Skip to content

Commit

Permalink
v1.35.0 fixups (#1110)
Browse files Browse the repository at this point in the history 
Fix a nil-pointer reference crash when clients create root spans without passing a context.

Add hotspots/endpoints to tracer startup message to help debugging.
  • Loading branch information
@felixge
felixge authored Dec 31, 2021
1 parent 44e0809 commit 84f1fa1
Show file tree
Hide file tree
Showing 4 changed files with 69 additions and 52 deletions.
100 changes: 52 additions & 48 deletions ddtrace/tracer/log.go
View file
Original file line number Diff line number Diff line change
Expand Up @@ -26,31 +26,33 @@ const (

// startupInfo contains various information about the status of the tracer on startup.
type startupInfo struct {
Date string `json:"date"` // ISO 8601 date and time of start
OSName string `json:"os_name"` // Windows, Darwin, Debian, etc.
OSVersion string `json:"os_version"` // Version of the OS
Version string `json:"version"` // Tracer version
Lang string `json:"lang"` // "Go"
LangVersion string `json:"lang_version"` // Go version, e.g. go1.13
Env string `json:"env"` // Tracer env
Service string `json:"service"` // Tracer Service
AgentURL string `json:"agent_url"` // The address of the agent
AgentError string `json:"agent_error"` // Any error that occurred trying to connect to agent
Debug bool `json:"debug"` // Whether debug mode is enabled
AnalyticsEnabled bool `json:"analytics_enabled"` // True if there is a global analytics rate set
SampleRate string `json:"sample_rate"` // The default sampling rate for the rules sampler
SamplingRules []SamplingRule `json:"sampling_rules"` // Rules used by the rules sampler
SamplingRulesError string `json:"sampling_rules_error"` // Any errors that occurred while parsing sampling rules
ServiceMappings map[string]string `json:"service_mappings"` // Service Mappings
Tags map[string]string `json:"tags"` // Global tags
RuntimeMetricsEnabled bool `json:"runtime_metrics_enabled"` // Whether or not runtime metrics are enabled
HealthMetricsEnabled bool `json:"health_metrics_enabled"` // Whether or not health metrics are enabled
ApplicationVersion string `json:"dd_version"` // Version of the user's application
Architecture string `json:"architecture"` // Architecture of host machine
GlobalService string `json:"global_service"` // Global service string. If not-nil should be same as Service. (#614)
LambdaMode string `json:"lambda_mode"` // Whether or not the client has enabled lambda mode
AppSec bool `json:"appsec"` // AppSec status: true when started, false otherwise.
AgentFeatures agentFeatures `json:"agent_features"` // Lists the capabilities of the agent.
Date string `json:"date"` // ISO 8601 date and time of start
OSName string `json:"os_name"` // Windows, Darwin, Debian, etc.
OSVersion string `json:"os_version"` // Version of the OS
Version string `json:"version"` // Tracer version
Lang string `json:"lang"` // "Go"
LangVersion string `json:"lang_version"` // Go version, e.g. go1.13
Env string `json:"env"` // Tracer env
Service string `json:"service"` // Tracer Service
AgentURL string `json:"agent_url"` // The address of the agent
AgentError string `json:"agent_error"` // Any error that occurred trying to connect to agent
Debug bool `json:"debug"` // Whether debug mode is enabled
AnalyticsEnabled bool `json:"analytics_enabled"` // True if there is a global analytics rate set
SampleRate string `json:"sample_rate"` // The default sampling rate for the rules sampler
SamplingRules []SamplingRule `json:"sampling_rules"` // Rules used by the rules sampler
SamplingRulesError string `json:"sampling_rules_error"` // Any errors that occurred while parsing sampling rules
ServiceMappings map[string]string `json:"service_mappings"` // Service Mappings
Tags map[string]string `json:"tags"` // Global tags
RuntimeMetricsEnabled bool `json:"runtime_metrics_enabled"` // Whether or not runtime metrics are enabled
HealthMetricsEnabled bool `json:"health_metrics_enabled"` // Whether or not health metrics are enabled
ProfilerCodeHotspotsEnabled bool `json:"profiler_code_hotspots_enabled"` // Whether or not profiler code hotspots are enabled
ProfilerEndpointsEnabled bool `json:"profiler_endpoints_enabled"` // Whether or not profiler endpoints are enabled
ApplicationVersion string `json:"dd_version"` // Version of the user's application
Architecture string `json:"architecture"` // Architecture of host machine
GlobalService string `json:"global_service"` // Global service string. If not-nil should be same as Service. (#614)
LambdaMode string `json:"lambda_mode"` // Whether or not the client has enabled lambda mode
AppSec bool `json:"appsec"` // AppSec status: true when started, false otherwise.
AgentFeatures agentFeatures `json:"agent_features"` // Lists the capabilities of the agent.
}

