECOND) AS ts, toString(multiIf(resource.`service.name`::String IS NOT NULL, resource.`service.name`::String, NULL)) AS `service.name`, count() AS __result_0 FROM signoz_logs.distributed_logs_v2 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND ts_bucket_start >= ? AND timestamp < ? AND ts_bucket_start <= ? AND (`service.name`) GLOBAL IN (SELECT `service.name` FROM __limit_cte) GROUP BY ts, `service.name` ORDER BY `service.name` desc, ts desc",
Args: []any{"cartservice", "%service.name%", "%service.name\":\"cartservice%", uint64(1705397400), uint64(1705485600), "1705399200000000000", uint64(1705397400), "1705485600000000000", uint64(1705485600), 10, "1705399200000000000", uint64(1705397400), "1705485600000000000", uint64(1705485600)},
},
expectedErr: nil,
@@ -160,7 +177,7 @@ func TestStatementBuilderTimeSeries(t *testing.T) {
},
},
expected: qbtypes.Statement{
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_logs.distributed_logs_v2_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), __limit_cte AS (SELECT toString(multiIf(`attribute_string_materialized$$key$$name_exists` = ?, `attribute_string_materialized$$key$$name`, NULL)) AS `materialized.key.name`, count() AS __result_0 FROM signoz_logs.distributed_logs_v2 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND ts_bucket_start >= ? AND timestamp < ? AND ts_bucket_start <= ? GROUP BY `materialized.key.name` ORDER BY __result_0 DESC LIMIT ?) SELECT toStartOfInterval(fromUnixTimestamp64Nano(timestamp), INTERVAL 30 SECOND) AS ts, toString(multiIf(`attribute_string_materialized$$key$$name_exists` = ?, `attribute_string_materialized$$key$$name`, NULL)) AS `materialized.key.name`, count() AS __result_0 FROM signoz_logs.distributed_logs_v2 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND ts_bucket_start >= ? AND timestamp < ? AND ts_bucket_start <= ? AND (`materialized.key.name`) GLOBAL IN (SELECT `materialized.key.name` FROM __limit_cte) GROUP BY ts, `materialized.key.name`",
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_logs.distributed_logs_v2_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint), __limit_cte AS (SELECT toString(multiIf(`attribute_string_materialized$$key$$name_exists` = ?, `attribute_string_materialized$$key$$name`, NULL)) AS `materialized.key.name`, count() AS __result_0 FROM signoz_logs.distributed_logs_v2 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND ts_bucket_start >= ? AND timestamp < ? AND ts_bucket_start <= ? GROUP BY `materialized.key.name` ORDER BY __result_0 DESC LIMIT ?) SELECT toStartOfInterval(fromUnixTimestamp64Nano(timestamp), INTERVAL 30 SECOND) AS ts, toString(multiIf(`attribute_string_materialized$$key$$name_exists` = ?, `attribute_string_materialized$$key$$name`, NULL)) AS `materialized.key.name`, count() AS __result_0 FROM signoz_logs.distributed_logs_v2 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND ts_bucket_start >= ? AND timestamp < ? AND ts_bucket_start <= ? AND (`materialized.key.name`) GLOBAL IN (SELECT `materialized.key.name` FROM __limit_cte) GROUP BY ts, `materialized.key.name`",
Args: []any{"cartservice", "%service.name%", "%service.name\":\"cartservice%", uint64(1705397400), uint64(1705485600), true, "1705399200000000000", uint64(1705397400), "1705485600000000000", uint64(1705485600), 10, true, "1705399200000000000", uint64(1705397400), "1705485600000000000", uint64(1705485600)},
},
},
@@ -212,6 +229,9 @@ func TestStatementBuilderTimeSeries(t *testing.T) {
DefaultFullTextColumn,
GetBodyJSONKey,
fl,
nil,
false,
100000,
)
for _, c := range cases {
@@ -253,7 +273,7 @@ func TestStatementBuilderListQuery(t *testing.T) {
Limit: 10,
},
expected: qbtypes.Statement{
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_logs.distributed_logs_v2_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?) SELECT timestamp, id, trace_id, span_id, trace_flags, severity_text, severity_number, scope_name, scope_version, body, attributes_string, attributes_number, attributes_bool, resources_string, scope_string FROM signoz_logs.distributed_logs_v2 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND ts_bucket_start >= ? AND timestamp < ? AND ts_bucket_start <= ? LIMIT ?",
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_logs.distributed_logs_v2_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint) SELECT timestamp, id, trace_id, span_id, trace_flags, severity_text, severity_number, scope_name, scope_version, body, attributes_string, attributes_number, attributes_bool, resources_string, scope_string FROM signoz_logs.distributed_logs_v2 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND ts_bucket_start >= ? AND timestamp < ? AND ts_bucket_start <= ? LIMIT ?",
Args: []any{"cartservice", "%service.name%", "%service.name\":\"cartservice%", uint64(1747945619), uint64(1747983448), "1747947419000000000", uint64(1747945619), "1747983448000000000", uint64(1747983448), 10},
},
expectedErr: nil,
@@ -281,7 +301,7 @@ func TestStatementBuilderListQuery(t *testing.T) {
},
},
expected: qbtypes.Statement{
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_logs.distributed_logs_v2_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?) SELECT timestamp, id, trace_id, span_id, trace_flags, severity_text, severity_number, scope_name, scope_version, body, attributes_string, attributes_number, attributes_bool, resources_string, scope_string FROM signoz_logs.distributed_logs_v2 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND ts_bucket_start >= ? AND timestamp < ? AND ts_bucket_start <= ? ORDER BY `attribute_string_materialized$$key$$name` AS `materialized.key.name` desc LIMIT ?",
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_logs.distributed_logs_v2_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint) SELECT timestamp, id, trace_id, span_id, trace_flags, severity_text, severity_number, scope_name, scope_version, body, attributes_string, attributes_number, attributes_bool, resources_string, scope_string FROM signoz_logs.distributed_logs_v2 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND ts_bucket_start >= ? AND timestamp < ? AND ts_bucket_start <= ? ORDER BY `attribute_string_materialized$$key$$name` AS `materialized.key.name` desc LIMIT ?",
Args: []any{"cartservice", "%service.name%", "%service.name\":\"cartservice%", uint64(1747945619), uint64(1747983448), "1747947419000000000", uint64(1747945619), "1747983448000000000", uint64(1747983448), 10},
},
expectedErr: nil,
@@ -353,6 +373,9 @@ func TestStatementBuilderListQuery(t *testing.T) {
DefaultFullTextColumn,
GetBodyJSONKey,
fl,
nil,
false,
100000,
)
for _, c := range cases {
@@ -424,7 +447,7 @@ func TestStatementBuilderListQueryResourceTests(t *testing.T) {
},
},
expected: qbtypes.Statement{
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_logs.distributed_logs_v2_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?) SELECT timestamp, id, trace_id, span_id, trace_flags, severity_text, severity_number, scope_name, scope_version, body, attributes_string, attributes_number, attributes_bool, resources_string, scope_string FROM signoz_logs.distributed_logs_v2 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND match(LOWER(body), LOWER(?)) AND timestamp >= ? AND ts_bucket_start >= ? AND timestamp < ? AND ts_bucket_start <= ? ORDER BY `attribute_string_materialized$$key$$name` AS `materialized.key.name` desc LIMIT ?",
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_logs.distributed_logs_v2_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint) SELECT timestamp, id, trace_id, span_id, trace_flags, severity_text, severity_number, scope_name, scope_version, body, attributes_string, attributes_number, attributes_bool, resources_string, scope_string FROM signoz_logs.distributed_logs_v2 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND match(LOWER(body), LOWER(?)) AND timestamp >= ? AND ts_bucket_start >= ? AND timestamp < ? AND ts_bucket_start <= ? ORDER BY `attribute_string_materialized$$key$$name` AS `materialized.key.name` desc LIMIT ?",
Args: []any{"cartservice", "%service.name%", "%service.name\":\"cartservice%", uint64(1747945619), uint64(1747983448), "hello", "1747947419000000000", uint64(1747945619), "1747983448000000000", uint64(1747983448), 10},
},
expectedErr: nil,
@@ -499,6 +522,9 @@ func TestStatementBuilderListQueryResourceTests(t *testing.T) {
DefaultFullTextColumn,
GetBodyJSONKey,
fl,
nil,
false,
100000,
)
for _, c := range cases {
@@ -575,6 +601,9 @@ func TestStatementBuilderTimeSeriesBodyGroupBy(t *testing.T) {
DefaultFullTextColumn,
GetBodyJSONKey,
fl,
nil,
false,
100000,
)
for _, c := range cases {
@@ -615,7 +644,7 @@ func TestStatementBuilderListQueryServiceCollision(t *testing.T) {
Limit: 10,
},
expected: qbtypes.Statement{
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_logs.distributed_logs_v2_resource WHERE ((simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?)) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?) SELECT timestamp, id, trace_id, span_id, trace_flags, severity_text, severity_number, scope_name, scope_version, body, attributes_string, attributes_number, attributes_bool, resources_string, scope_string FROM signoz_logs.distributed_logs_v2 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND (LOWER(body) LIKE LOWER(?)) AND timestamp >= ? AND ts_bucket_start >= ? AND timestamp < ? AND ts_bucket_start <= ? LIMIT ?",
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_logs.distributed_logs_v2_resource WHERE ((simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?)) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint) SELECT timestamp, id, trace_id, span_id, trace_flags, severity_text, severity_number, scope_name, scope_version, body, attributes_string, attributes_number, attributes_bool, resources_string, scope_string FROM signoz_logs.distributed_logs_v2 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND (LOWER(body) LIKE LOWER(?)) AND timestamp >= ? AND ts_bucket_start >= ? AND timestamp < ? AND ts_bucket_start <= ? LIMIT ?",
Args: []any{"cartservice", "%service.name%", "%service.name\":\"cartservice%", uint64(1747945619), uint64(1747983448), "%error%", "1747947419000000000", uint64(1747945619), "1747983448000000000", uint64(1747983448), 10},
},
expectedErr: nil,
@@ -644,7 +673,7 @@ func TestStatementBuilderListQueryServiceCollision(t *testing.T) {
},
},
expected: qbtypes.Statement{
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_logs.distributed_logs_v2_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?) SELECT timestamp, id, trace_id, span_id, trace_flags, severity_text, severity_number, scope_name, scope_version, body, attributes_string, attributes_number, attributes_bool, resources_string, scope_string FROM signoz_logs.distributed_logs_v2 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND LOWER(body) LIKE LOWER(?) AND timestamp >= ? AND ts_bucket_start >= ? AND timestamp < ? AND ts_bucket_start <= ? ORDER BY `attribute_string_materialized$$key$$name` AS `materialized.key.name` desc LIMIT ?",
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_logs.distributed_logs_v2_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint) SELECT timestamp, id, trace_id, span_id, trace_flags, severity_text, severity_number, scope_name, scope_version, body, attributes_string, attributes_number, attributes_bool, resources_string, scope_string FROM signoz_logs.distributed_logs_v2 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND LOWER(body) LIKE LOWER(?) AND timestamp >= ? AND ts_bucket_start >= ? AND timestamp < ? AND ts_bucket_start <= ? ORDER BY `attribute_string_materialized$$key$$name` AS `materialized.key.name` desc LIMIT ?",
Args: []any{"cartservice", "%service.name%", "%service.name\":\"cartservice%", uint64(1747945619), uint64(1747983448), "%error%", "1747947419000000000", uint64(1747945619), "1747983448000000000", uint64(1747983448), 10},
},
expectedErr: nil,
@@ -670,6 +699,9 @@ func TestStatementBuilderListQueryServiceCollision(t *testing.T) {
DefaultFullTextColumn,
GetBodyJSONKey,
fl,
nil,
false,
100000,
)
for _, c := range cases {
@@ -894,6 +926,9 @@ func TestAdjustKey(t *testing.T) {
DefaultFullTextColumn,
GetBodyJSONKey,
fl,
nil,
false,
100000,
)
for _, c := range cases {
@@ -1039,6 +1074,9 @@ func TestStmtBuilderBodyField(t *testing.T) {
DefaultFullTextColumn,
GetBodyJSONKey,
fl,
nil,
false,
100000,
)
q, err := statementBuilder.Build(context.Background(), 1747947419000, 1747983448000, c.requestType, c.query, nil)
@@ -1138,6 +1176,9 @@ func TestStmtBuilderBodyFullTextSearch(t *testing.T) {
DefaultFullTextColumn,
GetBodyJSONKey,
fl,
nil,
false,
100000,
)
q, err := statementBuilder.Build(context.Background(), 1747947419000, 1747983448000, c.requestType, c.query, nil)
@@ -1157,3 +1198,110 @@ func TestStmtBuilderBodyFullTextSearch(t *testing.T) {
})
}
}
// TestSkipResourceFingerprintLogs exercises the three resolver outcomes for
// logs: use-CTE (count < threshold), fallback (count >= threshold), and the
// legacy path (feature disabled).
func TestSkipResourceFingerprintLogs(t *testing.T) {
const (
startMs = uint64(1747947419000)
endMs = uint64(1747983448000)
threshold = uint64(10)
)
query := qbtypes.QueryBuilderQuery[qbtypes.LogAggregation]{
Signal: telemetrytypes.SignalLogs,
Filter: &qbtypes.Filter{
Expression: "service.name = 'redis-manual'",
},
Limit: 5,
}
t.Run("disabled uses the legacy CTE", func(t *testing.T) {
sb := newSkipResourceFingerprintLogsBuilder(t, nil, false, threshold)
stmt, err := sb.Build(context.Background(), startMs, endMs, qbtypes.RequestTypeRaw, query, nil)
require.NoError(t, err)
require.Contains(t, stmt.Query, "__resource_filter AS (SELECT fingerprint")
require.Contains(t, stmt.Query, "resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter)")
})
t.Run("CTE attached when count below threshold", func(t *testing.T) {
mockStore := telemetrystoretest.New(telemetrystore.Config{}, &regexQueryMatcher{})
mock := mockStore.Mock()
mock.ExpectQueryRow(`SELECT count\(\) FROM \(SELECT fingerprint FROM signoz_logs\.distributed_logs_v2_resource`).
WillReturnRow(cmock.NewRow([]cmock.ColumnType{
{Name: "count", Type: "UInt64"},
}, []any{uint64(2)}))
sb := newSkipResourceFingerprintLogsBuilder(t, mockStore, true, threshold)
stmt, err := sb.Build(context.Background(), startMs, endMs, qbtypes.RequestTypeRaw, query, nil)
require.NoError(t, err)
require.Contains(t, stmt.Query, "__resource_filter AS (SELECT fingerprint")
require.Contains(t, stmt.Query, "resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter)")
require.NoError(t, mock.ExpectationsWereMet())
})
t.Run("fallback when count at or above threshold", func(t *testing.T) {
mockStore := telemetrystoretest.New(telemetrystore.Config{}, &regexQueryMatcher{})
mock := mockStore.Mock()
mock.ExpectQueryRow(`SELECT count\(\) FROM \(SELECT fingerprint FROM signoz_logs\.distributed_logs_v2_resource`).
WillReturnRow(cmock.NewRow([]cmock.ColumnType{
{Name: "count", Type: "UInt64"},
}, []any{threshold}))
sb := newSkipResourceFingerprintLogsBuilder(t, mockStore, true, threshold)
stmt, err := sb.Build(context.Background(), startMs, endMs, qbtypes.RequestTypeRaw, query, nil)
require.NoError(t, err)
require.NotContains(t, stmt.Query, "__resource_filter AS")
require.NotContains(t, stmt.Query, "resource_fingerprint")
require.Contains(t, stmt.Query, "service.name")
require.NoError(t, mock.ExpectationsWereMet())
})
}
func newSkipResourceFingerprintLogsBuilder(
t *testing.T,
telemetryStore telemetrystore.TelemetryStore,
skipEnable bool,
threshold uint64,
) *logQueryStatementBuilder {
t.Helper()
fl := flaggertest.New(t)
fm := NewFieldMapper(fl)
cb := NewConditionBuilder(fm, fl)
mockMetadataStore := telemetrytypestest.NewMockMetadataStore()
mockMetadataStore.KeysMap = buildCompleteFieldKeyMap(time.Date(2024, 1, 15, 10, 0, 0, 0, time.UTC))
aggExprRewriter := querybuilder.NewAggExprRewriter(
instrumentationtest.New().ToProviderSettings(),
DefaultFullTextColumn,
fm,
cb,
GetBodyJSONKey,
fl,
)
return NewLogQueryStatementBuilder(
instrumentationtest.New().ToProviderSettings(),
mockMetadataStore,
fm,
cb,
aggExprRewriter,
DefaultFullTextColumn,
GetBodyJSONKey,
fl,
telemetryStore,
skipEnable,
threshold,
)
}

