Minor Improvements: Log Level and Metrics Documentation (#84)

* control log level via env var

Signed-off-by: Eric Pickard <piceri@github.com>

* rename deptracker_post_record_no_attestation to deptracker_post_record_unknown_artifact

Signed-off-by: Eric Pickard <piceri@github.com>

* add docs for deptracker_post_record_unknown_artifact_cache_hit

Signed-off-by: Eric Pickard <piceri@github.com>

* remove trailing space

Signed-off-by: Eric Pickard <piceri@github.com>

---------

Signed-off-by: Eric Pickard <piceri@github.com>

Author: piceri

README.md

@@ -65,6 +65,7 @@ Two modes of authentication are supported:
 
 | Variable               | Description                                | Default                                              |
 |------------------------|--------------------------------------------|------------------------------------------------------|
+| `LOG_LEVEL`            | log level: DEBUG, INFO, WARN, ERROR        | `INFO`                                               |
 | `ORG`                  | GitHub organization name                   | (required)                                           |
 | `BASE_URL`             | API base URL                               | `api.github.com`                                     |
 | `DN_TEMPLATE`          | Deployment name template                   | `{{namespace}}/{{deploymentName}}/{{containerName}}` |
@@ -192,9 +193,12 @@ The metrics exposed beyond the default Prometheus metrics are:
   record uploads.
 * `deptracker_post_record_rate_limited`: the number of post attempts
   that were rate limited.
-* `deptracker_post_record_no_attestation`: the number of attempts
+* `deptracker_post_record_unknown_artifact`: the number of attempts
   that resulted in no matching attestation for the container digest
-  (404 "no artifacts found" responses).
+  (404 "no artifacts found" responses) and an entry in the unknown
+  artifact cache.
+* `deptracker_post_record_unknown_artifact_cache_hit`: the number of
+  attempts to create new records prevented by the unknown artifact cache.
 * `deptracker_post_record_soft_fail`: the number of recoverable failed
   attempts to upload the deployment record.
 * `deptracker_post_record_hard_fail`: the number of failures to

cmd/deployment-tracker/main.go

@@ -10,6 +10,7 @@ import (
 	"net/http"
 	"os"
 	"os/signal"
+	"strings"
 	"syscall"
 	"time"
 
@@ -65,8 +66,13 @@ func main() {
 
 	// init logging
 	log.SetFlags(log.LstdFlags | log.Lshortfile | log.LUTC)
-	opts := slog.HandlerOptions{Level: slog.LevelInfo}
+	logLevelStr := getEnvOrDefault("LOG_LEVEL", "INFO")
+	level, msg := parseLogLevel(logLevelStr)
+	opts := slog.HandlerOptions{Level: level}
 	slog.SetDefault(slog.New(slog.NewJSONHandler(os.Stdout, &opts)))
+	if msg != "" {
+		slog.Warn(msg)
+	}
 
 	var ghAppPrivateKey []byte
 	if b64Key := os.Getenv("GH_APP_PRIVATE_KEY"); b64Key != "" {
@@ -220,3 +226,18 @@ func createK8sConfig(kubeconfig string) (*rest.Config, error) {
 	}
 	return clientcmd.BuildConfigFromFlags("", homeDir+"/.kube/config")
 }
+
+func parseLogLevel(logLevel string) (slog.Level, string) {
+	switch strings.ToUpper(logLevel) {
+	case "DEBUG":
+		return slog.LevelDebug, ""
+	case "INFO":
+		return slog.LevelInfo, ""
+	case "WARN":
+		return slog.LevelWarn, ""
+	case "ERROR":
+		return slog.LevelError, ""
+	default:
+		return slog.LevelInfo, fmt.Sprintf("%s is an unsupported log level (DEBUG, WARN, INFO, ERROR), using INFO...", logLevel)
+	}
+}

