feat: Optionally extract replication status for artifactory_replicati…

…on_enabled metric (#88)

Co-authored-by: Peiman Jafari <[email protected]>

README.md

@@ -126,6 +126,8 @@ Flags:
                                 URI on which to scrape JFrog Artifactory.
       --artifactory.ssl-verify  Flag that enables SSL certificate verification for the scrape URI
       --artifactory.timeout=5s  Timeout for trying to get stats from JFrog Artifactory.
+      --optional-metric=metric-name ...
+                                optional metric to be enabled. Pass multiple times to enable multiple optional metrics.
       --log.level=info          Only log messages with the given severity or above. One of: [debug, info, warn, error]
       --log.format=logfmt       Output format of log messages. One of: [logfmt, json]
       --version                 Show application version.
@@ -138,6 +140,7 @@ Flags:
 | `artifactory.scrape-uri`<br/>`ARTI_SCRAPE_URI` | No | `http://localhost:8081/artifactory` | URI on which to scrape JFrog Artifactory. |
 | `artifactory.ssl-verify`<br/>`ARTI_SSL_VERIFY` | No | `true` | Flag that enables SSL certificate verification for the scrape URI. |
 | `artifactory.timeout`<br/>`ARTI_TIMEOUT` | No | `5s` | Timeout for trying to get stats from JFrog Artifactory. |
+| `optional-metric`| No | | optional metric to be enabled. Pass multiple times to enable multiple optional metrics. |
 | `log.level` | No | `info` | Only log messages with the given severity or above. One of: [debug, info, warn, error]. |
 | `log.format` | No | `logfmt` | Output format of log messages. One of: [logfmt, json]. |
 | `ARTI_USERNAME` | *No | | User to access Artifactory |
@@ -157,7 +160,7 @@ Some metrics are not available with Artifactory OSS license. The exporter return
 | artifactory_exporter_total_scrapes | Current total artifactory scrapes. |  | &#9989; |
 | artifactory_exporter_total_api_errors | Current total Artifactory API errors when scraping for stats. |  | &#9989; |
 | artifactory_exporter_json_parse_failures |Number of errors while parsing Json. |  | &#9989; |
-| artifactory_replication_enabled | Replication status for an Artifactory repository (1 = enabled). | `name`, `type`, `cron_exp` | |
+| artifactory_replication_enabled | Replication status for an Artifactory repository (1 = enabled). | `name`, `type`, `cron_exp`, `status` | |
 | artifactory_security_groups | Number of Artifactory groups. | | |
 | artifactory_security_users | Number of Artifactory users for each realm. | `realm` | |
 | artifactory_storage_artifacts | Total artifacts count stored in Artifactory. |  | &#9989; |
@@ -181,6 +184,13 @@ Some metrics are not available with Artifactory OSS license. The exporter return
 | artifactory_system_license | License type and expiry as labels, seconds to expiration as value | `type`, `licensed_to`, `expires` | &#9989; |
 | artifactory_system_version | Version and revision of Artifactory as labels. | `version`, `revision` | &#9989; |
 
+#### Optional metrics
+
+Some metrics are expensive to compute and are disabled by default. To enable them, use `--optional-metric=metric_name` flag. Use this with caution as it may impact the performance in Artifactory instances with many repositories.
+
+Supported optional metrics:
+
+* `replication_status` - Extracts status of replication for each repository which has replication enabled. Enabling this will add the `status` label to `artifactory_replication_enabled` metric.
 
 ### Grafana Dashboard
 

artifactory/client.go

@@ -10,11 +10,12 @@ import (
 
 // Client represents Artifactory HTTP Client
 type Client struct {
-	URI        string
-	authMethod string
-	cred       config.Credentials
-	client     *http.Client
-	logger     log.Logger
+	URI             string
+	authMethod      string
+	cred            config.Credentials
+	optionalMetrics config.OptionalMetrics
+	client          *http.Client
+	logger          log.Logger
 }
 
 // NewClient returns an initialized Artifactory HTTP Client.
@@ -25,10 +26,11 @@ func NewClient(conf *config.Config) *Client {
 		Transport: tr,
 	}
 	return &Client{
-		URI:        conf.ArtiScrapeURI,
-		authMethod: conf.Credentials.AuthMethod,
-		cred:       *conf.Credentials,
-		client:     client,
-		logger:     conf.Logger,
+		URI:             conf.ArtiScrapeURI,
+		authMethod:      conf.Credentials.AuthMethod,
+		cred:            *conf.Credentials,
+		optionalMetrics: conf.OptionalMetrics,
+		client:          client,
+		logger:          conf.Logger,
 	}
 }

artifactory/replication.go

@@ -2,11 +2,13 @@ package artifactory
 
 import (
 	"encoding/json"
+	"fmt"
 