View File

@@ -0,0 +1,77 @@
package telemetryresourcefilter
import (
"context"
"github.com/SigNoz/signoz/pkg/factory"
"github.com/SigNoz/signoz/pkg/flagger"
"github.com/SigNoz/signoz/pkg/telemetrystore"
qbtypes "github.com/SigNoz/signoz/pkg/types/querybuildertypes/querybuildertypesv5"
"github.com/SigNoz/signoz/pkg/types/telemetrytypes"
)
type ResourceFingerprintResolver[T any] struct {
stmtBuilder *resourceFilterStatementBuilder[T]
telemetryStore telemetrystore.TelemetryStore
threshold uint64
}
func NewResolver[T any](
settings factory.ProviderSettings,
dbName string,
tableName string,
signal telemetrytypes.Signal,
source telemetrytypes.Source,
metadataStore telemetrytypes.MetadataStore,
fullTextColumn *telemetrytypes.TelemetryFieldKey,
jsonKeyToKey qbtypes.JsonKeyToFieldFunc,
fl flagger.Flagger,
telemetryStore telemetrystore.TelemetryStore,
threshold uint64,
) *ResourceFingerprintResolver[T] {
return &ResourceFingerprintResolver[T]{
stmtBuilder: New[T](
settings,
dbName,
tableName,
signal,
source,
metadataStore,
fullTextColumn,
jsonKeyToKey,
fl,
),
telemetryStore: telemetryStore,
threshold: threshold,
}
}
func (r *ResourceFingerprintResolver[T]) StatementBuilder() qbtypes.StatementBuilder[T] {
return r.stmtBuilder
}
func (r *ResourceFingerprintResolver[T]) Resolve(
ctx context.Context,
query qbtypes.QueryBuilderQuery[T],
start, end uint64,
variables map[string]qbtypes.VariableItem,
) (qbtypes.ResourceFilterResolveKind, error) {
countStmt, err := r.stmtBuilder.BuildCount(ctx, start, end, query, variables)
if err != nil {
return qbtypes.ResourceFilterResolveKindNoOp, err
}
if countStmt == nil {
return qbtypes.ResourceFilterResolveKindNoOp, nil
}
var count uint64
row := r.telemetryStore.ClickhouseDB().QueryRow(ctx, countStmt.Query, countStmt.Args...)
if err := row.Scan(&count); err != nil {
return qbtypes.ResourceFilterResolveKindNoOp, err
}
if count >= r.threshold {
return qbtypes.ResourceFilterResolveKindFallback, nil
}
return qbtypes.ResourceFilterResolveKindUseCTE, nil
}

View File

@@ -116,6 +116,10 @@ func (b *resourceFilterStatementBuilder[T]) Build(
return nil, nil //nolint:nilnil
}
// Group by fingerprint instead of using DISTINCT; on ClickHouse GROUP BY
// parallelizes across multiple threads and is faster for deduplication.
q.GroupBy("fingerprint")
stmt, args := q.BuildWithFlavor(sqlbuilder.ClickHouse)
return &qbtypes.Statement{
Query: stmt,
@@ -123,6 +127,25 @@ func (b *resourceFilterStatementBuilder[T]) Build(
}, nil
}
// BuildCount returns a statement that counts the distinct fingerprints matching
// the resource filter. Returns (nil, nil) when the filter is a no-op.
func (b *resourceFilterStatementBuilder[T]) BuildCount(
ctx context.Context,
start uint64,
end uint64,
query qbtypes.QueryBuilderQuery[T],
variables map[string]qbtypes.VariableItem,
) (*qbtypes.Statement, error) {
inner, err := b.Build(ctx, start, end, qbtypes.RequestTypeRaw, query, variables)
if err != nil || inner == nil {
return nil, err
}
return &qbtypes.Statement{
Query: fmt.Sprintf("SELECT count() FROM (%s)", inner.Query),
Args: inner.Args,
}, nil
}
// addConditions adds both filter and time conditions to the query.
// Returns true (isNoOp) when the filter expression evaluated to no resource conditions,
// meaning the CTE would select all fingerprints and should be skipped entirely.

View File

@@ -119,7 +119,7 @@ func TestResourceFilterStatementBuilder_Traces(t *testing.T) {
start: testStartNs,
end: testEndNs,
expected: &qbtypes.Statement{
Query: "SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?",
Query: "SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint",
Args: []any{"redis-manual", "%service.name%", "%service.name\":\"redis-manual%", expectedBucketStart, expectedBucketEnd},
},
},
@@ -134,7 +134,7 @@ func TestResourceFilterStatementBuilder_Traces(t *testing.T) {
start: testStartNs,
end: testEndNs,
expected: &qbtypes.Statement{
Query: "SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE ((simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND (simpleJSONExtractString(labels, 'k8s.namespace.name') = ? AND labels LIKE ? AND labels LIKE ?)) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?",
Query: "SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE ((simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND (simpleJSONExtractString(labels, 'k8s.namespace.name') = ? AND labels LIKE ? AND labels LIKE ?)) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint",
Args: []any{"redis-manual", "%service.name%", "%service.name\":\"redis-manual%", "production", "%k8s.namespace.name%", "%k8s.namespace.name\":\"production%", expectedBucketStart, expectedBucketEnd},
},
},
@@ -183,7 +183,7 @@ func TestResourceFilterStatementBuilder_Traces(t *testing.T) {
start: testStartNs,
end: testEndNs,
expected: &qbtypes.Statement{
Query: "SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (LOWER(simpleJSONExtractString(labels, 'service.name')) LIKE LOWER(?) AND labels LIKE ? AND LOWER(labels) LIKE LOWER(?)) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?",
Query: "SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (LOWER(simpleJSONExtractString(labels, 'service.name')) LIKE LOWER(?) AND labels LIKE ? AND LOWER(labels) LIKE LOWER(?)) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint",
Args: []any{"redis%", "%service.name%", "%service.name%redis%%", expectedBucketStart, expectedBucketEnd},
},
},
@@ -198,7 +198,7 @@ func TestResourceFilterStatementBuilder_Traces(t *testing.T) {
start: testStartNs,
end: testEndNs,
expected: &qbtypes.Statement{
Query: "SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONHas(labels, 'service.name') = ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?",
Query: "SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONHas(labels, 'service.name') = ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint",
Args: []any{true, "%service.name%", expectedBucketStart, expectedBucketEnd},
},
},
@@ -213,7 +213,7 @@ func TestResourceFilterStatementBuilder_Traces(t *testing.T) {
start: testStartNs,
end: testEndNs,
expected: &qbtypes.Statement{
Query: "SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONHas(labels, 'service.name') <> ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?",
Query: "SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONHas(labels, 'service.name') <> ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint",
Args: []any{true, expectedBucketStart, expectedBucketEnd},
},
},
@@ -228,7 +228,7 @@ func TestResourceFilterStatementBuilder_Traces(t *testing.T) {
start: testStartNs,
end: testEndNs,
expected: &qbtypes.Statement{
Query: "SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE ((simpleJSONExtractString(labels, 'service.name') = ? OR simpleJSONExtractString(labels, 'service.name') = ?) AND labels LIKE ? AND (labels LIKE ? OR labels LIKE ?)) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?",
Query: "SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE ((simpleJSONExtractString(labels, 'service.name') = ? OR simpleJSONExtractString(labels, 'service.name') = ?) AND labels LIKE ? AND (labels LIKE ? OR labels LIKE ?)) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint",
Args: []any{"redis", "postgres", "%service.name%", "%service.name\":\"redis%", "%service.name\":\"postgres%", expectedBucketStart, expectedBucketEnd},
},
},
@@ -243,7 +243,7 @@ func TestResourceFilterStatementBuilder_Traces(t *testing.T) {
start: testStartNs,
end: testEndNs,
expected: &qbtypes.Statement{
Query: "SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE ((simpleJSONExtractString(labels, 'service.name') <> ? AND simpleJSONExtractString(labels, 'service.name') <> ?) AND (labels NOT LIKE ? AND labels NOT LIKE ?)) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?",
Query: "SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE ((simpleJSONExtractString(labels, 'service.name') <> ? AND simpleJSONExtractString(labels, 'service.name') <> ?) AND (labels NOT LIKE ? AND labels NOT LIKE ?)) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint",
Args: []any{"redis", "postgres", "%service.name\":\"redis%", "%service.name\":\"postgres%", expectedBucketStart, expectedBucketEnd},
},
},
@@ -258,7 +258,7 @@ func TestResourceFilterStatementBuilder_Traces(t *testing.T) {
start: testStartNs,
end: testEndNs,
expected: &qbtypes.Statement{
Query: "SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (LOWER(simpleJSONExtractString(labels, 'service.name')) LIKE LOWER(?) AND labels LIKE ? AND LOWER(labels) LIKE LOWER(?)) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?",
Query: "SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (LOWER(simpleJSONExtractString(labels, 'service.name')) LIKE LOWER(?) AND labels LIKE ? AND LOWER(labels) LIKE LOWER(?)) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint",
Args: []any{"%redis%", "%service.name%", "%service.name%redis%", expectedBucketStart, expectedBucketEnd},
},
},
@@ -273,7 +273,7 @@ func TestResourceFilterStatementBuilder_Traces(t *testing.T) {
start: testStartNs,
end: testEndNs,
expected: &qbtypes.Statement{
Query: "SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (match(simpleJSONExtractString(labels, 'service.name'), ?) AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?",
Query: "SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (match(simpleJSONExtractString(labels, 'service.name'), ?) AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint",
Args: []any{"redis.*", "%service.name%", expectedBucketStart, expectedBucketEnd},
},
},
@@ -288,7 +288,7 @@ func TestResourceFilterStatementBuilder_Traces(t *testing.T) {
start: testStartNs,
end: testEndNs,
expected: &qbtypes.Statement{
Query: "SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') <> ? AND labels NOT LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?",
Query: "SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') <> ? AND labels NOT LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint",
Args: []any{"redis", "%service.name\":\"redis%", expectedBucketStart, expectedBucketEnd},
},
},
@@ -315,7 +315,7 @@ func TestResourceFilterStatementBuilder_Traces(t *testing.T) {
start: testStartNs,
end: 0,
expected: &qbtypes.Statement{
Query: "SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ?",
Query: "SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? GROUP BY fingerprint",
Args: []any{"redis", "%service.name%", "%service.name\":\"redis%", expectedBucketStart},
},
},
@@ -330,7 +330,7 @@ func TestResourceFilterStatementBuilder_Traces(t *testing.T) {
start: testStartNs,
end: testEndNs,
expected: &qbtypes.Statement{
Query: "SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE NOT (((simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?))) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?",
Query: "SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE NOT (((simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?))) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint",
Args: []any{"redis", "%service.name%", "%service.name\":\"redis%", expectedBucketStart, expectedBucketEnd},
},
},
@@ -406,7 +406,7 @@ func TestResourceFilterStatementBuilder_Logs(t *testing.T) {
start: testStartNs,
end: testEndNs,
expected: &qbtypes.Statement{
Query: "SELECT fingerprint FROM signoz_logs.distributed_logs_v2_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?",
Query: "SELECT fingerprint FROM signoz_logs.distributed_logs_v2_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint",
Args: []any{"redis-manual", "%service.name%", "%service.name\":\"redis-manual%", expectedBucketStart, expectedBucketEnd},
},
},
@@ -443,7 +443,7 @@ func TestResourceFilterStatementBuilder_Logs(t *testing.T) {
start: testStartNs,
end: testEndNs,
expected: &qbtypes.Statement{
Query: "SELECT fingerprint FROM signoz_logs.distributed_logs_v2_resource WHERE ((simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND (simpleJSONExtractString(labels, 'k8s.namespace.name') = ? AND labels LIKE ? AND labels LIKE ?)) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?",
Query: "SELECT fingerprint FROM signoz_logs.distributed_logs_v2_resource WHERE ((simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND (simpleJSONExtractString(labels, 'k8s.namespace.name') = ? AND labels LIKE ? AND labels LIKE ?)) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint",
Args: []any{"redis", "%service.name%", "%service.name\":\"redis%", "default", "%k8s.namespace.name%", "%k8s.namespace.name\":\"default%", expectedBucketStart, expectedBucketEnd},
},
},
@@ -461,7 +461,7 @@ func TestResourceFilterStatementBuilder_Logs(t *testing.T) {
start: uint64(1769976178000000000), // These will give bucket start 1769974378 and end 1770062578
end: uint64(1770062578000000000),
expected: &qbtypes.Statement{
Query: "SELECT fingerprint FROM signoz_logs.distributed_logs_v2_resource WHERE ((simpleJSONExtractString(labels, 'env') = ? AND labels LIKE ? AND labels LIKE ?) AND (simpleJSONExtractString(labels, 'k8s.deployment.name') = ? AND labels LIKE ? AND labels LIKE ?)) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?",
Query: "SELECT fingerprint FROM signoz_logs.distributed_logs_v2_resource WHERE ((simpleJSONExtractString(labels, 'env') = ? AND labels LIKE ? AND labels LIKE ?) AND (simpleJSONExtractString(labels, 'k8s.deployment.name') = ? AND labels LIKE ? AND labels LIKE ?)) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint",
Args: []any{"prod", "%env%", "%env\":\"prod%", "prod-deployment", "%k8s.deployment.name%", "%k8s.deployment.name\":\"prod-deployment%", uint64(1769974378), uint64(1770062578)},
},
},
@@ -606,7 +606,7 @@ func TestResourceFilterStatementBuilder_Variables(t *testing.T) {
start: testStartNs,
end: testEndNs,
expected: &qbtypes.Statement{
Query: "SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?",
Query: "SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint",
Args: []any{"redis-manual", "%service.name%", "%service.name\":\"redis-manual%", expectedBucketStart, expectedBucketEnd},
},
},

