Skip to content

homedepot/artifactory_exporter

 
 

Repository files navigation

JFrog Artifactory Exporter

Go Build Publish Image Go Report Card

A Prometheus exporter for JFrog Artifactory stats.

Note

This exporter is under development and more metrics will be added later. Tested on Artifactory Commercial, Enterprise and OSS version 6.x and 7.x.

Authentication

The Artifactory exporter requires admin user and it supports multiple means of authentication. The following methods are supported:

  • Basic Auth
  • Bearer Token

Basic Auth

Basic auth may be used by setting ARTI_USERNAME and ARTI_PASSWORD environment variables.

Bearer Token

Artifactory access tokens may be used via the Authorization header by setting ARTI_ACCESS_TOKEN environment variable.

Usage

Binary

Download the binary for your operation system from release page and run it:

$ ./artifactory_exporter <flags>

Docker

Set the credentials in env_file_name and you can deploy this exporter using the peimanja/artifactory_exporter Docker image:

$ docker run --env-file=env_file_name -p 9531:9531 peimanja/artifactory_exporter:latest <flags>

Docker Compose

Running the exporter against an Artifactory instance with millions of artifacts will cause performance issues in case Prometheus will scrape too often.

To avoid such situations you can run nginx in front of the exporter and use it as a cache. The Artifactory responses will be cached in nginx and kept valid for PROXY_CACHE_VALID seconds. After that time any new request from Prometheus will re-request metrics from Artifactory and store again in the nginx cache.

Set the credentials in an environment file as described in the Docker section and store the file as .env next to docker-compose.yml and run the following command:

docker-compose up -d

Install with Helm

Helm must be installed to use the charts. Please refer to Helm's documentation to get started.

Source code for the exporter helm chart can be found here peimanja/helm-charts

Once Helm is set up properly, add the repo as follows:

Prerequisites

  • Kubernetes 1.8+ with Beta APIs enabled

Add Repo

helm repo add peimanja https://peimanja.github.io/helm-charts
helm repo update

See helm repo for command documentation.

Configuration

See Customizing the Chart Before Installing. To see all configurable options with detailed comments, visit the chart's values.yaml, or run these configuration commands:

# Helm 3
helm show values peimanja/prometheus-artifactory-exporter

Set your values in myvals.yaml:

artifactory:
  url: http://artifactory:8081/artifactory
  accessToken: "xxxxxxxxxxxxxxxxxxxx"
  existingSecret: false

options:
  logLevel: info
  logFormat: logfmt
  telemetryPath: /metrics
  verifySSL: false
  timeout: 5s
  optionalMetrics:
    - replication_status
    - federation_status

Install Chart

# Helm 3
helm install -f myvals.yaml [RELEASE_NAME] peimanja/prometheus-artifactory-exporter

See configuration below.

See helm install for command documentation.

Flags

$  docker run peimanja/artifactory_exporter:latest -h
usage: artifactory_exporter [<flags>]

Flags:
  -h, --help                    Show context-sensitive help (also try --help-long and --help-man).
      --web.listen-address=":9531"
                                Address to listen on for web interface and telemetry.
      --web.telemetry-path="/metrics"
                                Path under which to expose metrics.
      --artifactory.scrape-uri="http://localhost:8081/artifactory"
                                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.
Flag / Environment Variable Required Default Description
web.listen-address
WEB_LISTEN_ADDR
No :9531 Address to listen on for web interface and telemetry.
web.telemetry-path
WEB_TELEMETRY_PATH
No /metrics Path under which to expose metrics.
artifactory.scrape-uri
ARTI_SCRAPE_URI
No http://localhost:8081/artifactory URI on which to scrape JFrog Artifactory.
artifactory.ssl-verify
ARTI_SSL_VERIFY
No true Flag that enables SSL certificate verification for the scrape URI.
artifactory.timeout
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
ARTI_PASSWORD *No Password of the user accessing the Artifactory
ARTI_ACCESS_TOKEN *No Access token for accessing the Artifactory
  • Either ARTI_USERNAME and ARTI_PASSWORD or ARTI_ACCESS_TOKEN environment variables has to be set.