// checkEndpoint tries to connect to the URL specified by endpoint.
Expand Down Expand Up @@ -79,29 +81,31 @@ func logStartup(t *tracer) {
}

info := startupInfo{
Date: time.Now().Format(time.RFC3339),
OSName: osName(),
OSVersion: osVersion(),
Version: version.Tag,
Lang: "Go",
LangVersion: runtime.Version(),
Env: t.config.env,
Service: t.config.serviceName,
AgentURL: t.config.transport.endpoint(),
Debug: t.config.debug,
AnalyticsEnabled: !math.IsNaN(globalconfig.AnalyticsRate()),
SampleRate: fmt.Sprintf("%f", t.rulesSampling.globalRate),
SamplingRules: t.rulesSampling.rules,
ServiceMappings: t.config.serviceMappings,
Tags: tags,
RuntimeMetricsEnabled: t.config.runtimeMetrics,
HealthMetricsEnabled: t.config.runtimeMetrics,
ApplicationVersion: t.config.version,
Architecture: runtime.GOARCH,
GlobalService: globalconfig.ServiceName(),
LambdaMode: fmt.Sprintf("%t", t.config.logToStdout),
AgentFeatures: t.config.agent,
AppSec: appsec.Enabled(),
Date: time.Now().Format(time.RFC3339),
OSName: osName(),
OSVersion: osVersion(),
Version: version.Tag,
Lang: "Go",
LangVersion: runtime.Version(),
Env: t.config.env,
Service: t.config.serviceName,
AgentURL: t.config.transport.endpoint(),
Debug: t.config.debug,
AnalyticsEnabled: !math.IsNaN(globalconfig.AnalyticsRate()),
SampleRate: fmt.Sprintf("%f", t.rulesSampling.globalRate),
SamplingRules: t.rulesSampling.rules,
ServiceMappings: t.config.serviceMappings,
Tags: tags,
RuntimeMetricsEnabled: t.config.runtimeMetrics,
HealthMetricsEnabled: t.config.runtimeMetrics,
ApplicationVersion: t.config.version,
ProfilerCodeHotspotsEnabled: t.config.profilerHotspots,
ProfilerEndpointsEnabled: t.config.profilerEndpoints,
Architecture: runtime.GOARCH,
GlobalService: globalconfig.ServiceName(),
LambdaMode: fmt.Sprintf("%t", t.config.logToStdout),
AgentFeatures: t.config.agent,
AppSec: appsec.Enabled(),
}
if _, err := samplingRulesFromEnv(); err != nil {
info.SamplingRulesError = fmt.Sprintf("%s", err)
Expand Down
8 changes: 4 additions & 4 deletions ddtrace/tracer/log_test.go
View file
Original file line number Diff line number Diff line change
Expand Up @@ -26,7 +26,7 @@ func TestStartupLog(t *testing.T) {
logStartup(tracer)
lines := removeAppSec(tp.Lines())
assert.Len(lines, 2)
assert.Regexp(`Datadog Tracer v[0-9]+\.[0-9]+\.[0-9]+ INFO: DATADOG TRACER CONFIGURATION {"date":"[^"]*","os_name":"[^"]*","os_version":"[^"]*","version":"[^"]*","lang":"Go","lang_version":"[^"]*","env":"","service":"tracer\.test","agent_url":"http://localhost:9/v0.4/traces","agent_error":"Post .*","debug":false,"analytics_enabled":false,"sample_rate":"NaN","sampling_rules":null,"sampling_rules_error":"","service_mappings":null,"tags":{"runtime-id":"[^"]*"},"runtime_metrics_enabled":false,"health_metrics_enabled":false,"dd_version":"","architecture":"[^"]*","global_service":"","lambda_mode":"false","appsec":((true)|(false)),"agent_features":{"DropP0s":false,"Stats":false,"StatsdPort":0}}`, lines[1])
assert.Regexp(`Datadog Tracer v[0-9]+\.[0-9]+\.[0-9]+ INFO: DATADOG TRACER CONFIGURATION {"date":"[^"]*","os_name":"[^"]*","os_version":"[^"]*","version":"[^"]*","lang":"Go","lang_version":"[^"]*","env":"","service":"tracer\.test","agent_url":"http://localhost:9/v0.4/traces","agent_error":"Post .*","debug":false,"analytics_enabled":false,"sample_rate":"NaN","sampling_rules":null,"sampling_rules_error":"","service_mappings":null,"tags":{"runtime-id":"[^"]*"},"runtime_metrics_enabled":false,"health_metrics_enabled":false,"profiler_code_hotspots_enabled":false,"profiler_endpoints_enabled":false,"dd_version":"","architecture":"[^"]*","global_service":"","lambda_mode":"false","appsec":((true)|(false)),"agent_features":{"DropP0s":false,"Stats":false,"StatsdPort":0}}`, lines[1])
})