View File

@@ -24,13 +24,13 @@ var (
)
type traceQueryStatementBuilder struct {
logger *slog.Logger
metadataStore telemetrytypes.MetadataStore
fm qbtypes.FieldMapper
cb qbtypes.ConditionBuilder
resourceFilterStmtBuilder qbtypes.StatementBuilder[qbtypes.TraceAggregation]
aggExprRewriter qbtypes.AggExprRewriter
telemetryStore telemetrystore.TelemetryStore
logger *slog.Logger
metadataStore telemetrytypes.MetadataStore
fm qbtypes.FieldMapper
cb qbtypes.ConditionBuilder
resourceFilterResolver *telemetryresourcefilter.ResourceFingerprintResolver[qbtypes.TraceAggregation]
aggExprRewriter qbtypes.AggExprRewriter
skipResourceFingerprintEnabled bool
}
var _ qbtypes.StatementBuilder[qbtypes.TraceAggregation] = (*traceQueryStatementBuilder)(nil)
@@ -43,10 +43,12 @@ func NewTraceQueryStatementBuilder(
aggExprRewriter qbtypes.AggExprRewriter,
telemetryStore telemetrystore.TelemetryStore,
flagger flagger.Flagger,
skipResourceFingerprintEnable bool,
skipResourceFingerprintThreshold uint64,
) *traceQueryStatementBuilder {
tracesSettings := factory.NewScopedProviderSettings(settings, "github.com/SigNoz/signoz/pkg/telemetrytraces")
resourceFilterStmtBuilder := telemetryresourcefilter.New[qbtypes.TraceAggregation](
resourceFilterResolver := telemetryresourcefilter.NewResolver[qbtypes.TraceAggregation](
settings,
DBName,
TracesResourceV3TableName,
@@ -56,16 +58,18 @@ func NewTraceQueryStatementBuilder(
nil,
nil,
flagger,
telemetryStore,
skipResourceFingerprintThreshold,
)
return &traceQueryStatementBuilder{
logger: tracesSettings.Logger(),
metadataStore: metadataStore,
fm: fieldMapper,
cb: conditionBuilder,
resourceFilterStmtBuilder: resourceFilterStmtBuilder,
aggExprRewriter: aggExprRewriter,
telemetryStore: telemetryStore,
logger: tracesSettings.Logger(),
metadataStore: metadataStore,
fm: fieldMapper,
cb: conditionBuilder,
resourceFilterResolver: resourceFilterResolver,
aggExprRewriter: aggExprRewriter,
skipResourceFingerprintEnabled: skipResourceFingerprintEnable,
}
}
@@ -302,9 +306,11 @@ func (b *traceQueryStatementBuilder) buildListQuery(
cteArgs [][]any
)
if frag, args, err := b.maybeAttachResourceFilter(ctx, sb, query, start, end, variables); err != nil {
frag, args, skipResourceFilter, err := b.maybeAttachResourceFilter(ctx, sb, query, start, end, variables)
if err != nil {
return nil, err
} else if frag != "" {
}
if frag != "" {
cteFragments = append(cteFragments, frag)
cteArgs = append(cteArgs, args)
}
@@ -322,7 +328,7 @@ func (b *traceQueryStatementBuilder) buildListQuery(
sb.From(fmt.Sprintf("%s.%s", DBName, SpanIndexV3TableName))
// Add filter conditions
preparedWhereClause, err := b.addFilterCondition(ctx, sb, start, end, query, keys, variables)
preparedWhereClause, err := b.addFilterCondition(ctx, sb, start, end, query, keys, variables, skipResourceFilter)
if err != nil {
return nil, err
}
@@ -383,15 +389,17 @@ func (b *traceQueryStatementBuilder) buildTraceQuery(
cteArgs [][]any
)
if frag, args, err := b.maybeAttachResourceFilter(ctx, distSB, query, start, end, variables); err != nil {
frag, args, skipResourceFilter, err := b.maybeAttachResourceFilter(ctx, distSB, query, start, end, variables)
if err != nil {
return nil, err
} else if frag != "" {
}
if frag != "" {
cteFragments = append(cteFragments, frag)
cteArgs = append(cteArgs, args)
}
// Add filter conditions
preparedWhereClause, err := b.addFilterCondition(ctx, distSB, start, end, query, keys, variables)
preparedWhereClause, err := b.addFilterCondition(ctx, distSB, start, end, query, keys, variables, skipResourceFilter)
if err != nil {
return nil, err
}
@@ -492,9 +500,11 @@ func (b *traceQueryStatementBuilder) buildTimeSeriesQuery(
cteArgs [][]any
)
if frag, args, err := b.maybeAttachResourceFilter(ctx, sb, query, start, end, variables); err != nil {
frag, args, skipResourceFilter, err := b.maybeAttachResourceFilter(ctx, sb, query, start, end, variables)
if err != nil {
return nil, err
} else if frag != "" {
}
if frag != "" {
cteFragments = append(cteFragments, frag)
cteArgs = append(cteArgs, args)
}
@@ -535,7 +545,7 @@ func (b *traceQueryStatementBuilder) buildTimeSeriesQuery(
}
sb.From(fmt.Sprintf("%s.%s", DBName, SpanIndexV3TableName))
preparedWhereClause, err := b.addFilterCondition(ctx, sb, start, end, query, keys, variables)
preparedWhereClause, err := b.addFilterCondition(ctx, sb, start, end, query, keys, variables, skipResourceFilter)
if err != nil {
return nil, err
}
@@ -644,9 +654,11 @@ func (b *traceQueryStatementBuilder) buildScalarQuery(
cteArgs [][]any
)
if frag, args, err := b.maybeAttachResourceFilter(ctx, sb, query, start, end, variables); err != nil {
frag, args, skipResourceFilter, err := b.maybeAttachResourceFilter(ctx, sb, query, start, end, variables)
if err != nil {
return nil, err
} else if frag != "" && !skipResourceCTE {
}
if frag != "" && !skipResourceCTE {
cteFragments = append(cteFragments, frag)
cteArgs = append(cteArgs, args)
}
@@ -688,7 +700,7 @@ func (b *traceQueryStatementBuilder) buildScalarQuery(
sb.From(fmt.Sprintf("%s.%s", DBName, SpanIndexV3TableName))
// Add filter conditions
preparedWhereClause, err := b.addFilterCondition(ctx, sb, start, end, query, keys, variables)
preparedWhereClause, err := b.addFilterCondition(ctx, sb, start, end, query, keys, variables, skipResourceFilter)
if err != nil {
return nil, err
}
@@ -751,6 +763,7 @@ func (b *traceQueryStatementBuilder) addFilterCondition(
query qbtypes.QueryBuilderQuery[qbtypes.TraceAggregation],
keys map[string][]*telemetrytypes.TelemetryFieldKey,
variables map[string]qbtypes.VariableItem,
skipResourceFilter bool,
) (querybuilder.PreparedWhereClause, error) {
var preparedWhereClause querybuilder.PreparedWhereClause
@@ -764,7 +777,7 @@ func (b *traceQueryStatementBuilder) addFilterCondition(
FieldMapper: b.fm,
ConditionBuilder: b.cb,
FieldKeys: keys,
SkipResourceFilter: true,
SkipResourceFilter: skipResourceFilter,
Variables: variables,
StartNs: start,
EndNs: end,
@@ -805,34 +818,30 @@ func (b *traceQueryStatementBuilder) maybeAttachResourceFilter(
query qbtypes.QueryBuilderQuery[qbtypes.TraceAggregation],
start, end uint64,
variables map[string]qbtypes.VariableItem,
) (cteSQL string, cteArgs []any, err error) {
) (cteSQL string, cteArgs []any, skipResourceFilter bool, err error) {
stmt, err := b.buildResourceFilterCTE(ctx, query, start, end, variables)
if b.skipResourceFingerprintEnabled {
decision, err := b.resourceFilterResolver.Resolve(ctx, query, start, end, variables)
if err != nil {
return "", nil, true, err
}
switch decision {
case qbtypes.ResourceFilterResolveKindNoOp:
return "", nil, true, nil
case qbtypes.ResourceFilterResolveKindFallback:
return "", nil, false, nil
}
}
stmt, err := b.resourceFilterResolver.StatementBuilder().Build(
ctx, start, end, qbtypes.RequestTypeRaw, query, variables,
)
if err != nil {
return "", nil, err
return "", nil, true, err
}
if stmt == nil {
return "", nil, nil
return "", nil, true, nil
}
sb.Where("resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter)")
return fmt.Sprintf("__resource_filter AS (%s)", stmt.Query), stmt.Args, nil
}
func (b *traceQueryStatementBuilder) buildResourceFilterCTE(
ctx context.Context,
query qbtypes.QueryBuilderQuery[qbtypes.TraceAggregation],
start, end uint64,
variables map[string]qbtypes.VariableItem,
) (*qbtypes.Statement, error) {
return b.resourceFilterStmtBuilder.Build(
ctx,
start,
end,
qbtypes.RequestTypeRaw,
query,
variables,
)
return fmt.Sprintf("__resource_filter AS (%s)", stmt.Query), stmt.Args, true, nil
}