Metrics

Some metrics are not available with Artifactory OSS license. The exporter returns the following metrics:

Metric Description Labels OSS support
artifactory_up Was the last scrape of Artifactory successful.
artifactory_exporter_build_info Exporter build information. version, revision, branch, goversion
artifactory_exporter_total_scrapes Current total artifactory scrapes.
artifactory_exporter_total_api_errors Current total Artifactory API errors when scraping for stats.
artifactory_exporter_json_parse_failures Number of errors while parsing Json.
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.
artifactory_storage_artifacts_size_bytes Total artifacts Size stored in Artifactory in bytes.
artifactory_storage_binaries Total binaries count stored in Artifactory.
artifactory_storage_binaries_size_bytes Total binaries Size stored in Artifactory in bytes.
artifactory_storage_filestore_bytes Total space in the file store in bytes. storage_dir, storage_type
artifactory_storage_filestore_used_bytes Space used in the file store in bytes. storage_dir, storage_type
artifactory_storage_filestore_free_bytes Space free in the file store in bytes. storage_dir, storage_type
artifactory_storage_repo_used_bytes Space used by an Artifactory repository in bytes. name, package_type, type
artifactory_storage_repo_folders Number of folders in an Artifactory repository. name, package_type, type
artifactory_storage_repo_files Number files in an Artifactory repository. name, package_type, type
artifactory_storage_repo_items Number Items in an Artifactory repository. name, package_type, type
artifactory_artifacts_created_1m Number of artifacts created in the repo (last 1 minute). name, package_type, type
artifactory_artifacts_created_5m Number of artifacts created in the repo (last 5 minutes). name, package_type, type
artifactory_artifacts_created_15m Number of artifacts created in the repo (last 15 minutes). name, package_type, type
artifactory_artifacts_downloaded_1m Number of artifacts downloaded from the repository (last 1 minute). name, package_type, type
artifactory_artifacts_downloaded_5m Number of artifacts downloaded from the repository (last 5 minutes). name, package_type, type
artifactory_artifacts_downloaded_15m Number of artifacts downloaded from the repository (last 15 minute). name, package_type, type
artifactory_system_healthy Is Artifactory working properly (1 = healthy).
artifactory_system_license License type and expiry as labels, seconds to expiration as value type, licensed_to, expires
artifactory_system_version Version and revision of Artifactory as labels. version, revision
artifactory_federation_mirror_lag Federation mirror lag in milliseconds. name, remote_url, remote_name
artifactory_federation_unavailable_mirror Unsynchronized federated mirror status. status, name, remote_url, remote_name
  • Common labels:
    • node_id: Artifactory node ID that the metric is scraped from.

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:

  • artifacts - Extracts number of artifacts created/downloaded for each repository. Enabling this will add artifactory_artifacts_* metrics. Please note that on large Artifactory instances, this may impact the performance.
  • 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.
  • federation_status - Extracts federation metrics. Enabling this will add two new metrics: artifactory_federation_mirror_lag, and artifactory_federation_unavailable_mirror. Please note that these metrics are only available in Artifactory Enterprise Plus and version 7.18.3 and above.
  • open_metrics - Exposes Open Metrics from the JFrog Platform. For more information about Open Metrics, please refer to JFrog Platform Open Metrics.

Grafana Dashboard

Dashboard can be found here.

Grafana dDashboard Grafana dDashboard

Common Issues

In most cases enabling debug logs will help to identify the issue. To enable debug logs, use --log.level=debug flag.

No metrics are being scraped

  • Check if the exporter is running and listening on the port specified by --web.listen-address flag.
  • Check if artifactory_up metric is 1 or 0. If it is 0, check the logs for the error message.
  • Check if artifactory_exporter_total_api_errors metric is 0. If it is not 0 and it is increasing, check the logs for the error message.

Some metrics or labels are missing

  • Check the logs to see if there are any timeouts or errors while scraping for metrics. In a large Artifactory instance, it may take a long time to scrape for all metrics especially artifactory_artifacts_* metrics. If there are any errors, try increasing the default timeout(5s) using --artifactory.timeout flag.
  • Some metrics are not available based on your version or license type. Check the metrics section to see if the metric is available for your license type.
  • Some metrics are optional and are disabled by default. Check the optional metrics section to see available optional metrics. You can enable them using --optional-metric=metric_name flag. You can pass this flag multiple times to enable multiple optional metrics.

There was an error when trying to unmarshal the API Error

This error is usually means that the exporter is not able to properly reach the Artifactory endpoint. One of the common reasons for this is that the Artifactory URL is not set properly. Make sure your are not missing /artifactory at the end of the URL which is how most implementations of Artifactory are configured. (e.g. http://artifactory.yourdomain.com/artifactory)

About

JFrog Artifactory Prometheus Exporter written in Go

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Go 98.7%
  • Dockerfile 1.3%