pkg/deploymentrecord/client.go

@@ -265,7 +265,7 @@ func (c *Client) PostOne(ctx context.Context, record *DeploymentRecord) error {
 		switch {
 		case resp.StatusCode == 404:
 			// No artifact found - do not retry
-			dtmetrics.PostDeploymentRecordNoAttestation.Inc()
+			dtmetrics.PostDeploymentRecordUnknownArtifact.Inc()
 			slog.Debug("no artifact attestation found, no record created",
 				"attempt", attempt,
 				"status_code", resp.StatusCode,

pkg/deploymentrecord/client_test.go

@@ -298,7 +298,7 @@ func testRecord() *DeploymentRecord {
 func allCounters() []prometheus.Counter {
 	return []prometheus.Counter{
 		dtmetrics.PostDeploymentRecordOk,
-		dtmetrics.PostDeploymentRecordNoAttestation,
+		dtmetrics.PostDeploymentRecordUnknownArtifact,
 		dtmetrics.PostDeploymentRecordRateLimited,
 		dtmetrics.PostDeploymentRecordSoftFail,
 		dtmetrics.PostDeploymentRecordHardFail,
@@ -308,19 +308,19 @@ func allCounters() []prometheus.Counter {
 
 func TestPostOne(t *testing.T) {
 	tests := []struct {
-		name              string
-		record            *DeploymentRecord
-		retries           int
-		handler           http.HandlerFunc
-		wantErr           bool
-		errType           any // expected error type for errors.As
-		errContain        string
-		wantOk            float64
-		wantNoAttestation float64
-		wantRateLimited   float64
-		wantSoftFail      float64
-		wantHardFail      float64
-		wantClientError   float64
+		name                string
+		record              *DeploymentRecord
+		retries             int
+		handler             http.HandlerFunc
+		wantErr             bool
+		errType             any // expected error type for errors.As
+		errContain          string
+		wantOk              float64
+		wantUnknownArtifact float64
+		wantRateLimited     float64
+		wantSoftFail        float64
+		wantHardFail        float64
+		wantClientError     float64
 	}{
 		{
 			name:   "success on 200",
@@ -354,10 +354,10 @@ func TestPostOne(t *testing.T) {
 				w.WriteHeader(http.StatusNotFound)
 				_, _ = w.Write([]byte(`{"message":"no artifacts found"}`))
 			},
-			wantErr:           true,
-			errType:           &NoArtifactError{},
-			errContain:        "sha256:abc123",
-			wantNoAttestation: 1,
+			wantErr:             true,
+			errType:             &NoArtifactError{},
+			errContain:          "sha256:abc123",
+			wantUnknownArtifact: 1,
 		},
 		{
 			name:   "400 returns ClientError",
@@ -555,15 +555,15 @@ func TestPostOne(t *testing.T) {
 			// Assert all metric deltas
 			wantDeltas := []float64{
 				tt.wantOk,
-				tt.wantNoAttestation,
+				tt.wantUnknownArtifact,
 				tt.wantRateLimited,
 				tt.wantSoftFail,
 				tt.wantHardFail,
 				tt.wantClientError,
 			}
 			names := []string{
 				"PostDeploymentRecordOk",
-				"PostDeploymentRecordNoAttestation",
+				"PostDeploymentRecordUnknownArtifact",
 				"PostDeploymentRecordRateLimited",
 				"PostDeploymentRecordSoftFail",
 				"PostDeploymentRecordHardFail",

pkg/dtmetrics/prom.go

@@ -50,9 +50,9 @@ var (
 	)
 
 	//nolint: revive
-	PostDeploymentRecordNoAttestation = promauto.NewCounter(
+	PostDeploymentRecordUnknownArtifact = promauto.NewCounter(
 		prometheus.CounterOpts{
-			Name: "deptracker_post_record_no_attestation",
+			Name: "deptracker_post_record_unknown_artifact",
 			Help: "The total number of post attempts that resulted in no matching attestation for the container digest (404 'no artifacts found' responses)",
 		},
 	)