View File

@@ -2,19 +2,36 @@ package telemetrytraces
import (
"context"
"regexp"
"testing"
"time"
cmock "github.com/SigNoz/clickhouse-go-mock"
"github.com/SigNoz/signoz/pkg/errors"
"github.com/SigNoz/signoz/pkg/flagger/flaggertest"
"github.com/SigNoz/signoz/pkg/instrumentation/instrumentationtest"
"github.com/SigNoz/signoz/pkg/querybuilder"
"github.com/SigNoz/signoz/pkg/telemetrystore"
"github.com/SigNoz/signoz/pkg/telemetrystore/telemetrystoretest"
qbtypes "github.com/SigNoz/signoz/pkg/types/querybuildertypes/querybuildertypesv5"
"github.com/SigNoz/signoz/pkg/types/telemetrytypes"
"github.com/SigNoz/signoz/pkg/types/telemetrytypes/telemetrytypestest"
"github.com/stretchr/testify/require"
)
type regexQueryMatcher struct{}
func (m *regexQueryMatcher) Match(expectedSQL, actualSQL string) error {
re, err := regexp.Compile(expectedSQL)
if err != nil {
return err
}
if !re.MatchString(actualSQL) {
return errors.NewInvalidInputf(errors.CodeInvalidInput, "expected query to match %s, got %s", expectedSQL, actualSQL)
}
return nil
}
func TestStatementBuilder(t *testing.T) {
cases := []struct {
name string
@@ -47,7 +64,7 @@ func TestStatementBuilder(t *testing.T) {
},
},
expected: qbtypes.Statement{
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), __limit_cte AS (SELECT toString(multiIf(multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL) IS NOT NULL, multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL), NULL)) AS `service.name`, count() AS __result_0 FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? GROUP BY `service.name` ORDER BY __result_0 DESC LIMIT ?) SELECT toStartOfInterval(timestamp, INTERVAL 30 SECOND) AS ts, toString(multiIf(multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL) IS NOT NULL, multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL), NULL)) AS `service.name`, count() AS __result_0 FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? AND (`service.name`) GLOBAL IN (SELECT `service.name` FROM __limit_cte) GROUP BY ts, `service.name`",
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint), __limit_cte AS (SELECT toString(multiIf(multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL) IS NOT NULL, multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL), NULL)) AS `service.name`, count() AS __result_0 FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? GROUP BY `service.name` ORDER BY __result_0 DESC LIMIT ?) SELECT toStartOfInterval(timestamp, INTERVAL 30 SECOND) AS ts, toString(multiIf(multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL) IS NOT NULL, multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL), NULL)) AS `service.name`, count() AS __result_0 FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? AND (`service.name`) GLOBAL IN (SELECT `service.name` FROM __limit_cte) GROUP BY ts, `service.name`",
Args: []any{"redis-manual", "%service.name%", "%service.name\":\"redis-manual%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), 10, "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448)},
},
expectedErr: nil,
@@ -136,7 +153,7 @@ func TestStatementBuilder(t *testing.T) {
},
},
expected: qbtypes.Statement{
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), __limit_cte AS (SELECT toString(multiIf(attribute_string_http$$route <> ?, attribute_string_http$$route, NULL)) AS `httpRoute`, count() AS __result_0 FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? GROUP BY `httpRoute` ORDER BY __result_0 DESC LIMIT ?) SELECT toStartOfInterval(timestamp, INTERVAL 30 SECOND) AS ts, toString(multiIf(attribute_string_http$$route <> ?, attribute_string_http$$route, NULL)) AS `httpRoute`, count() AS __result_0 FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? AND (`httpRoute`) GLOBAL IN (SELECT `httpRoute` FROM __limit_cte) GROUP BY ts, `httpRoute`",
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint), __limit_cte AS (SELECT toString(multiIf(attribute_string_http$$route <> ?, attribute_string_http$$route, NULL)) AS `httpRoute`, count() AS __result_0 FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? GROUP BY `httpRoute` ORDER BY __result_0 DESC LIMIT ?) SELECT toStartOfInterval(timestamp, INTERVAL 30 SECOND) AS ts, toString(multiIf(attribute_string_http$$route <> ?, attribute_string_http$$route, NULL)) AS `httpRoute`, count() AS __result_0 FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? AND (`httpRoute`) GLOBAL IN (SELECT `httpRoute` FROM __limit_cte) GROUP BY ts, `httpRoute`",
Args: []any{"redis-manual", "%service.name%", "%service.name\":\"redis-manual%", uint64(1747945619), uint64(1747983448), "", "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), 10, "", "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448)},
},
expectedErr: nil,
@@ -215,7 +232,7 @@ func TestStatementBuilder(t *testing.T) {
},
},
expected: qbtypes.Statement{
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), __limit_cte AS (SELECT toString(multiIf(multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL) IS NOT NULL, multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL), NULL)) AS `service.name`, sum(multiIf(mapContains(attributes_number, 'metric.max_count') = ?, toFloat64(attributes_number['metric.max_count']), NULL)) AS __result_0 FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? GROUP BY `service.name` ORDER BY __result_0 desc LIMIT ?) SELECT toStartOfInterval(timestamp, INTERVAL 30 SECOND) AS ts, toString(multiIf(multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL) IS NOT NULL, multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL), NULL)) AS `service.name`, sum(multiIf(mapContains(attributes_number, 'metric.max_count') = ?, toFloat64(attributes_number['metric.max_count']), NULL)) AS __result_0 FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? AND (`service.name`) GLOBAL IN (SELECT `service.name` FROM __limit_cte) GROUP BY ts, `service.name` ORDER BY ts desc",
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint), __limit_cte AS (SELECT toString(multiIf(multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL) IS NOT NULL, multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL), NULL)) AS `service.name`, sum(multiIf(mapContains(attributes_number, 'metric.max_count') = ?, toFloat64(attributes_number['metric.max_count']), NULL)) AS __result_0 FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? GROUP BY `service.name` ORDER BY __result_0 desc LIMIT ?) SELECT toStartOfInterval(timestamp, INTERVAL 30 SECOND) AS ts, toString(multiIf(multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL) IS NOT NULL, multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL), NULL)) AS `service.name`, sum(multiIf(mapContains(attributes_number, 'metric.max_count') = ?, toFloat64(attributes_number['metric.max_count']), NULL)) AS __result_0 FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? AND (`service.name`) GLOBAL IN (SELECT `service.name` FROM __limit_cte) GROUP BY ts, `service.name` ORDER BY ts desc",
Args: []any{"redis-manual", "%service.name%", "%service.name\":\"redis-manual%", uint64(1747945619), uint64(1747983448), true, "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), 10, true, "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448)},
},
expectedErr: nil,
@@ -244,7 +261,7 @@ func TestStatementBuilder(t *testing.T) {
},
},
expected: qbtypes.Statement{
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), __limit_cte AS (SELECT toString(multiIf(multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL) IS NOT NULL, multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL), NULL)) AS `service.name`, sum(multiIf(`attribute_number_cart$$items_count_exists` = ?, toFloat64(`attribute_number_cart$$items_count`), NULL)) AS __result_0 FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? GROUP BY `service.name` ORDER BY __result_0 DESC LIMIT ?) SELECT toStartOfInterval(timestamp, INTERVAL 30 SECOND) AS ts, toString(multiIf(multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL) IS NOT NULL, multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL), NULL)) AS `service.name`, sum(multiIf(`attribute_number_cart$$items_count_exists` = ?, toFloat64(`attribute_number_cart$$items_count`), NULL)) AS __result_0 FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? AND (`service.name`) GLOBAL IN (SELECT `service.name` FROM __limit_cte) GROUP BY ts, `service.name`",
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint), __limit_cte AS (SELECT toString(multiIf(multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL) IS NOT NULL, multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL), NULL)) AS `service.name`, sum(multiIf(`attribute_number_cart$$items_count_exists` = ?, toFloat64(`attribute_number_cart$$items_count`), NULL)) AS __result_0 FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? GROUP BY `service.name` ORDER BY __result_0 DESC LIMIT ?) SELECT toStartOfInterval(timestamp, INTERVAL 30 SECOND) AS ts, toString(multiIf(multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL) IS NOT NULL, multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL), NULL)) AS `service.name`, sum(multiIf(`attribute_number_cart$$items_count_exists` = ?, toFloat64(`attribute_number_cart$$items_count`), NULL)) AS __result_0 FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? AND (`service.name`) GLOBAL IN (SELECT `service.name` FROM __limit_cte) GROUP BY ts, `service.name`",
Args: []any{"redis-manual", "%service.name%", "%service.name\":\"redis-manual%", uint64(1747945619), uint64(1747983448), true, "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), 10, true, "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448)},
},
expectedErr: nil,
@@ -283,7 +300,7 @@ func TestStatementBuilder(t *testing.T) {
},
},
expected: qbtypes.Statement{
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), __limit_cte AS (SELECT toString(multiIf(multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL) IS NOT NULL, multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL), NULL)) AS `service.name`, sum(multiIf(`attribute_number_cart$$items_count_exists` = ?, toFloat64(`attribute_number_cart$$items_count`), NULL)) AS __result_0 FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? GROUP BY `service.name` ORDER BY `service.name` desc LIMIT ?) SELECT toStartOfInterval(timestamp, INTERVAL 30 SECOND) AS ts, toString(multiIf(multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL) IS NOT NULL, multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL), NULL)) AS `service.name`, sum(multiIf(`attribute_number_cart$$items_count_exists` = ?, toFloat64(`attribute_number_cart$$items_count`), NULL)) AS __result_0 FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? AND (`service.name`) GLOBAL IN (SELECT `service.name` FROM __limit_cte) GROUP BY ts, `service.name` ORDER BY `service.name` desc, ts desc",
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint), __limit_cte AS (SELECT toString(multiIf(multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL) IS NOT NULL, multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL), NULL)) AS `service.name`, sum(multiIf(`attribute_number_cart$$items_count_exists` = ?, toFloat64(`attribute_number_cart$$items_count`), NULL)) AS __result_0 FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? GROUP BY `service.name` ORDER BY `service.name` desc LIMIT ?) SELECT toStartOfInterval(timestamp, INTERVAL 30 SECOND) AS ts, toString(multiIf(multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL) IS NOT NULL, multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL), NULL)) AS `service.name`, sum(multiIf(`attribute_number_cart$$items_count_exists` = ?, toFloat64(`attribute_number_cart$$items_count`), NULL)) AS __result_0 FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? AND (`service.name`) GLOBAL IN (SELECT `service.name` FROM __limit_cte) GROUP BY ts, `service.name` ORDER BY `service.name` desc, ts desc",
Args: []any{"redis-manual", "%service.name%", "%service.name\":\"redis-manual%", uint64(1747945619), uint64(1747983448), true, "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), 10, true, "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448)},
},
expectedErr: nil,
@@ -314,7 +331,7 @@ func TestStatementBuilder(t *testing.T) {
},
},
expected: qbtypes.Statement{
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), __limit_cte AS (SELECT toString(multiIf(response_status_code <> ?, response_status_code, NULL)) AS `responseStatusCode`, count() AS __result_0 FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? GROUP BY `responseStatusCode` ORDER BY __result_0 DESC LIMIT ?) SELECT toStartOfInterval(timestamp, INTERVAL 30 SECOND) AS ts, toString(multiIf(response_status_code <> ?, response_status_code, NULL)) AS `responseStatusCode`, count() AS __result_0 FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? AND (`responseStatusCode`) GLOBAL IN (SELECT `responseStatusCode` FROM __limit_cte) GROUP BY ts, `responseStatusCode`",
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint), __limit_cte AS (SELECT toString(multiIf(response_status_code <> ?, response_status_code, NULL)) AS `responseStatusCode`, count() AS __result_0 FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? GROUP BY `responseStatusCode` ORDER BY __result_0 DESC LIMIT ?) SELECT toStartOfInterval(timestamp, INTERVAL 30 SECOND) AS ts, toString(multiIf(response_status_code <> ?, response_status_code, NULL)) AS `responseStatusCode`, count() AS __result_0 FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? AND (`responseStatusCode`) GLOBAL IN (SELECT `responseStatusCode` FROM __limit_cte) GROUP BY ts, `responseStatusCode`",
Args: []any{"redis-manual", "%service.name%", "%service.name\":\"redis-manual%", uint64(1747945619), uint64(1747983448), "", "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), 10, "", "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448)},
},
expectedErr: nil,
@@ -345,7 +362,7 @@ func TestStatementBuilder(t *testing.T) {
},
},
expected: qbtypes.Statement{
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), __limit_cte AS (SELECT toString(multiIf(response_status_code <> ?, response_status_code, NULL)) AS `responseStatusCode`, quantile(0.90)(multiIf(duration_nano <> ?, duration_nano, NULL)) AS __result_0 FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? GROUP BY `responseStatusCode` ORDER BY __result_0 DESC LIMIT ?) SELECT toStartOfInterval(timestamp, INTERVAL 30 SECOND) AS ts, toString(multiIf(response_status_code <> ?, response_status_code, NULL)) AS `responseStatusCode`, quantile(0.90)(multiIf(duration_nano <> ?, duration_nano, NULL)) AS __result_0 FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? AND (`responseStatusCode`) GLOBAL IN (SELECT `responseStatusCode` FROM __limit_cte) GROUP BY ts, `responseStatusCode`",
