Stage 12: observability & performance (OTel/OTLP, domain metrics, guest GC)
- pkg/telemetry: shared OTel provider bootstrap (none/stdout/otlp + W3C
propagators + Go runtime metrics); backend/internal/telemetry becomes a thin
facade keeping its gin middleware.
- Telemetry parity: gateway and the Telegram connector gain telemetry runtimes
and config (GATEWAY_/TELEGRAM_ SERVICE_NAME + OTEL_*); otelgrpc instruments the
backend push server, the gateway's backend+connector clients and the connector
server. Default exporter stays none (collector/dashboards are Stage 14).
- Operational metrics (variant attribute on game-scoped ones): game_replay_duration,
game_move_validate_duration, games_started_total, games_abandoned_total,
game_cache_active, chat_messages_total{kind}, gateway edge_request_duration.
Wired via the SetMetrics setter pattern (default no-op meter).
- TODO-3: account.GuestReaper deletes guests with no game seat past
BACKEND_GUEST_RETENTION (default 30d, swept every BACKEND_GUEST_REAP_INTERVAL).
- Tests: pkg/telemetry exporter selection; game/social/edge metric recording via
a manual reader; config (otlp accepted, guest knobs); inttest guest reaper.
- Docs: PLAN.md re-scopes Stage 12 and adds Stage 13 (alphabet-on-wire) + Stage 14
(CI/deploy) with the agreed dictionary-versioning resolution; ARCHITECTURE 11/13,
TESTING, the three READMEs and FUNCTIONAL(+ru) updated.
This commit is contained in:
@@ -0,0 +1,84 @@
|
||||
package telemetry
|
||||
|
||||
import (
|
||||
"context"
|
||||
"testing"
|
||||
"time"
|
||||
)
|
||||
|
||||
// TestConfigValidate covers the supported and rejected exporter selections.
|
||||
func TestConfigValidate(t *testing.T) {
|
||||
for _, exp := range []string{ExporterNone, ExporterStdout, ExporterOTLP} {
|
||||
c := DefaultConfig("svc")
|
||||
c.TracesExporter = exp
|
||||
c.MetricsExporter = exp
|
||||
if err := c.Validate(); err != nil {
|
||||
t.Errorf("exporter %q must be valid: %v", exp, err)
|
||||
}
|
||||
}
|
||||
bad := DefaultConfig("svc")
|
||||
bad.TracesExporter = "prometheus"
|
||||
if err := bad.Validate(); err == nil {
|
||||
t.Error("unsupported exporter must be rejected")
|
||||
}
|
||||
if err := DefaultConfig("").Validate(); err == nil {
|
||||
t.Error("empty service name must be rejected")
|
||||
}
|
||||
}
|
||||
|
||||
// TestNewNoneAndStdout builds the providers with the none and stdout exporters,
|
||||
// starts the runtime metrics and shuts them down.
|
||||
func TestNewNoneAndStdout(t *testing.T) {
|
||||
for _, exp := range []string{ExporterNone, ExporterStdout} {
|
||||
cfg := DefaultConfig("svc")
|
||||
cfg.TracesExporter = exp
|
||||
cfg.MetricsExporter = exp
|
||||
rt, err := New(context.Background(), cfg)
|
||||
if err != nil {
|
||||
t.Fatalf("New(%q): %v", exp, err)
|
||||
}
|
||||
if rt.TracerProvider() == nil || rt.MeterProvider() == nil {
|
||||
t.Errorf("New(%q): nil provider", exp)
|
||||
}
|
||||
if err := rt.StartRuntimeMetrics(); err != nil {
|
||||
t.Errorf("StartRuntimeMetrics(%q): %v", exp, err)
|
||||
}
|
||||
if err := rt.Shutdown(context.Background()); err != nil {
|
||||
t.Errorf("Shutdown(%q): %v", exp, err)
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
// TestNewOTLPConstructs verifies the otlp exporter selection builds providers
|
||||
// without a reachable collector (the gRPC exporters dial lazily). Shutdown is
|
||||
// best-effort: with no collector the final export may error, so it is bounded and
|
||||
// its result ignored.
|
||||
func TestNewOTLPConstructs(t *testing.T) {
|
||||
cfg := DefaultConfig("svc")
|
||||
cfg.TracesExporter = ExporterOTLP
|
||||
cfg.MetricsExporter = ExporterOTLP
|
||||
rt, err := New(context.Background(), cfg)
|
||||
if err != nil {
|
||||
t.Fatalf("New(otlp): %v", err)
|
||||
}
|
||||
if rt.TracerProvider() == nil || rt.MeterProvider() == nil {
|
||||
t.Error("New(otlp): nil provider")
|
||||
}
|
||||
ctx, cancel := context.WithTimeout(context.Background(), 2*time.Second)
|
||||
defer cancel()
|
||||
_ = rt.Shutdown(ctx)
|
||||
}
|
||||
|
||||
// TestNilRuntime checks the nil-receiver fallbacks used before initialisation.
|
||||
func TestNilRuntime(t *testing.T) {
|
||||
var rt *Runtime
|
||||
if rt.TracerProvider() == nil {
|
||||
t.Error("nil runtime must fall back to the global tracer provider")
|
||||
}
|
||||
if rt.MeterProvider() == nil {
|
||||
t.Error("nil runtime must fall back to the global meter provider")
|
||||
}
|
||||
if err := rt.Shutdown(context.Background()); err != nil {
|
||||
t.Errorf("nil runtime Shutdown: %v", err)
|
||||
}
|
||||
}
|
||||
Reference in New Issue
Block a user