t.Run("configured", func(t *testing.T) {
Expand Down Expand Up @@ -55,7 +55,7 @@ func TestStartupLog(t *testing.T) {
tp.Reset()
logStartup(tracer)
assert.Len(tp.Lines(), 2)
assert.Regexp(`Datadog Tracer v[0-9]+\.[0-9]+\.[0-9]+ INFO: DATADOG TRACER CONFIGURATION {"date":"[^"]*","os_name":"[^"]*","os_version":"[^"]*","version":"[^"]*","lang":"Go","lang_version":"[^"]*","env":"configuredEnv","service":"configured.service","agent_url":"http://localhost:9/v0.4/traces","agent_error":"Post .*","debug":true,"analytics_enabled":true,"sample_rate":"0\.123000","sampling_rules":\[{"service":"mysql","name":"","sample_rate":0\.75}\],"sampling_rules_error":"","service_mappings":{"initial_service":"new_service"},"tags":{"runtime-id":"[^"]*","tag":"value","tag2":"NaN"},"runtime_metrics_enabled":true,"health_metrics_enabled":true,"dd_version":"2.3.4","architecture":"[^"]*","global_service":"configured.service","lambda_mode":"false","appsec":((true)|(false)),"agent_features":{"DropP0s":false,"Stats":false,"StatsdPort":0}}`, tp.Lines()[1])
assert.Regexp(`Datadog Tracer v[0-9]+\.[0-9]+\.[0-9]+ INFO: DATADOG TRACER CONFIGURATION {"date":"[^"]*","os_name":"[^"]*","os_version":"[^"]*","version":"[^"]*","lang":"Go","lang_version":"[^"]*","env":"configuredEnv","service":"configured.service","agent_url":"http://localhost:9/v0.4/traces","agent_error":"Post .*","debug":true,"analytics_enabled":true,"sample_rate":"0\.123000","sampling_rules":\[{"service":"mysql","name":"","sample_rate":0\.75}\],"sampling_rules_error":"","service_mappings":{"initial_service":"new_service"},"tags":{"runtime-id":"[^"]*","tag":"value","tag2":"NaN"},"runtime_metrics_enabled":true,"health_metrics_enabled":true,"profiler_code_hotspots_enabled":false,"profiler_endpoints_enabled":false,"dd_version":"2.3.4","architecture":"[^"]*","global_service":"configured.service","lambda_mode":"false","appsec":((true)|(false)),"agent_features":{"DropP0s":false,"Stats":false,"StatsdPort":0}}`, tp.Lines()[1])
})

t.Run("errors", func(t *testing.T) {
Expand All @@ -69,7 +69,7 @@ func TestStartupLog(t *testing.T) {
tp.Reset()
logStartup(tracer)
assert.Len(tp.Lines(), 2)
assert.Regexp(`Datadog Tracer v[0-9]+\.[0-9]+\.[0-9]+ INFO: DATADOG TRACER CONFIGURATION {"date":"[^"]*","os_name":"[^"]*","os_version":"[^"]*","version":"[^"]*","lang":"Go","lang_version":"[^"]*","env":"","service":"tracer\.test","agent_url":"http://localhost:9/v0.4/traces","agent_error":"Post .*","debug":false,"analytics_enabled":false,"sample_rate":"NaN","sampling_rules":\[{"service":"some.service","name":"","sample_rate":0\.234}\],"sampling_rules_error":"found errors:\\n\\tat index 1: rate not provided","service_mappings":null,"tags":{"runtime-id":"[^"]*"},"runtime_metrics_enabled":false,"health_metrics_enabled":false,"dd_version":"","architecture":"[^"]*","global_service":"","lambda_mode":"false","appsec":((true)|(false)),"agent_features":{"DropP0s":false,"Stats":false,"StatsdPort":0}}`, tp.Lines()[1])
assert.Regexp(`Datadog Tracer v[0-9]+\.[0-9]+\.[0-9]+ INFO: DATADOG TRACER CONFIGURATION {"date":"[^"]*","os_name":"[^"]*","os_version":"[^"]*","version":"[^"]*","lang":"Go","lang_version":"[^"]*","env":"","service":"tracer\.test","agent_url":"http://localhost:9/v0.4/traces","agent_error":"Post .*","debug":false,"analytics_enabled":false,"sample_rate":"NaN","sampling_rules":\[{"service":"some.service","name":"","sample_rate":0\.234}\],"sampling_rules_error":"found errors:\\n\\tat index 1: rate not provided","service_mappings":null,"tags":{"runtime-id":"[^"]*"},"runtime_metrics_enabled":false,"health_metrics_enabled":false,"profiler_code_hotspots_enabled":false,"profiler_endpoints_enabled":false,"dd_version":"","architecture":"[^"]*","global_service":"","lambda_mode":"false","appsec":((true)|(false)),"agent_features":{"DropP0s":false,"Stats":false,"StatsdPort":0}}`, tp.Lines()[1])
})