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint), __limit_cte AS (SELECT toString(multiIf(response_status_code <> ?, response_status_code, NULL)) AS `responseStatusCode`, quantile(0.90)(multiIf(duration_nano <> ?, duration_nano, NULL)) AS __result_0 FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? GROUP BY `responseStatusCode` ORDER BY __result_0 DESC LIMIT ?) SELECT toStartOfInterval(timestamp, INTERVAL 30 SECOND) AS ts, toString(multiIf(response_status_code <> ?, response_status_code, NULL)) AS `responseStatusCode`, quantile(0.90)(multiIf(duration_nano <> ?, duration_nano, NULL)) AS __result_0 FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? AND (`responseStatusCode`) GLOBAL IN (SELECT `responseStatusCode` FROM __limit_cte) GROUP BY ts, `responseStatusCode`",
Args: []any{"redis-manual", "%service.name%", "%service.name\":\"redis-manual%", uint64(1747945619), uint64(1747983448), "", 0, "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), 10, "", 0, "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448)},
},
expectedErr: nil,
@@ -367,6 +384,8 @@ func TestStatementBuilder(t *testing.T) {
aggExprRewriter,
nil,
fl,
false,
100000,
)
vars := map[string]qbtypes.VariableItem{
@@ -439,7 +458,7 @@ func TestStatementBuilderListQuery(t *testing.T) {
},
},
expected: qbtypes.Statement{
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?) SELECT name AS `name`, multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL) AS `service.name`, duration_nano AS `duration_nano`, `attribute_number_cart$$items_count` AS `cart.items_count`, timestamp AS `timestamp`, span_id AS `span_id`, trace_id AS `trace_id` FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? LIMIT ?",
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint) SELECT name AS `name`, multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL) AS `service.name`, duration_nano AS `duration_nano`, `attribute_number_cart$$items_count` AS `cart.items_count`, timestamp AS `timestamp`, span_id AS `span_id`, trace_id AS `trace_id` FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? LIMIT ?",
Args: []any{"redis-manual", "%service.name%", "%service.name\":\"redis-manual%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), 10},
},
expectedErr: nil,
@@ -468,7 +487,7 @@ func TestStatementBuilderListQuery(t *testing.T) {
Limit: 10,
},
expected: qbtypes.Statement{
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?) SELECT duration_nano AS `duration_nano`, name AS `name`, response_status_code AS `response_status_code`, multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL) AS `service.name`, span_id AS `span_id`, timestamp AS `timestamp`, trace_id AS `trace_id` FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? ORDER BY attributes_string['user.id'] AS `user.id` desc LIMIT ?",
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint) SELECT duration_nano AS `duration_nano`, name AS `name`, response_status_code AS `response_status_code`, multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL) AS `service.name`, span_id AS `span_id`, timestamp AS `timestamp`, trace_id AS `trace_id` FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? ORDER BY attributes_string['user.id'] AS `user.id` desc LIMIT ?",
Args: []any{"redis-manual", "%service.name%", "%service.name\":\"redis-manual%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), 10},
},
expectedErr: nil,
@@ -512,7 +531,7 @@ func TestStatementBuilderListQuery(t *testing.T) {
Limit: 10,
},
expected: qbtypes.Statement{
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?) SELECT name AS `name`, resource_string_service$$name AS `serviceName`, duration_nano AS `durationNano`, http_method AS `httpMethod`, response_status_code AS `responseStatusCode`, timestamp AS `timestamp`, span_id AS `span_id`, trace_id AS `trace_id` FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? LIMIT ?",
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint) SELECT name AS `name`, resource_string_service$$name AS `serviceName`, duration_nano AS `durationNano`, http_method AS `httpMethod`, response_status_code AS `responseStatusCode`, timestamp AS `timestamp`, span_id AS `span_id`, trace_id AS `trace_id` FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? LIMIT ?",
Args: []any{"redis-manual", "%service.name%", "%service.name\":\"redis-manual%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), 10},
},
expectedErr: nil,
@@ -556,7 +575,7 @@ func TestStatementBuilderListQuery(t *testing.T) {
Limit: 10,
},
expected: qbtypes.Statement{
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?) SELECT name AS `name`, resource_string_service$$name AS `serviceName`, duration_nano AS `durationNano`, http_method AS `httpMethod`, multiIf(toString(`attribute_string_mixed$$materialization$$key`) != '', toString(`attribute_string_mixed$$materialization$$key`), toString(multiIf(resource.`mixed.materialization.key` IS NOT NULL, resource.`mixed.materialization.key`::String, mapContains(resources_string, 'mixed.materialization.key'), resources_string['mixed.materialization.key'], NULL)) != '', toString(multiIf(resource.`mixed.materialization.key` IS NOT NULL, resource.`mixed.materialization.key`::String, mapContains(resources_string, 'mixed.materialization.key'), resources_string['mixed.materialization.key'], NULL)), NULL) AS `mixed.materialization.key`, timestamp AS `timestamp`, span_id AS `span_id`, trace_id AS `trace_id` FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? LIMIT ?",
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint) SELECT name AS `name`, resource_string_service$$name AS `serviceName`, duration_nano AS `durationNano`, http_method AS `httpMethod`, multiIf(toString(`attribute_string_mixed$$materialization$$key`) != '', toString(`attribute_string_mixed$$materialization$$key`), toString(multiIf(resource.`mixed.materialization.key` IS NOT NULL, resource.`mixed.materialization.key`::String, mapContains(resources_string, 'mixed.materialization.key'), resources_string['mixed.materialization.key'], NULL)) != '', toString(multiIf(resource.`mixed.materialization.key` IS NOT NULL, resource.`mixed.materialization.key`::String, mapContains(resources_string, 'mixed.materialization.key'), resources_string['mixed.materialization.key'], NULL)), NULL) AS `mixed.materialization.key`, timestamp AS `timestamp`, span_id AS `span_id`, trace_id AS `trace_id` FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? LIMIT ?",
Args: []any{"redis-manual", "%service.name%", "%service.name\":\"redis-manual%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), 10},
},
expectedErr: nil,
@@ -601,7 +620,7 @@ func TestStatementBuilderListQuery(t *testing.T) {
Limit: 10,
},
expected: qbtypes.Statement{
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?) SELECT name AS `name`, resource_string_service$$name AS `serviceName`, duration_nano AS `durationNano`, http_method AS `httpMethod`, `attribute_string_mixed$$materialization$$key` AS `mixed.materialization.key`, timestamp AS `timestamp`, span_id AS `span_id`, trace_id AS `trace_id` FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? LIMIT ?",
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint) SELECT name AS `name`, resource_string_service$$name AS `serviceName`, duration_nano AS `durationNano`, http_method AS `httpMethod`, `attribute_string_mixed$$materialization$$key` AS `mixed.materialization.key`, timestamp AS `timestamp`, span_id AS `span_id`, trace_id AS `trace_id` FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? LIMIT ?",
Args: []any{"redis-manual", "%service.name%", "%service.name\":\"redis-manual%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), 10},
},
expectedErr: nil,
@@ -662,6 +681,8 @@ func TestStatementBuilderListQuery(t *testing.T) {
aggExprRewriter,
nil,
fl,
false,
100000,
)
for _, c := range cases {
@@ -771,6 +792,8 @@ func TestStatementBuilderListQueryWithCorruptData(t *testing.T) {
aggExprRewriter,
nil,
fl,
false,
100000,
)
q, err := statementBuilder.Build(context.Background(), 1747947419000, 1747983448000, c.requestType, c.query, nil)
@@ -807,7 +830,7 @@ func TestStatementBuilderTraceQuery(t *testing.T) {
Limit: 10,
},
expected: qbtypes.Statement{
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), __toe AS (SELECT trace_id FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __toe_duration_sorted AS (SELECT trace_id, duration_nano, resource_string_service$$name as `service.name`, name FROM signoz_traces.distributed_signoz_index_v3 WHERE parent_span_id = '' AND trace_id GLOBAL IN __toe AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? ORDER BY duration_nano DESC LIMIT 1 BY trace_id) SELECT __toe_duration_sorted.`service.name` AS `service.name`, __toe_duration_sorted.name AS `name`, count() AS span_count, __toe_duration_sorted.duration_nano AS `duration_nano`, __toe_duration_sorted.trace_id AS `trace_id` FROM __toe INNER JOIN __toe_duration_sorted ON __toe.trace_id = __toe_duration_sorted.trace_id GROUP BY trace_id, duration_nano, name, `service.name` ORDER BY duration_nano DESC LIMIT 1 BY trace_id LIMIT ? SETTINGS distributed_product_mode='allow', max_memory_usage=10000000000",
Query: "WITH __resource_filter AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint), __toe AS (SELECT trace_id FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __toe_duration_sorted AS (SELECT trace_id, duration_nano, resource_string_service$$name as `service.name`, name FROM signoz_traces.distributed_signoz_index_v3 WHERE parent_span_id = '' AND trace_id GLOBAL IN __toe AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? ORDER BY duration_nano DESC LIMIT 1 BY trace_id) SELECT __toe_duration_sorted.`service.name` AS `service.name`, __toe_duration_sorted.name AS `name`, count() AS span_count, __toe_duration_sorted.duration_nano AS `duration_nano`, __toe_duration_sorted.trace_id AS `trace_id` FROM __toe INNER JOIN __toe_duration_sorted ON __toe.trace_id = __toe_duration_sorted.trace_id GROUP BY trace_id, duration_nano, name, `service.name` ORDER BY duration_nano DESC LIMIT 1 BY trace_id LIMIT ? SETTINGS distributed_product_mode='allow', max_memory_usage=10000000000",
Args: []any{"redis-manual", "%service.name%", "%service.name\":\"redis-manual%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), 10},
},
expectedErr: nil,
@@ -923,6 +946,8 @@ func TestStatementBuilderTraceQuery(t *testing.T) {
aggExprRewriter,
nil,
fl,
false,
100000,
)
for _, c := range cases {
@@ -1138,6 +1163,8 @@ func TestAdjustKey(t *testing.T) {
aggExprRewriter,
nil,
fl,
false,
100000,
)
for _, c := range cases {
@@ -1412,6 +1439,8 @@ func TestAdjustKeys(t *testing.T) {
aggExprRewriter,
nil,
fl,
false,
100000,
)
for _, c := range cases {
@@ -1483,3 +1512,114 @@ func TestAdjustKeys(t *testing.T) {
})
}
}
// TestSkipResourceFingerprint exercises the three resolver outcomes when
// skip_resource_fingerprint is enabled: use-CTE (count < threshold),
// fallback (count >= threshold), and the legacy path (feature disabled).
func TestSkipResourceFingerprint(t *testing.T) {
const (
startMs = uint64(1747947419000)
endMs = uint64(1747983448000)
threshold = uint64(10)
)
query := qbtypes.QueryBuilderQuery[qbtypes.TraceAggregation]{
Signal: telemetrytypes.SignalTraces,
Filter: &qbtypes.Filter{
Expression: "service.name = 'redis-manual'",
},
SelectFields: []telemetrytypes.TelemetryFieldKey{
{
Name: "name",
FieldContext: telemetrytypes.FieldContextSpan,
FieldDataType: telemetrytypes.FieldDataTypeString,
},
},
Limit: 5,
}
t.Run("disabled uses the legacy CTE", func(t *testing.T) {
sb := newSkipResourceFingerprintBuilder(t, nil, false, threshold)
stmt, err := sb.Build(context.Background(), startMs, endMs, qbtypes.RequestTypeRaw, query, nil)
require.NoError(t, err)
require.Contains(t, stmt.Query, "__resource_filter AS (SELECT fingerprint")
require.Contains(t, stmt.Query, "resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter)")
})
t.Run("CTE attached when count below threshold", func(t *testing.T) {
mockStore := telemetrystoretest.New(telemetrystore.Config{}, &regexQueryMatcher{})
mock := mockStore.Mock()
// Only the count query runs against the telemetry store; the CTE
// itself is embedded as SQL in the main query (no extra round trip).
mock.ExpectQueryRow(`SELECT count\(\) FROM \(SELECT fingerprint FROM signoz_traces\.distributed_traces_v3_resource`).
WillReturnRow(cmock.NewRow([]cmock.ColumnType{
{Name: "count", Type: "UInt64"},
}, []any{uint64(2)}))
sb := newSkipResourceFingerprintBuilder(t, mockStore, true, threshold)
stmt, err := sb.Build(context.Background(), startMs, endMs, qbtypes.RequestTypeRaw, query, nil)
require.NoError(t, err)
require.Contains(t, stmt.Query, "__resource_filter AS (SELECT fingerprint")
require.Contains(t, stmt.Query, "resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter)")
require.NoError(t, mock.ExpectationsWereMet())
})
t.Run("fallback when count at or above threshold", func(t *testing.T) {
mockStore := telemetrystoretest.New(telemetrystore.Config{}, &regexQueryMatcher{})
mock := mockStore.Mock()
mock.ExpectQueryRow(`SELECT count\(\) FROM \(SELECT fingerprint FROM signoz_traces\.distributed_traces_v3_resource`).
WillReturnRow(cmock.NewRow([]cmock.ColumnType{
{Name: "count", Type: "UInt64"},
}, []any{threshold}))
sb := newSkipResourceFingerprintBuilder(t, mockStore, true, threshold)
stmt, err := sb.Build(context.Background(), startMs, endMs, qbtypes.RequestTypeRaw, query, nil)
require.NoError(t, err)
require.NotContains(t, stmt.Query, "__resource_filter AS")
require.NotContains(t, stmt.Query, "resource_fingerprint")
// resource conditions are pushed onto the main table via the
// resource.`service.name` / resources_string lookup
require.Contains(t, stmt.Query, "service.name")
require.NoError(t, mock.ExpectationsWereMet())
})
}
func newSkipResourceFingerprintBuilder(
t *testing.T,
telemetryStore telemetrystore.TelemetryStore,
skipEnable bool,
threshold uint64,
) *traceQueryStatementBuilder {
t.Helper()
fm := NewFieldMapper()
cb := NewConditionBuilder(fm)
mockMetadataStore := telemetrytypestest.NewMockMetadataStore()
mockMetadataStore.KeysMap = buildCompleteFieldKeyMap()
fl := flaggertest.New(t)
aggExprRewriter := querybuilder.NewAggExprRewriter(
instrumentationtest.New().ToProviderSettings(), nil, fm, cb, nil, fl,
)
return NewTraceQueryStatementBuilder(
instrumentationtest.New().ToProviderSettings(),
mockMetadataStore,
fm,
cb,
aggExprRewriter,
telemetryStore,
fl,
skipEnable,
threshold,
)
}