 	"github.com/go-kit/kit/log/level"
 )
 
 const replicationEndpoint = "replications"
+const replicationStatusEndpoint = "replication"
 
 // Replication represents single element of API respond from replication endpoint
 type Replication struct {
@@ -21,12 +23,17 @@ type Replication struct {
 	EnableEventReplication          bool   `json:"enableEventReplication"`
 	CheckBinaryExistenceInFilestore bool   `json:"checkBinaryExistenceInFilestore"`
 	SyncStatistics                  bool   `json:"syncStatistics"`
+	Status                          string `json:"status"`
 }
 type Replications struct {
 	Replications []Replication
 	NodeId       string
 }
 
+type ReplicationStatus struct {
+	Status string `json:"status"`
+}
+
 // FetchReplications makes the API call to replication endpoint and returns []Replication
 func (c *Client) FetchReplications() (Replications, error) {
 	var replications Replications
@@ -47,5 +54,27 @@ func (c *Client) FetchReplications() (Replications, error) {
 			endpoint: replicationEndpoint,
 		}
 	}
+
+	if c.optionalMetrics.ReplicationStatus {
+		level.Debug(c.logger).Log("msg", "Fetching replications status")
+		for i, replication := range replications.Replications {
+			var status ReplicationStatus
+			if replication.Enabled {
+				statusResp, err := c.FetchHTTP(fmt.Sprintf("%s/%s", replicationStatusEndpoint, replication.RepoKey))
+				if err != nil {
+					return replications, err
+				}
+				if err := json.Unmarshal(statusResp.Body, &status); err != nil {
+					level.Error(c.logger).Log("msg", "There was an issue when try to unmarshal replication status respond")
+					return replications, &UnmarshalError{
+						message:  err.Error(),
+						endpoint: fmt.Sprintf("%s/%s", replicationStatusEndpoint, replication.RepoKey),
+					}
+				}
+				replications.Replications[i].Status = status.Status
+			}
+		}
+	}
+
 	return replications, nil
 }

collector/collector.go

@@ -15,7 +15,7 @@ var (
 	defaultLabelNames     = []string{"node_id"}
 	filestoreLabelNames   = append([]string{"storage_type", "storage_dir"}, defaultLabelNames...)
 	repoLabelNames        = append([]string{"name", "type", "package_type"}, defaultLabelNames...)
-	replicationLabelNames = append([]string{"name", "type", "url", "cron_exp"}, defaultLabelNames...)
+	replicationLabelNames = append([]string{"name", "type", "url", "cron_exp", "status"}, defaultLabelNames...)
 )
 
 func newMetric(metricName string, subsystem string, docString string, labelNames []string) *prometheus.Desc {

collector/replication.go

@@ -28,8 +28,9 @@ func (e *Exporter) exportReplications(ch chan<- prometheus.Metric) error {
 				rType := strings.ToLower(replication.ReplicationType)
 				rURL := strings.ToLower(replication.URL)
 				cronExp := replication.CronExp
-				level.Debug(e.logger).Log("msg", "Registering metric", "metric", metricName, "repo", replication.RepoKey, "type", rType, "url", rURL, "cron", cronExp, "value", enabled)
-				ch <- prometheus.MustNewConstMetric(metric, prometheus.GaugeValue, enabled, repo, rType, rURL, cronExp, replications.NodeId)
+				status := replication.Status
+				level.Debug(e.logger).Log("msg", "Registering metric", "metric", metricName, "repo", replication.RepoKey, "type", rType, "url", rURL, "cron", cronExp, "status", status, "value", enabled)
+				ch <- prometheus.MustNewConstMetric(metric, prometheus.GaugeValue, enabled, repo, rType, rURL, cronExp, status, replications.NodeId)
 			}
 		}
 	}

config/config.go

@@ -14,13 +14,16 @@ import (
 )
 