t.Run("lambda", func(t *testing.T) {
Expand All @@ -81,7 +81,7 @@ func TestStartupLog(t *testing.T) {
tp.Reset()
logStartup(tracer)
assert.Len(tp.Lines(), 1)
assert.Regexp(`Datadog Tracer v[0-9]+\.[0-9]+\.[0-9]+ INFO: DATADOG TRACER CONFIGURATION {"date":"[^"]*","os_name":"[^"]*","os_version":"[^"]*","version":"[^"]*","lang":"Go","lang_version":"[^"]*","env":"","service":"tracer\.test","agent_url":"http://localhost:9/v0.4/traces","agent_error":"","debug":false,"analytics_enabled":false,"sample_rate":"NaN","sampling_rules":null,"sampling_rules_error":"","service_mappings":null,"tags":{"runtime-id":"[^"]*"},"runtime_metrics_enabled":false,"health_metrics_enabled":false,"dd_version":"","architecture":"[^"]*","global_service":"","lambda_mode":"true","appsec":((true)|(false)),"agent_features":{"DropP0s":false,"Stats":false,"StatsdPort":0}}`, tp.Lines()[0])
assert.Regexp(`Datadog Tracer v[0-9]+\.[0-9]+\.[0-9]+ INFO: DATADOG TRACER CONFIGURATION {"date":"[^"]*","os_name":"[^"]*","os_version":"[^"]*","version":"[^"]*","lang":"Go","lang_version":"[^"]*","env":"","service":"tracer\.test","agent_url":"http://localhost:9/v0.4/traces","agent_error":"","debug":false,"analytics_enabled":false,"sample_rate":"NaN","sampling_rules":null,"sampling_rules_error":"","service_mappings":null,"tags":{"runtime-id":"[^"]*"},"runtime_metrics_enabled":false,"health_metrics_enabled":false,"profiler_code_hotspots_enabled":false,"profiler_endpoints_enabled":false,"dd_version":"","architecture":"[^"]*","global_service":"","lambda_mode":"true","appsec":((true)|(false)),"agent_features":{"DropP0s":false,"Stats":false,"StatsdPort":0}}`, tp.Lines()[0])
})
}

Expand Down
10 changes: 10 additions & 0 deletions ddtrace/tracer/tracer.go
View file
Original file line number Diff line number Diff line change
Expand Up @@ -341,6 +341,16 @@ func (t *tracer) StartSpan(operationName string, options ...ddtrace.StartSpanOpt
}
}
}
if pprofContext == nil {
// For root span's without context, there is no pprofContext, but we need
// one to avoid a panic() in pprof.WithLabels(). Using context.Background()
// is not ideal here, as it will cause us to remove all labels from the
// goroutine when the span finishes. However, the alternatives of not
// applying labels for such spans or to leave the endpoint/hotspot labels
// on the goroutine after it finishes are even less appealing. We'll have
// to properly document this for users.
pprofContext = gocontext.Background()
}
id := opts.SpanID
if id == 0 {
id = random.Uint64()
Expand Down
3 changes: 3 additions & 0 deletions internal/traceprof/traceproftest/app.go
View file
Original file line number Diff line number Diff line change
Expand Up @@ -251,6 +251,9 @@ func (a *App) Work(ctx context.Context, req *pb.WorkReq) (*pb.WorkRes, error) {
close(stop)
cpuSpan.Finish()

orphanSpan := tracer.StartSpan("orphan")
orphanSpan.Finish()

return &pb.WorkRes{
LocalRootSpanId: fmt.Sprintf("%d", localRootSpan.Context().SpanID()),
SpanId: fmt.Sprintf("%d", cpuSpan.Context().SpanID()),
Expand Down

0 comments on commit 84f1fa1

Please sign in to comment.