View File

@@ -67,7 +67,7 @@ func TestTraceOperatorStatementBuilder(t *testing.T) {
},
},
expected: qbtypes.Statement{
Query: "WITH toDateTime64(1747947419000000000, 9) AS t_from, toDateTime64(1747983448000000000, 9) AS t_to, 1747945619 AS bucket_from, 1747983448 AS bucket_to, all_spans AS (SELECT *, resource_string_service$$name AS `service.name` FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_A AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), A AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_A) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_B AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), B AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_B) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), A_DIR_DESC_B AS (SELECT p.* FROM A AS p INNER JOIN B AS c ON p.trace_id = c.trace_id AND p.span_id = c.parent_span_id) SELECT timestamp, trace_id, span_id, name, duration_nano, parent_span_id, multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL) AS `service.name` FROM A_DIR_DESC_B ORDER BY timestamp DESC LIMIT ? SETTINGS distributed_product_mode='allow', max_memory_usage=10000000000",
Query: "WITH toDateTime64(1747947419000000000, 9) AS t_from, toDateTime64(1747983448000000000, 9) AS t_to, 1747945619 AS bucket_from, 1747983448 AS bucket_to, all_spans AS (SELECT *, resource_string_service$$name AS `service.name` FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_A AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint), A AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_A) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_B AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint), B AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_B) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), A_DIR_DESC_B AS (SELECT p.* FROM A AS p INNER JOIN B AS c ON p.trace_id = c.trace_id AND p.span_id = c.parent_span_id) SELECT timestamp, trace_id, span_id, name, duration_nano, parent_span_id, multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL) AS `service.name` FROM A_DIR_DESC_B ORDER BY timestamp DESC LIMIT ? SETTINGS distributed_product_mode='allow', max_memory_usage=10000000000",
Args: []any{"1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "frontend", "%service.name%", "%service.name\":\"frontend%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "backend", "%service.name%", "%service.name\":\"backend%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), 10},
},
expectedErr: nil,
@@ -104,7 +104,7 @@ func TestTraceOperatorStatementBuilder(t *testing.T) {
},
},
expected: qbtypes.Statement{
Query: "WITH toDateTime64(1747947419000000000, 9) AS t_from, toDateTime64(1747983448000000000, 9) AS t_to, 1747945619 AS bucket_from, 1747983448 AS bucket_to, all_spans AS (SELECT *, resource_string_service$$name AS `service.name` FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_A AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), A AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_A) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_B AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), B AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_B) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), A_INDIR_DESC_B AS (WITH RECURSIVE up AS (SELECT d.trace_id, d.span_id, d.parent_span_id, 0 AS depth FROM B AS d UNION ALL SELECT p.trace_id, p.span_id, p.parent_span_id, up.depth + 1 FROM all_spans AS p JOIN up ON p.trace_id = up.trace_id AND p.span_id = up.parent_span_id WHERE up.depth < 100) SELECT DISTINCT a.* FROM A AS a GLOBAL INNER JOIN (SELECT DISTINCT trace_id, span_id FROM up WHERE depth > 0 ) AS ancestors ON ancestors.trace_id = a.trace_id AND ancestors.span_id = a.span_id) SELECT timestamp, trace_id, span_id, name, duration_nano, parent_span_id FROM A_INDIR_DESC_B ORDER BY timestamp DESC LIMIT ? SETTINGS distributed_product_mode='allow', max_memory_usage=10000000000",
Query: "WITH toDateTime64(1747947419000000000, 9) AS t_from, toDateTime64(1747983448000000000, 9) AS t_to, 1747945619 AS bucket_from, 1747983448 AS bucket_to, all_spans AS (SELECT *, resource_string_service$$name AS `service.name` FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_A AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint), A AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_A) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_B AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint), B AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_B) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), A_INDIR_DESC_B AS (WITH RECURSIVE up AS (SELECT d.trace_id, d.span_id, d.parent_span_id, 0 AS depth FROM B AS d UNION ALL SELECT p.trace_id, p.span_id, p.parent_span_id, up.depth + 1 FROM all_spans AS p JOIN up ON p.trace_id = up.trace_id AND p.span_id = up.parent_span_id WHERE up.depth < 100) SELECT DISTINCT a.* FROM A AS a GLOBAL INNER JOIN (SELECT DISTINCT trace_id, span_id FROM up WHERE depth > 0 ) AS ancestors ON ancestors.trace_id = a.trace_id AND ancestors.span_id = a.span_id) SELECT timestamp, trace_id, span_id, name, duration_nano, parent_span_id FROM A_INDIR_DESC_B ORDER BY timestamp DESC LIMIT ? SETTINGS distributed_product_mode='allow', max_memory_usage=10000000000",
Args: []any{"1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "gateway", "%service.name%", "%service.name\":\"gateway%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "database", "%service.name%", "%service.name\":\"database%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), 5},
},
expectedErr: nil,
@@ -141,7 +141,7 @@ func TestTraceOperatorStatementBuilder(t *testing.T) {
},
},
expected: qbtypes.Statement{
Query: "WITH toDateTime64(1747947419000000000, 9) AS t_from, toDateTime64(1747983448000000000, 9) AS t_to, 1747945619 AS bucket_from, 1747983448 AS bucket_to, all_spans AS (SELECT *, resource_string_service$$name AS `service.name` FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_A AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), A AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_A) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_B AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), B AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_B) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), A_AND_B AS (SELECT l.* FROM A AS l INNER JOIN B AS r ON l.trace_id = r.trace_id) SELECT timestamp, trace_id, span_id, name, duration_nano, parent_span_id FROM A_AND_B ORDER BY timestamp DESC LIMIT ? SETTINGS distributed_product_mode='allow', max_memory_usage=10000000000",
Query: "WITH toDateTime64(1747947419000000000, 9) AS t_from, toDateTime64(1747983448000000000, 9) AS t_to, 1747945619 AS bucket_from, 1747983448 AS bucket_to, all_spans AS (SELECT *, resource_string_service$$name AS `service.name` FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_A AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint), A AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_A) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_B AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint), B AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_B) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), A_AND_B AS (SELECT l.* FROM A AS l INNER JOIN B AS r ON l.trace_id = r.trace_id) SELECT timestamp, trace_id, span_id, name, duration_nano, parent_span_id FROM A_AND_B ORDER BY timestamp DESC LIMIT ? SETTINGS distributed_product_mode='allow', max_memory_usage=10000000000",
Args: []any{"1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "frontend", "%service.name%", "%service.name\":\"frontend%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "backend", "%service.name%", "%service.name\":\"backend%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), 15},
},
expectedErr: nil,
@@ -178,7 +178,7 @@ func TestTraceOperatorStatementBuilder(t *testing.T) {
},
},
expected: qbtypes.Statement{
Query: "WITH toDateTime64(1747947419000000000, 9) AS t_from, toDateTime64(1747983448000000000, 9) AS t_to, 1747945619 AS bucket_from, 1747983448 AS bucket_to, all_spans AS (SELECT *, resource_string_service$$name AS `service.name` FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_A AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), A AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_A) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_B AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), B AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_B) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), A_OR_B AS (SELECT * FROM A UNION DISTINCT SELECT * FROM B) SELECT timestamp, trace_id, span_id, name, duration_nano, parent_span_id FROM A_OR_B ORDER BY timestamp DESC LIMIT ? SETTINGS distributed_product_mode='allow', max_memory_usage=10000000000",
Query: "WITH toDateTime64(1747947419000000000, 9) AS t_from, toDateTime64(1747983448000000000, 9) AS t_to, 1747945619 AS bucket_from, 1747983448 AS bucket_to, all_spans AS (SELECT *, resource_string_service$$name AS `service.name` FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_A AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint), A AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_A) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_B AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint), B AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_B) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), A_OR_B AS (SELECT * FROM A UNION DISTINCT SELECT * FROM B) SELECT timestamp, trace_id, span_id, name, duration_nano, parent_span_id FROM A_OR_B ORDER BY timestamp DESC LIMIT ? SETTINGS distributed_product_mode='allow', max_memory_usage=10000000000",
Args: []any{"1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "frontend", "%service.name%", "%service.name\":\"frontend%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "backend", "%service.name%", "%service.name\":\"backend%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), 20},
},
expectedErr: nil,
@@ -215,7 +215,7 @@ func TestTraceOperatorStatementBuilder(t *testing.T) {
},
},
expected: qbtypes.Statement{
Query: "WITH toDateTime64(1747947419000000000, 9) AS t_from, toDateTime64(1747983448000000000, 9) AS t_to, 1747945619 AS bucket_from, 1747983448 AS bucket_to, all_spans AS (SELECT *, resource_string_service$$name AS `service.name` FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_A AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), A AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_A) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_B AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), B AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_B) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), A_not_B AS (SELECT l.* FROM A AS l WHERE l.trace_id GLOBAL NOT IN (SELECT DISTINCT trace_id FROM B)) SELECT timestamp, trace_id, span_id, name, duration_nano, parent_span_id FROM A_not_B ORDER BY timestamp DESC LIMIT ? SETTINGS distributed_product_mode='allow', max_memory_usage=10000000000",
Query: "WITH toDateTime64(1747947419000000000, 9) AS t_from, toDateTime64(1747983448000000000, 9) AS t_to, 1747945619 AS bucket_from, 1747983448 AS bucket_to, all_spans AS (SELECT *, resource_string_service$$name AS `service.name` FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_A AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint), A AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_A) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_B AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint), B AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_B) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), A_not_B AS (SELECT l.* FROM A AS l WHERE l.trace_id GLOBAL NOT IN (SELECT DISTINCT trace_id FROM B)) SELECT timestamp, trace_id, span_id, name, duration_nano, parent_span_id FROM A_not_B ORDER BY timestamp DESC LIMIT ? SETTINGS distributed_product_mode='allow', max_memory_usage=10000000000",
Args: []any{"1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "frontend", "%service.name%", "%service.name\":\"frontend%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "backend", "%service.name%", "%service.name\":\"backend%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), 10},
},
expectedErr: nil,
@@ -264,7 +264,7 @@ func TestTraceOperatorStatementBuilder(t *testing.T) {
},
},
expected: qbtypes.Statement{
Query: "WITH toDateTime64(1747947419000000000, 9) AS t_from, toDateTime64(1747983448000000000, 9) AS t_to, 1747945619 AS bucket_from, 1747983448 AS bucket_to, all_spans AS (SELECT *, resource_string_service$$name AS `service.name` FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_A AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), A AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_A) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_B AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), B AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_B) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), A_DIR_DESC_B AS (SELECT p.* FROM A AS p INNER JOIN B AS c ON p.trace_id = c.trace_id AND p.span_id = c.parent_span_id) SELECT toStartOfInterval(timestamp, INTERVAL 60 SECOND) AS ts, toString(multiIf(multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL) IS NOT NULL, multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL), NULL)) AS `service.name`, count() AS __result_0 FROM A_DIR_DESC_B GROUP BY ts, `service.name` ORDER BY ts desc SETTINGS distributed_product_mode='allow', max_memory_usage=10000000000",
Query: "WITH toDateTime64(1747947419000000000, 9) AS t_from, toDateTime64(1747983448000000000, 9) AS t_to, 1747945619 AS bucket_from, 1747983448 AS bucket_to, all_spans AS (SELECT *, resource_string_service$$name AS `service.name` FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_A AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint), A AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_A) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_B AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint), B AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_B) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), A_DIR_DESC_B AS (SELECT p.* FROM A AS p INNER JOIN B AS c ON p.trace_id = c.trace_id AND p.span_id = c.parent_span_id) SELECT toStartOfInterval(timestamp, INTERVAL 60 SECOND) AS ts, toString(multiIf(multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL) IS NOT NULL, multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL), NULL)) AS `service.name`, count() AS __result_0 FROM A_DIR_DESC_B GROUP BY ts, `service.name` ORDER BY ts desc SETTINGS distributed_product_mode='allow', max_memory_usage=10000000000",
Args: []any{"1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "frontend", "%service.name%", "%service.name\":\"frontend%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "backend", "%service.name%", "%service.name\":\"backend%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448)},
},
expectedErr: nil,
@@ -323,7 +323,7 @@ func TestTraceOperatorStatementBuilder(t *testing.T) {
},
},
expected: qbtypes.Statement{
Query: "WITH toDateTime64(1747947419000000000, 9) AS t_from, toDateTime64(1747983448000000000, 9) AS t_to, 1747945619 AS bucket_from, 1747983448 AS bucket_to, all_spans AS (SELECT *, resource_string_service$$name AS `service.name` FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_A AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), A AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_A) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), B AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? AND toFloat64(response_status_code) < ?), A_AND_B AS (SELECT l.* FROM A AS l INNER JOIN B AS r ON l.trace_id = r.trace_id) SELECT toString(multiIf(multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL) IS NOT NULL, multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL), NULL)) AS `service.name`, avg(multiIf(duration_nano <> ?, duration_nano, NULL)) AS __result_0 FROM A_AND_B GROUP BY `service.name` ORDER BY __result_0 desc SETTINGS distributed_product_mode='allow', max_memory_usage=10000000000",
Query: "WITH toDateTime64(1747947419000000000, 9) AS t_from, toDateTime64(1747983448000000000, 9) AS t_to, 1747945619 AS bucket_from, 1747983448 AS bucket_to, all_spans AS (SELECT *, resource_string_service$$name AS `service.name` FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_A AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint), A AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_A) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), B AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ? AND toFloat64(response_status_code) < ?), A_AND_B AS (SELECT l.* FROM A AS l INNER JOIN B AS r ON l.trace_id = r.trace_id) SELECT toString(multiIf(multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL) IS NOT NULL, multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL), NULL)) AS `service.name`, avg(multiIf(duration_nano <> ?, duration_nano, NULL)) AS __result_0 FROM A_AND_B GROUP BY `service.name` ORDER BY __result_0 desc SETTINGS distributed_product_mode='allow', max_memory_usage=10000000000",
Args: []any{"1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "frontend", "%service.name%", "%service.name\":\"frontend%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), float64(400), 0},
},
expectedErr: nil,
@@ -380,7 +380,7 @@ func TestTraceOperatorStatementBuilder(t *testing.T) {
},
},
expected: qbtypes.Statement{
Query: "WITH toDateTime64(1747947419000000000, 9) AS t_from, toDateTime64(1747983448000000000, 9) AS t_to, 1747945619 AS bucket_from, 1747983448 AS bucket_to, all_spans AS (SELECT *, resource_string_service$$name AS `service.name` FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_A AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), A AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_A) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_B AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), B AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_B) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), A_DIR_DESC_B AS (SELECT p.* FROM A AS p INNER JOIN B AS c ON p.trace_id = c.trace_id AND p.span_id = c.parent_span_id), __resource_filter_C AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), C AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_C) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_D AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), D AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_D) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), C_DIR_DESC_D AS (SELECT p.* FROM C AS p INNER JOIN D AS c ON p.trace_id = c.trace_id AND p.span_id = c.parent_span_id), A_DIR_DESC_B_AND_C_DIR_DESC_D AS (SELECT l.* FROM A_DIR_DESC_B AS l INNER JOIN C_DIR_DESC_D AS r ON l.trace_id = r.trace_id) SELECT timestamp, trace_id, span_id, name, duration_nano, parent_span_id FROM A_DIR_DESC_B_AND_C_DIR_DESC_D ORDER BY timestamp DESC LIMIT ? SETTINGS distributed_product_mode='allow', max_memory_usage=10000000000",
Query: "WITH toDateTime64(1747947419000000000, 9) AS t_from, toDateTime64(1747983448000000000, 9) AS t_to, 1747945619 AS bucket_from, 1747983448 AS bucket_to, all_spans AS (SELECT *, resource_string_service$$name AS `service.name` FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_A AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint), A AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_A) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_B AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint), B AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_B) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), A_DIR_DESC_B AS (SELECT p.* FROM A AS p INNER JOIN B AS c ON p.trace_id = c.trace_id AND p.span_id = c.parent_span_id), __resource_filter_C AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint), C AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_C) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_D AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint), D AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_D) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), C_DIR_DESC_D AS (SELECT p.* FROM C AS p INNER JOIN D AS c ON p.trace_id = c.trace_id AND p.span_id = c.parent_span_id), A_DIR_DESC_B_AND_C_DIR_DESC_D AS (SELECT l.* FROM A_DIR_DESC_B AS l INNER JOIN C_DIR_DESC_D AS r ON l.trace_id = r.trace_id) SELECT timestamp, trace_id, span_id, name, duration_nano, parent_span_id FROM A_DIR_DESC_B_AND_C_DIR_DESC_D ORDER BY timestamp DESC LIMIT ? SETTINGS distributed_product_mode='allow', max_memory_usage=10000000000",
Args: []any{"1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "frontend", "%service.name%", "%service.name\":\"frontend%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "backend", "%service.name%", "%service.name\":\"backend%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "auth", "%service.name%", "%service.name\":\"auth%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "database", "%service.name%", "%service.name\":\"database%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), 5},
},
expectedErr: nil,
@@ -414,7 +414,7 @@ func TestTraceOperatorStatementBuilder(t *testing.T) {
},
},
expected: qbtypes.Statement{
Query: "WITH toDateTime64(1747947419000000000, 9) AS t_from, toDateTime64(1747983448000000000, 9) AS t_to, 1747945619 AS bucket_from, 1747983448 AS bucket_to, all_spans AS (SELECT *, resource_string_service$$name AS `service.name` FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_A AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), A AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_A) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_B AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), B AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_B) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), A_INDIR_DESC_B AS (WITH RECURSIVE up AS (SELECT d.trace_id, d.span_id, d.parent_span_id, 0 AS depth FROM B AS d UNION ALL SELECT p.trace_id, p.span_id, p.parent_span_id, up.depth + 1 FROM all_spans AS p JOIN up ON p.trace_id = up.trace_id AND p.span_id = up.parent_span_id WHERE up.depth < 100) SELECT DISTINCT a.* FROM A AS a GLOBAL INNER JOIN (SELECT DISTINCT trace_id, span_id FROM up WHERE depth > 0 ) AS ancestors ON ancestors.trace_id = a.trace_id AND ancestors.span_id = a.span_id), __return_from_B AS (SELECT * FROM B WHERE trace_id IN (SELECT DISTINCT trace_id FROM A_INDIR_DESC_B)) SELECT timestamp, trace_id, span_id, name, duration_nano, parent_span_id FROM __return_from_B ORDER BY timestamp DESC LIMIT ? SETTINGS distributed_product_mode='allow', max_memory_usage=10000000000",
Query: "WITH toDateTime64(1747947419000000000, 9) AS t_from, toDateTime64(1747983448000000000, 9) AS t_to, 1747945619 AS bucket_from, 1747983448 AS bucket_to, all_spans AS (SELECT *, resource_string_service$$name AS `service.name` FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_A AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint), A AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_A) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_B AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint), B AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_B) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), A_INDIR_DESC_B AS (WITH RECURSIVE up AS (SELECT d.trace_id, d.span_id, d.parent_span_id, 0 AS depth FROM B AS d UNION ALL SELECT p.trace_id, p.span_id, p.parent_span_id, up.depth + 1 FROM all_spans AS p JOIN up ON p.trace_id = up.trace_id AND p.span_id = up.parent_span_id WHERE up.depth < 100) SELECT DISTINCT a.* FROM A AS a GLOBAL INNER JOIN (SELECT DISTINCT trace_id, span_id FROM up WHERE depth > 0 ) AS ancestors ON ancestors.trace_id = a.trace_id AND ancestors.span_id = a.span_id), __return_from_B AS (SELECT * FROM B WHERE trace_id IN (SELECT DISTINCT trace_id FROM A_INDIR_DESC_B)) SELECT timestamp, trace_id, span_id, name, duration_nano, parent_span_id FROM __return_from_B ORDER BY timestamp DESC LIMIT ? SETTINGS distributed_product_mode='allow', max_memory_usage=10000000000",
Args: []any{"1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "gateway", "%service.name%", "%service.name\":\"gateway%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "database", "%service.name%", "%service.name\":\"database%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), 10},
},
expectedErr: nil,
@@ -456,7 +456,7 @@ func TestTraceOperatorStatementBuilder(t *testing.T) {
},
},
expected: qbtypes.Statement{
Query: "WITH toDateTime64(1747947419000000000, 9) AS t_from, toDateTime64(1747983448000000000, 9) AS t_to, 1747945619 AS bucket_from, 1747983448 AS bucket_to, all_spans AS (SELECT *, resource_string_service$$name AS `service.name` FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_A AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), A AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_A) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_B AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), B AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_B) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), A_INDIR_DESC_B AS (WITH RECURSIVE up AS (SELECT d.trace_id, d.span_id, d.parent_span_id, 0 AS depth FROM B AS d UNION ALL SELECT p.trace_id, p.span_id, p.parent_span_id, up.depth + 1 FROM all_spans AS p JOIN up ON p.trace_id = up.trace_id AND p.span_id = up.parent_span_id WHERE up.depth < 100) SELECT DISTINCT a.* FROM A AS a GLOBAL INNER JOIN (SELECT DISTINCT trace_id, span_id FROM up WHERE depth > 0 ) AS ancestors ON ancestors.trace_id = a.trace_id AND ancestors.span_id = a.span_id), __resource_filter_C AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), C AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_C) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), A_INDIR_DESC_B_AND_C AS (SELECT l.* FROM A_INDIR_DESC_B AS l INNER JOIN C AS r ON l.trace_id = r.trace_id), __return_from_C AS (SELECT * FROM C WHERE trace_id IN (SELECT DISTINCT trace_id FROM A_INDIR_DESC_B_AND_C)) SELECT timestamp, trace_id, span_id, name, duration_nano, parent_span_id FROM __return_from_C ORDER BY timestamp DESC LIMIT ? SETTINGS distributed_product_mode='allow', max_memory_usage=10000000000",
Query: "WITH toDateTime64(1747947419000000000, 9) AS t_from, toDateTime64(1747983448000000000, 9) AS t_to, 1747945619 AS bucket_from, 1747983448 AS bucket_to, all_spans AS (SELECT *, resource_string_service$$name AS `service.name` FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_A AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint), A AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_A) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_B AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint), B AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_B) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), A_INDIR_DESC_B AS (WITH RECURSIVE up AS (SELECT d.trace_id, d.span_id, d.parent_span_id, 0 AS depth FROM B AS d UNION ALL SELECT p.trace_id, p.span_id, p.parent_span_id, up.depth + 1 FROM all_spans AS p JOIN up ON p.trace_id = up.trace_id AND p.span_id = up.parent_span_id WHERE up.depth < 100) SELECT DISTINCT a.* FROM A AS a GLOBAL INNER JOIN (SELECT DISTINCT trace_id, span_id FROM up WHERE depth > 0 ) AS ancestors ON ancestors.trace_id = a.trace_id AND ancestors.span_id = a.span_id), __resource_filter_C AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ? GROUP BY fingerprint), C AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_C) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), A_INDIR_DESC_B_AND_C AS (SELECT l.* FROM A_INDIR_DESC_B AS l INNER JOIN C AS r ON l.trace_id = r.trace_id), __return_from_C AS (SELECT * FROM C WHERE trace_id IN (SELECT DISTINCT trace_id FROM A_INDIR_DESC_B_AND_C)) SELECT timestamp, trace_id, span_id, name, duration_nano, parent_span_id FROM __return_from_C ORDER BY timestamp DESC LIMIT ? SETTINGS distributed_product_mode='allow', max_memory_usage=10000000000",
Args: []any{"1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "gateway", "%service.name%", "%service.name\":\"gateway%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "database", "%service.name%", "%service.name\":\"database%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "auth", "%service.name%", "%service.name\":\"auth%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), 10},
},
expectedErr: nil,
@@ -478,6 +478,8 @@ func TestTraceOperatorStatementBuilder(t *testing.T) {
aggExprRewriter,
nil,
fl,
false,
100000,
)
statementBuilder := NewTraceOperatorStatementBuilder(
@@ -594,6 +596,8 @@ func TestTraceOperatorStatementBuilderErrors(t *testing.T) {
aggExprRewriter,
nil,
fl,
false,
100000,
)
statementBuilder := NewTraceOperatorStatementBuilder(

View File

@@ -5,13 +5,13 @@ import (
"strings"
"testing"
"github.com/SigNoz/signoz/pkg/flagger/flaggertest"
"github.com/SigNoz/signoz/pkg/instrumentation/instrumentationtest"
"github.com/SigNoz/signoz/pkg/querybuilder"
qbtypes "github.com/SigNoz/signoz/pkg/types/querybuildertypes/querybuildertypesv5"
"github.com/SigNoz/signoz/pkg/types/telemetrytypes"
"github.com/SigNoz/signoz/pkg/types/telemetrytypes/telemetrytypestest"
"github.com/stretchr/testify/assert"
"github.com/SigNoz/signoz/pkg/flagger/flaggertest"
"github.com/stretchr/testify/require"
)
@@ -44,8 +44,10 @@ func TestTraceTimeRangeOptimization(t *testing.T) {
fm,
cb,
aggExprRewriter,
nil, // telemetryStore is nil - optimization won't happen but code path is tested
nil, // telemetryStore is nil - adaptive path is disabled
fl,
false,
100000,
)
tests := []struct {

View File

@@ -18,16 +18,14 @@ var (
type StorableIntegrationDashboard struct {
bun.BaseModel `bun:"table:integration_dashboard"`
ID string `json:"id" bun:"id,pk,type:text"`
DashboardID string `json:"dashboardId" bun:"dashboard_id,type:text"`
Provider IntegrationDashboardProviderType `json:"provider" bun:"provider,type:text"`
Slug string `json:"slug" bun:"slug,type:text"`
CreatedAt time.Time `json:"createdAt" bun:"created_at"`
UpdatedAt time.Time `json:"updatedAt" bun:"updated_at"`
ID string `bun:"id,pk,type:text"`
DashboardID string `bun:"dashboard_id,type:text"`
Provider IntegrationDashboardProviderType `bun:"provider,type:text"`
Slug string `bun:"slug,type:text"`
CreatedAt time.Time `bun:"created_at"`
UpdatedAt time.Time `bun:"updated_at"`
}
type IntegrationDashboard = StorableIntegrationDashboard
func NewStorableIntegrationDashboard(dashboardID string, provider IntegrationDashboardProviderType, slug string) *StorableIntegrationDashboard {
now := time.Now()
return &StorableIntegrationDashboard{