 var (
-	listenAddress = kingpin.Flag("web.listen-address", "Address to listen on for web interface and telemetry.").Envar("WEB_LISTEN_ADDR").Default(":9531").String()
-	metricsPath   = kingpin.Flag("web.telemetry-path", "Path under which to expose metrics.").Envar("WEB_TELEMETRY_PATH").Default("/metrics").String()
-	artiScrapeURI = kingpin.Flag("artifactory.scrape-uri", "URI on which to scrape JFrog Artifactory.").Envar("ARTI_SCRAPE_URI").Default("http://localhost:8081/artifactory").String()
-	artiSSLVerify = kingpin.Flag("artifactory.ssl-verify", "Flag that enables SSL certificate verification for the scrape URI").Envar("ARTI_SSL_VERIFY").Default("false").Bool()
-	artiTimeout   = kingpin.Flag("artifactory.timeout", "Timeout for trying to get stats from JFrog Artifactory.").Envar("ARTI_TIMEOUT").Default("5s").Duration()
+	listenAddress   = kingpin.Flag("web.listen-address", "Address to listen on for web interface and telemetry.").Envar("WEB_LISTEN_ADDR").Default(":9531").String()
+	metricsPath     = kingpin.Flag("web.telemetry-path", "Path under which to expose metrics.").Envar("WEB_TELEMETRY_PATH").Default("/metrics").String()
+	artiScrapeURI   = kingpin.Flag("artifactory.scrape-uri", "URI on which to scrape JFrog Artifactory.").Envar("ARTI_SCRAPE_URI").Default("http://localhost:8081/artifactory").String()
+	artiSSLVerify   = kingpin.Flag("artifactory.ssl-verify", "Flag that enables SSL certificate verification for the scrape URI").Envar("ARTI_SSL_VERIFY").Default("false").Bool()
+	artiTimeout     = kingpin.Flag("artifactory.timeout", "Timeout for trying to get stats from JFrog Artifactory.").Envar("ARTI_TIMEOUT").Default("5s").Duration()
+	optionalMetrics = kingpin.Flag("optional-metric", "optional metric to be enabled. Pass multiple times to enable multiple optional metrics.").PlaceHolder("metric-name").Strings()
 )
 
+var optionalMetricsList = []string{"replication_status"}
+
 // Credentials represents Username and Password or API Key for
 // Artifactory Authentication
 type Credentials struct {
@@ -30,15 +33,20 @@ type Credentials struct {
 	AccessToken string `required:"false" envconfig:"ARTI_ACCESS_TOKEN"`
 }
 
+type OptionalMetrics struct {
+	ReplicationStatus bool
+}
+
 // Config represents all configuration options for running the Exporter.
 type Config struct {
-	ListenAddress string
-	MetricsPath   string
-	ArtiScrapeURI string
-	Credentials   *Credentials
-	ArtiSSLVerify bool
-	ArtiTimeout   time.Duration
-	Logger        log.Logger
+	ListenAddress   string
+	MetricsPath     string
+	ArtiScrapeURI   string
+	Credentials     *Credentials
+	ArtiSSLVerify   bool
+	ArtiTimeout     time.Duration
+	OptionalMetrics OptionalMetrics
+	Logger          log.Logger
 }
 
 // NewConfig Creates new Artifactory exporter Config
@@ -69,14 +77,25 @@ func NewConfig() (*Config, error) {
 		return nil, err
 	}
 
+	optMetrics := OptionalMetrics{}
+	for _, metric := range *optionalMetrics {
+		switch metric {
+		case "replication_status":
+			optMetrics.ReplicationStatus = true
+		default:
+			return nil, fmt.Errorf("unknown optional metric: %s. Valid optional metrics are: %s", metric, optionalMetricsList)
+		}
+	}
+
 	return &Config{
-		ListenAddress: *listenAddress,
-		MetricsPath:   *metricsPath,
-		ArtiScrapeURI: *artiScrapeURI,
-		Credentials:   &credentials,
-		ArtiSSLVerify: *artiSSLVerify,
-		ArtiTimeout:   *artiTimeout,
-		Logger:        logger,
+		ListenAddress:   *listenAddress,
+		MetricsPath:     *metricsPath,
+		ArtiScrapeURI:   *artiScrapeURI,
+		Credentials:     &credentials,
+		ArtiSSLVerify:   *artiSSLVerify,
+		ArtiTimeout:     *artiTimeout,
+		OptionalMetrics: optMetrics,
+		Logger:          logger,
 	}, nil
 
 }