View File

@@ -50,18 +50,10 @@ type ListServicesMetadataParams struct {
// Service represents a cloud integration service with its definition,
// cloud integration service is non nil only when the service entry exists in DB with ANY config (enabled or disabled).
type Service struct {
ServiceDefinitionMetadata
Overview string `json:"overview" required:"true"` // markdown
ServiceAssets ServiceAssets `json:"assets" required:"true"`
SupportedSignals SupportedSignals `json:"supportedSignals" required:"true"`
DataCollected DataCollected `json:"dataCollected" required:"true"`
ServiceDefinition
CloudIntegrationService *CloudIntegrationService `json:"cloudIntegrationService" required:"true" nullable:"true"`
}
type ServiceAssets struct {
Dashboards []*ServiceDashboard `json:"dashboards" required:"true" nullable:"false"`
}
type GetServiceParams struct {
CloudIntegrationID valuer.UUID `query:"cloud_integration_id" required:"false"`
}
@@ -129,12 +121,6 @@ type Dashboard struct {
Definition dashboardtypes.StorableDashboardData `json:"definition,omitempty"`
}
type ServiceDashboard struct {
Title string `json:"title"`
Description string `json:"description"`
IntegrationDashboard *IntegrationDashboard `json:"integrationDashboard,omitempty" required:"false"`
}
func NewCloudIntegrationService(serviceID ServiceID, cloudIntegrationID valuer.UUID, provider CloudProviderType, config *ServiceConfig) (*CloudIntegrationService, error) {
switch provider {
case CloudProviderTypeAWS:
@@ -178,41 +164,11 @@ func NewServiceMetadata(definition ServiceDefinition, enabled bool) *ServiceMeta
}
}
func NewService(provider CloudProviderType, def *ServiceDefinition, integrationService *CloudIntegrationService, integrationDashboards []*StorableIntegrationDashboard) *Service {
service := &Service{
ServiceDefinitionMetadata: def.ServiceDefinitionMetadata,
Overview: def.Overview,
SupportedSignals: def.SupportedSignals,
DataCollected: def.DataCollected,
CloudIntegrationService: integrationService,
ServiceAssets: ServiceAssets{Dashboards: make([]*ServiceDashboard, 0, len(def.Assets.Dashboards))},
func NewService(def ServiceDefinition, storableService *CloudIntegrationService) *Service {
return &Service{
ServiceDefinition: def,
CloudIntegrationService: storableService,
}
integrationDashboardsMap := make(map[string]*IntegrationDashboard)
for _, d := range integrationDashboards {
integrationDashboardsMap[d.Slug] = d
}
for _, d := range def.Assets.Dashboards {
dashboard := &ServiceDashboard{
Title: d.Title,
Description: d.Description,
}
if integrationService != nil {
slug := CloudIntegrationDashboardSlug(provider, integrationService.Type, d.ID)
if integrationDashboard, exists := integrationDashboardsMap[slug]; exists {
if integrationDashboard != nil {
dashboard.IntegrationDashboard = integrationDashboard
}
}
}
service.ServiceAssets.Dashboards = append(service.ServiceAssets.Dashboards, dashboard)
}
return service
}
func NewGettableServicesMetadata(services []*ServiceMetadata) *GettableServicesMetadata {

View File

@@ -0,0 +1,9 @@
package querybuildertypesv5
type ResourceFilterResolveKind int
const (
ResourceFilterResolveKindNoOp ResourceFilterResolveKind = iota
ResourceFilterResolveKindUseCTE
ResourceFilterResolveKindFallback
)

View File

@@ -28,6 +28,42 @@ class TelemetryFieldKey:
}
class RequestType:
RAW = "raw"
TIME_SERIES = "time_series"
SCALAR = "scalar"
TABLE = "table"
@dataclass
class Aggregation:
expression: str
alias: str | None = None
def to_dict(self) -> dict:
agg: dict[str, Any] = {"expression": self.expression}
if self.alias:
agg["alias"] = self.alias
return agg
@dataclass
class MetricAggregation:
metric_name: str
time_aggregation: str
space_aggregation: str
temporality: str = "cumulative"
def to_dict(self) -> dict:
agg: dict[str, Any] = {
"metricName": self.metric_name,
"timeAggregation": self.time_aggregation,
"spaceAggregation": self.space_aggregation,
"temporality": self.temporality,
}
return agg
@dataclass
class OrderBy:
key: TelemetryFieldKey
@@ -46,6 +82,8 @@ class BuilderQuery:
filter_expression: str | None = None
select_fields: list[TelemetryFieldKey] | None = None
order: list[OrderBy] | None = None
aggregations: list[Aggregation | MetricAggregation] | None = None
step_interval: int | None = None
def to_dict(self) -> dict:
spec: dict[str, Any] = {
@@ -62,6 +100,11 @@ class BuilderQuery:
spec["selectFields"] = [f.to_dict() for f in self.select_fields]
if self.order:
spec["order"] = [o.to_dict() if hasattr(o, "to_dict") else o for o in self.order]
if self.aggregations:
spec["aggregations"] = [agg.to_dict() if hasattr(agg, "to_dict") else agg for agg in self.aggregations]
if self.step_interval is not None:
spec["stepInterval"] = self.step_interval
return {"type": "builder_query", "spec": spec}
@@ -117,7 +160,7 @@ def make_query_request(
end_ms: int,
queries: list[dict],
*,
request_type: str = "time_series",
request_type: str = RequestType.TIME_SERIES,
format_options: dict | None = None,
variables: dict | None = None,
no_cache: bool = True,

View File

@@ -15,7 +15,6 @@ from fixtures.querier import (
build_group_by_field,
build_logs_aggregation,
build_order_by,
build_raw_query,
build_scalar_query,
find_named_result,
index_series_by_label,
@@ -2626,43 +2625,3 @@ def test_logs_aggregation_filter_by_trace_id(
orphan_count, orphan_warnings = _count(narrow_start_ms, now_ms, orphan_trace_id)
assert orphan_count == 1, f"Expected count=1 for orphan trace_id aggregation, got {orphan_count} — query may have been incorrectly short-circuited"
assert not any(outside_range_msg in m for m in orphan_warnings), f"Did not expect outside-range warning for orphan trace_id, got {orphan_warnings}"
def test_logs_list_ambigous_warnings(
signoz: types.SigNoz,
create_user_admin: None, # pylint: disable=unused-argument
get_token: Callable[[str, str], str],
insert_logs: Callable[[list[Logs]], None],
) -> None:
insert_logs(
[
Logs(
timestamp=datetime.now(tz=UTC) - timedelta(seconds=1),
resources={
"service.name": "java",
},
attributes={
"service.name": "java",
},
body="This is a log message, coming from a java application",
severity_text="DEBUG",
),
]
)
response = make_query_request(
signoz,
get_token(USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD),
start_ms=int((datetime.now(tz=UTC) - timedelta(minutes=1)).timestamp() * 1000),
end_ms=int(datetime.now(tz=UTC).timestamp() * 1000),
request_type="raw",
queries=[build_raw_query(name="A", signal="logs", filter_expression='service.name = "java"')],
)
assert response.status_code == HTTPStatus.OK
assert response.json()["status"] == "success"
warning = response.json()["data"].get("warning", None)
assert warning is not None
assert warning["message"] == "Encountered warnings"
assert len(warning.get("warnings")) > 0
assert any(["ambiguous" in w["message"] for w in warning.get("warnings")])

View File

@@ -0,0 +1,386 @@
from collections.abc import Callable
from datetime import UTC, datetime, timedelta
from http import HTTPStatus
from fixtures import types
from fixtures.auth import USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD
from fixtures.logs import Logs
from fixtures.meter import MeterSample, make_meter_samples
from fixtures.querier import (
Aggregation,
BuilderQuery,
MetricAggregation,
RequestType,
make_query_request,
)
def test_resource_default_warning(
signoz: types.SigNoz,
create_user_admin: None, # pylint: disable=unused-argument
get_token: Callable[[str, str], str],
insert_logs: Callable[[list[Logs]], None],
) -> None:
insert_logs(
[
Logs(
timestamp=datetime.now(tz=UTC),
resources={
"service.name": "java",
},
attributes={
"service.name": "java",
},
body="This is a log message, coming from a java application",
severity_text="DEBUG",
),
]
)
token = get_token(USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD)
response = make_query_request(
signoz,
token,
start_ms=int((datetime.now(tz=UTC) - timedelta(minutes=20)).timestamp() * 1000),
end_ms=int(datetime.now(tz=UTC).timestamp() * 1000),
request_type=RequestType.RAW,
queries=[
BuilderQuery(
name="A",
signal="logs",
filter_expression="service.name = 'java'",
).to_dict()
],
)
assert response.status_code == HTTPStatus.OK
assert response.json()["status"] == "success"
warning = response.json()["data"].get("warning", None)
assert warning is not None
assert warning["message"] == "Encountered warnings"
expected_service_name_warning = (
"Key `service.name` is ambiguous, found 2 different combinations of "
"field context / data type: [name=service.name,context=resource,datatype=string "
"name=service.name,context=attribute,datatype=string]. Using `resource` context "
"by default. To query attributes explicitly, use the fully qualified name "
"(e.g., 'attribute.service.name')"
)
assert warning["warnings"] == [
{"message": expected_service_name_warning},
]
def test_key_collision_warning(
signoz: types.SigNoz,
create_user_admin: None, # pylint: disable=unused-argument
get_token: Callable[[str, str], str],
insert_logs: Callable[[list[Logs]], None],
) -> None:
insert_logs(
[
Logs(
timestamp=datetime.now(tz=UTC),
resources={
"service.name": "java",
},
attributes={
"service.name": "java",
"http.status_code": 200,
},
body="This is a log message, coming from a java application",
severity_text="DEBUG",
),
Logs(
timestamp=datetime.now(tz=UTC),
resources={
"service.name": "java",
},
attributes={
"service.name": "java",
"http.status_code": "200",
},
body="This is a log message, coming from a java application",
severity_text="DEBUG",
),
]
)
token = get_token(USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD)
response = make_query_request(
signoz,
token,
start_ms=int((datetime.now(tz=UTC) - timedelta(minutes=20)).timestamp() * 1000),
end_ms=int(datetime.now(tz=UTC).timestamp() * 1000),
request_type=RequestType.RAW,
queries=[
BuilderQuery(
name="A",
signal="logs",
filter_expression="http.status_code = '200'",
).to_dict()
],
)
assert response.status_code == HTTPStatus.OK
assert response.json()["status"] == "success"
warning = response.json()["data"].get("warning", None)
assert warning is not None
assert warning["message"] == "Encountered warnings"
expected_http_status_code_warning = "Key `http.status_code` is ambiguous, found 2 different combinations of field context / data type: [name=http.status_code,context=attribute,datatype=number name=http.status_code,context=attribute,datatype=string]."
assert warning["warnings"] == [
{"message": expected_http_status_code_warning},
]
def test_deduped_warnings_for_single_query(
signoz: types.SigNoz,
create_user_admin: None, # pylint: disable=unused-argument
get_token: Callable[[str, str], str],
insert_logs: Callable[[list[Logs]], None],
) -> None:
insert_logs(
[
*[
Logs(
timestamp=datetime.now(tz=UTC) - timedelta(minutes=i * 10),
resources={
"service.name": "java",
},
attributes={
"service.name": "java",
"http.status_code": 200,
},
body="This is a log message, coming from a java application",
severity_text="DEBUG",
)
for i in range(10)
],
Logs(
timestamp=datetime.now(tz=UTC),
resources={
"service.name": "java",
},
attributes={
"service.name": "java",
"http.status_code": "200",
},
body="This is a log message, coming from a java application",
severity_text="DEBUG",
),
]
)
token = get_token(USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD)
# `service.name` (resource vs attribute) and `http.status_code` (number vs
# string) are both ambiguous keys, so referencing them in the filter makes
# the querier emit an "is ambiguous" warning while *building* the query.
#
# This test targets the bucket-cache warning-merge path: the cache stores a
# query's warnings alongside its buckets, and on a partial cache hit
# executeWithCache merges the cached warnings with the warnings re-emitted by
# each freshly executed missing range (querier.mergeResults). Before the
# dedup fix the same message appeared once per executed range.
query = BuilderQuery(
name="A",
signal="logs",
step_interval=600,
filter_expression="service.name = 'java' and http.status_code = 200",
aggregations=[Aggregation(expression="count()")],
).to_dict()
# Anchor both windows to a single "now" so their step-aligned boundaries
# match. Both windows end well before the 5m flux interval so the results
# are cacheable.
now_ms = int(datetime.now(tz=UTC).timestamp() * 1000)
minute = 60 * 1000
# First request populates the cache for [-90m, -30m], storing the warnings.
first = make_query_request(
signoz,
token,
start_ms=now_ms - 90 * minute,
end_ms=now_ms - 30 * minute,
request_type=RequestType.TIME_SERIES,
queries=[query],
no_cache=False,
)
assert first.status_code == HTTPStatus.OK
assert first.json()["status"] == "success"
# Second request reuses the cached [-90m, -30m] buckets (which carry the
# cached warnings) and executes only the trailing [-30m, -20m] step fresh,
# which re-emits the same warnings — exercising the cache/fresh merge.
response = make_query_request(
signoz,
token,
start_ms=now_ms - 90 * minute,
end_ms=now_ms - 20 * minute,
request_type=RequestType.TIME_SERIES,
queries=[query],
no_cache=False,
)
assert response.status_code == HTTPStatus.OK
assert response.json()["status"] == "success"
warning = response.json()["data"].get("warning", None)
assert warning is not None
assert warning["message"] == "Encountered warnings"
# Each ambiguity warning arrives from both the cached portion and the fresh
# missing range; after deduplication each distinct message appears once.
expected_service_name_warning = (
"Key `service.name` is ambiguous, found 2 different combinations of "
"field context / data type: [name=service.name,context=resource,datatype=string "
"name=service.name,context=attribute,datatype=string]. Using `resource` context "
"by default. To query attributes explicitly, use the fully qualified name "
"(e.g., 'attribute.service.name')"
)
expected_status_code_warning = "Key `http.status_code` is ambiguous, found 2 different combinations of field context / data type: [name=http.status_code,context=attribute,datatype=number name=http.status_code,context=attribute,datatype=string]."
assert warning["warnings"] == [
{"message": expected_service_name_warning},
{"message": expected_status_code_warning},
]
def test_deduped_warnings_for_multiple_queries(
signoz: types.SigNoz,
create_user_admin: None, # pylint: disable=unused-argument
get_token: Callable[[str, str], str],
insert_logs: Callable[[list[Logs]], None],
) -> None:
insert_logs(
[
*[
Logs(
timestamp=datetime.now(tz=UTC) - timedelta(minutes=i * 10),
resources={
"service.name": "java",
},
attributes={
"service.name": "java",
"http.status_code": 200,
},
body="This is a log message, coming from a java application",
severity_text="DEBUG",
)
for i in range(10)
],
Logs(
timestamp=datetime.now(tz=UTC),
resources={
"service.name": "java",
},
attributes={
"service.name": "java",
"http.status_code": "200",
},
body="This is a log message, coming from a java application",
severity_text="DEBUG",
),
]
)
token = get_token(USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD)
# `service.name` (resource vs attribute) and `http.status_code` (number vs
# string) are both ambiguous keys, so referencing them in the filter makes
# the querier emit an "is ambiguous" warning while *building* the query.
query_1 = BuilderQuery(
name="A",
signal="logs",
step_interval=600,
filter_expression="service.name = 'java' and http.status_code = 200",
aggregations=[Aggregation(expression="count()")],
).to_dict()
query_2 = BuilderQuery(
name="B",
signal="logs",
step_interval=600,
filter_expression="service.name != '_java' and http.status_code = 200",
aggregations=[Aggregation(expression="count()")],
).to_dict()
now_ms = int(datetime.now(tz=UTC).timestamp() * 1000)
minute = 60 * 1000
response = make_query_request(
signoz,
token,
start_ms=now_ms - 90 * minute,
end_ms=now_ms - 20 * minute,
request_type=RequestType.TIME_SERIES,
queries=[query_1, query_2],
no_cache=False,
)
assert response.status_code == HTTPStatus.OK
assert response.json()["status"] == "success"
warning = response.json()["data"].get("warning", None)
assert warning is not None
assert warning["message"] == "Encountered warnings"
expected_service_name_warning = (
"Key `service.name` is ambiguous, found 2 different combinations of "
"field context / data type: [name=service.name,context=resource,datatype=string "
"name=service.name,context=attribute,datatype=string]. Using `resource` context "
"by default. To query attributes explicitly, use the fully qualified name "
"(e.g., 'attribute.service.name')"
)
expected_status_code_warning = "Key `http.status_code` is ambiguous, found 2 different combinations of field context / data type: [name=http.status_code,context=attribute,datatype=number name=http.status_code,context=attribute,datatype=string]."
assert warning["warnings"] == [
{"message": expected_service_name_warning},
{"message": expected_status_code_warning},
]
def test_no_warnings_for_meter_query(
signoz: types.SigNoz,
create_user_admin: None, # pylint: disable=unused-argument
get_token: Callable[[str, str], str],
insert_meter_samples: Callable[[list[MeterSample]], None],
) -> None:
# Meter queries deliberately suppress the step-interval clamp warning. The
# minimum allowed step for a meter is 1h, so a 10m (600s) step over this
# range would normally clamp and emit a warning for a regular metric — for a
# meter source the querier discards it. This asserts no warning leaks out.
now = datetime.now(tz=UTC).replace(second=0, microsecond=0)
metric_name = "signoz.meter.log.size"
insert_meter_samples(
make_meter_samples(
metric_name,
{"service": "test-service"},
now,
count=60,
temporality="Delta",
type_="Sum",
is_monotonic=True,
)
)
token = get_token(USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD)
response = make_query_request(
signoz,
token,
start_ms=int((now - timedelta(minutes=20)).timestamp() * 1000),
end_ms=int(now.timestamp() * 1000),
request_type=RequestType.TIME_SERIES,
queries=[
BuilderQuery(
name="A",
signal="metrics",
source="meter",
step_interval=600,
aggregations=[MetricAggregation(metric_name=metric_name, time_aggregation="sum", space_aggregation="sum")],
).to_dict()
],
)
assert response.status_code == HTTPStatus.OK
assert response.json()["status"] == "success"
assert "warning" not in response.json()["data"]

View File

@@ -0,0 +1,191 @@
"""
End-to-end coverage for the skip_resource_fingerprint querier optimization.
The conftest in this package boots SigNoz with:
- skip_resource_fingerprint.enabled = true
- skip_resource_fingerprint.threshold = 2
With that configuration the two non-trivial resolver branches are reachable
from a single SigNoz instance:
- count < 2 -> the resolver attaches the fingerprint CTE (same shape as the
legacy path; cheap because the fingerprint set is small).
- count >= 2 -> fallback path: no fingerprint subquery, resource conditions
are evaluated directly on the main spans table.
These tests assert end-to-end correctness — the optimization must be
semantically transparent.
"""
from collections.abc import Callable
from datetime import UTC, datetime, timedelta
from http import HTTPStatus
from fixtures import types
from fixtures.auth import USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD
from fixtures.querier import make_query_request
from fixtures.traces import TraceIdGenerator, Traces, TracesKind, TracesStatusCode
def _span(
*,
timestamp: datetime,
service_name: str,
name: str = "span",
duration_seconds: float = 1.0,
extra_resources: dict | None = None,
attributes: dict | None = None,
) -> Traces:
resources = {"service.name": service_name}
if extra_resources:
resources.update(extra_resources)
return Traces(
timestamp=timestamp,
duration=timedelta(seconds=duration_seconds),
trace_id=TraceIdGenerator.trace_id(),
span_id=TraceIdGenerator.span_id(),
parent_span_id="",
name=name,
kind=TracesKind.SPAN_KIND_SERVER,
status_code=TracesStatusCode.STATUS_CODE_OK,
status_message="",
resources=resources,
attributes=attributes or {},
)
def test_skip_resource_fingerprint_use_cte_path(
signoz: types.SigNoz,
create_user_admin: None, # pylint: disable=unused-argument
get_token: Callable[[str, str], str],
insert_traces: Callable[[list[Traces]], None],
) -> None:
"""
A filter that matches a single unique resource fingerprint (count = 1 < 2)
keeps the legacy CTE attached. The query should still return only the rows
belonging to that resource.
"""
now = datetime.now(tz=UTC).replace(second=0, microsecond=0)
insert_traces(
[
_span(timestamp=now - timedelta(seconds=10), service_name="skip-cte-svc", name="span-1"),
_span(timestamp=now - timedelta(seconds=8), service_name="skip-cte-svc", name="span-2"),
# Noise from a different resource — must be filtered out.
_span(timestamp=now - timedelta(seconds=6), service_name="skip-cte-noise", name="span-noise"),
]
)
token = get_token(USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD)
response = make_query_request(
signoz,
token,
start_ms=int((datetime.now(tz=UTC) - timedelta(minutes=5)).timestamp() * 1000),
end_ms=int(datetime.now(tz=UTC).timestamp() * 1000),
request_type="raw",
queries=[
{
"type": "builder_query",
"spec": {
"name": "A",
"signal": "traces",
"limit": 50,
"order": [{"key": {"name": "timestamp"}, "direction": "asc"}],
"filter": {"expression": "service.name = 'skip-cte-svc'"},
"selectFields": [
{
"name": "service.name",
"fieldDataType": "string",
"fieldContext": "resource",
},
{"name": "name", "fieldContext": "span"},
],
"aggregations": [{"expression": "count()"}],
},
}
],
)
assert response.status_code == HTTPStatus.OK
assert response.json()["status"] == "success"
rows = response.json()["data"]["data"]["results"][0]["rows"]
assert len(rows) == 2, f"expected only the 2 'skip-cte-svc' spans, got {len(rows)}"
names = [row["data"]["name"] for row in rows]
assert names == ["span-1", "span-2"]
services = {row["data"]["service.name"] for row in rows}
assert services == {"skip-cte-svc"}
def test_skip_resource_fingerprint_fallback_path(
signoz: types.SigNoz,
create_user_admin: None, # pylint: disable=unused-argument
get_token: Callable[[str, str], str],
insert_traces: Callable[[list[Traces]], None],
) -> None:
"""
A filter that matches multiple unique resource fingerprints (count >= 2)
drives the resolver down the fallback path: no fingerprint subquery, and
the resource condition is evaluated directly on the main spans table.
The result must still be correct (no over- or under-matching).
"""
now = datetime.now(tz=UTC).replace(second=0, microsecond=0)
# 3 services share the same deployment.environment, so the resource filter
# selects 3 fingerprints, exceeding our threshold of 2.
fallback_env = {"deployment.environment": "skip-fallback"}
insert_traces(
[
_span(timestamp=now - timedelta(seconds=10), service_name="skip-fb-svc-a", extra_resources=fallback_env),
_span(timestamp=now - timedelta(seconds=9), service_name="skip-fb-svc-b", extra_resources=fallback_env),
_span(timestamp=now - timedelta(seconds=8), service_name="skip-fb-svc-c", extra_resources=fallback_env),
# Noise without the fallback env — must be filtered out.
_span(
timestamp=now - timedelta(seconds=7),
service_name="skip-fb-other",
extra_resources={"deployment.environment": "skip-other"},
),
]
)
token = get_token(USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD)
response = make_query_request(
signoz,
token,
start_ms=int((datetime.now(tz=UTC) - timedelta(minutes=5)).timestamp() * 1000),
end_ms=int(datetime.now(tz=UTC).timestamp() * 1000),
request_type="raw",
queries=[
{
"type": "builder_query",
"spec": {
"name": "A",
"signal": "traces",
"limit": 50,
"order": [{"key": {"name": "timestamp"}, "direction": "asc"}],
"filter": {"expression": "deployment.environment = 'skip-fallback'"},
"selectFields": [
{
"name": "service.name",
"fieldDataType": "string",
"fieldContext": "resource",
},
],
"aggregations": [{"expression": "count()"}],
},
}
],
)
assert response.status_code == HTTPStatus.OK
assert response.json()["status"] == "success"
rows = response.json()["data"]["data"]["results"][0]["rows"]
assert len(rows) == 3, f"expected 3 spans tagged with skip-fallback, got {len(rows)}"
services = sorted(row["data"]["service.name"] for row in rows)
assert services == ["skip-fb-svc-a", "skip-fb-svc-b", "skip-fb-svc-c"]

View File

@@ -0,0 +1,175 @@
"""
End-to-end coverage for the skip_resource_fingerprint optimization on logs.
The conftest boots SigNoz with threshold=2, so:
- count < 2 -> resolver attaches the fingerprint CTE (same shape as legacy).
- count >= 2 -> fallback: resource conditions evaluated on the main logs table.
Both branches must return the same correct rows.
"""
from collections.abc import Callable
from datetime import UTC, datetime, timedelta
from http import HTTPStatus
from fixtures import types
from fixtures.auth import USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD
from fixtures.logs import Logs
from fixtures.querier import make_query_request
def _log(
*,
timestamp: datetime,
service_name: str,
body: str,
extra_resources: dict | None = None,
) -> Logs:
resources = {"service.name": service_name}
if extra_resources:
resources.update(extra_resources)
return Logs(
timestamp=timestamp,
resources=resources,
attributes={},
body=body,
severity_text="INFO",
)
def test_skip_resource_fingerprint_logs_use_cte_path(
signoz: types.SigNoz,
create_user_admin: None, # pylint: disable=unused-argument
get_token: Callable[[str, str], str],
insert_logs: Callable[[list[Logs]], None],
) -> None:
"""
A filter matching a single resource fingerprint (count = 1 < 2) keeps the
legacy CTE attached. The result must only include rows for that resource.
"""
now = datetime.now(tz=UTC)
insert_logs(
[
_log(timestamp=now - timedelta(seconds=10), service_name="skip-logs-cte-svc", body="log-1"),
_log(timestamp=now - timedelta(seconds=8), service_name="skip-logs-cte-svc", body="log-2"),
# Noise from a different resource — must not appear.
_log(timestamp=now - timedelta(seconds=6), service_name="skip-logs-cte-noise", body="noise"),
]
)
token = get_token(USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD)
response = make_query_request(
signoz,
token,
start_ms=int((datetime.now(tz=UTC) - timedelta(minutes=5)).timestamp() * 1000),
end_ms=int(datetime.now(tz=UTC).timestamp() * 1000),
request_type="raw",
queries=[
{
"type": "builder_query",
"spec": {
"name": "A",
"signal": "logs",
"limit": 50,
"order": [{"key": {"name": "timestamp"}, "direction": "asc"}],
"filter": {"expression": "service.name = 'skip-logs-cte-svc'"},
"selectFields": [
{
"name": "service.name",
"fieldDataType": "string",
"fieldContext": "resource",
},
{"name": "body"},
],
"aggregations": [{"expression": "count()"}],
},
}
],
)
assert response.status_code == HTTPStatus.OK
assert response.json()["status"] == "success"
rows = response.json()["data"]["data"]["results"][0]["rows"]
assert len(rows) == 2, f"expected 2 'skip-logs-cte-svc' rows, got {len(rows)}"
bodies = [row["data"]["body"] for row in rows]
assert bodies == ["log-1", "log-2"]
services = {row["data"]["service.name"] for row in rows}
assert services == {"skip-logs-cte-svc"}
def test_skip_resource_fingerprint_logs_fallback_path(
signoz: types.SigNoz,
create_user_admin: None, # pylint: disable=unused-argument
get_token: Callable[[str, str], str],
insert_logs: Callable[[list[Logs]], None],
) -> None:
"""
A filter matching multiple resource fingerprints (count >= 2) drives the
fallback path: no CTE, resource conditions evaluated on the main logs
table. Result must still be correct (no over- or under-matching).
"""
now = datetime.now(tz=UTC)
fallback_env = {"deployment.environment": "skip-logs-fallback"}
insert_logs(
[
_log(timestamp=now - timedelta(seconds=10), service_name="skip-logs-fb-svc-a", body="a", extra_resources=fallback_env),
_log(timestamp=now - timedelta(seconds=9), service_name="skip-logs-fb-svc-b", body="b", extra_resources=fallback_env),
_log(timestamp=now - timedelta(seconds=8), service_name="skip-logs-fb-svc-c", body="c", extra_resources=fallback_env),
# Noise without the fallback env — must be filtered out.
_log(
timestamp=now - timedelta(seconds=7),
service_name="skip-logs-fb-other",
body="noise",
extra_resources={"deployment.environment": "skip-logs-other"},
),
]
)
token = get_token(USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD)
response = make_query_request(
signoz,
token,
start_ms=int((datetime.now(tz=UTC) - timedelta(minutes=5)).timestamp() * 1000),
end_ms=int(datetime.now(tz=UTC).timestamp() * 1000),
request_type="raw",
queries=[
{
"type": "builder_query",
"spec": {
"name": "A",
"signal": "logs",
"limit": 50,
"order": [{"key": {"name": "timestamp"}, "direction": "asc"}],
"filter": {"expression": "deployment.environment = 'skip-logs-fallback'"},
"selectFields": [
{
"name": "service.name",
"fieldDataType": "string",
"fieldContext": "resource",
},
{"name": "body"},
],
"aggregations": [{"expression": "count()"}],
},
}
],
)
assert response.status_code == HTTPStatus.OK
assert response.json()["status"] == "success"
rows = response.json()["data"]["data"]["results"][0]["rows"]
assert len(rows) == 3, f"expected 3 fallback rows, got {len(rows)}"
services = sorted(row["data"]["service.name"] for row in rows)
assert services == ["skip-logs-fb-svc-a", "skip-logs-fb-svc-b", "skip-logs-fb-svc-c"]
bodies = sorted(row["data"]["body"] for row in rows)
assert bodies == ["a", "b", "c"]

View File

@@ -0,0 +1,37 @@
import pytest
from testcontainers.core.container import Network
from fixtures import types
from fixtures.signoz import create_signoz
@pytest.fixture(name="signoz", scope="package")
def signoz_skip_resource_fingerprint(
network: Network,
migrator: types.Operation, # pylint: disable=unused-argument
zeus: types.TestContainerDocker,
gateway: types.TestContainerDocker,
sqlstore: types.TestContainerSQL,
clickhouse: types.TestContainerClickhouse,
request: pytest.FixtureRequest,
pytestconfig: pytest.Config,
) -> types.SigNoz:
"""
Package-scoped SigNoz instance with the skip_resource_fingerprint
optimization enabled and a low threshold so both the materialized and
fallback resolver paths are exercised by sibling tests.
"""
return create_signoz(
network=network,
zeus=zeus,
gateway=gateway,
sqlstore=sqlstore,
clickhouse=clickhouse,
request=request,
pytestconfig=pytestconfig,
cache_key="signoz-skip-resource-fingerprint",
env_overrides={
"SIGNOZ_QUERIER_SKIP__RESOURCE__FINGERPRINT_ENABLED": True,
"SIGNOZ_QUERIER_SKIP__RESOURCE__FINGERPRINT_THRESHOLD": 2,
},
)