freshservice-exporter

Prometheus exporter for Freshservice with a broad metric surface for Grafana Cloud — ITSM operations, SLA tracking, and DORA (MTTR, deployment frequency, change failure rate, lead time).

PyPI: pip install freshservice-exporter · Image: ghcr.io/polarpoint-io/freshservice-exporter:latest · Chart: ghcr.io/polarpoint-io/charts/freshservice-exporter

Repository layout

├── exporter.py              # Prometheus collector entrypoint
├── client.py                # Freshservice API v2 client
├── metrics.py / dora.py     # Aggregation logic
├── charts/                  # Helm chart for Kubernetes
├── dashboards/              # Grafana dashboard JSON
├── examples/                # Prometheus / Grafana Agent scrape configs
├── scripts/scrape_once.py   # One-shot local test scrape
└── tests/

Installation

From PyPI:

pip install freshservice-exporter

From GHCR:

docker run -e FRESHSERVICE_API_KEY=... -e FRESHSERVICE_DOMAIN=company.freshservice.com \
  -p 9192:9192 ghcr.io/polarpoint-io/freshservice-exporter:latest

Docker image tags

Quick start

pip install -e .

export FRESHSERVICE_API_KEY=your_api_key
export FRESHSERVICE_DOMAIN=your_company.freshservice.com
export FRESHSERVICE_INCLUDE_STATS=true
export FRESHSERVICE_DORA_LOOKBACK_DAYS=30

freshservice-exporter

Metrics: http://localhost:9192/metrics

Grafana query patterns

Use sum/count pairs for averages:

# Mean ticket resolution time (seconds)
sum(freshservice_ticket_resolution_duration_seconds_sum)
/
sum(freshservice_ticket_resolution_duration_seconds_count)

# Mean MTTR for incidents only (DORA)
freshservice_dora_mttr_seconds

# Deployment success rate
sum(freshservice_releases_completed_in_window_total)
/
(
  sum(freshservice_releases_completed_in_window_total)
  + sum(freshservice_releases_failed_in_window_total)
)

# Open ticket backlog by priority
sum by (priority) (freshservice_tickets_open_by_priority_total)

Metric catalog

Tickets

Changes

Releases (deployments)

Problems

DORA

Exporter diagnostics

Configuration

Development

make install
make test
make lint
make scrape    # requires .env with API credentials

Publish to GitHub (polarpoint-io)

gh auth login

gh repo create polarpoint-io/freshservice-exporter --private --source=. --remote=origin
git push -u origin main

Required GitHub secrets

Add these at GitHub repo → Settings → Secrets and variables → Actions:

Release process

Merges to main run semantic-release (ci.yml), which versions from conventional commits, publishes to PyPI, and creates a GitHub release. Publishing a release then triggers release.yml, which uploads to PyPI (idempotent, --skip-existing) and pushes the semver-tagged multi-arch image to GHCR built from the published package.

Helm

helm registry login ghcr.io --username <github-user> --password <github-pat>

helm install freshservice-exporter oci://ghcr.io/polarpoint-io/charts/freshservice-exporter \
  --namespace monitoring --create-namespace \
  --set freshservice.apiKey=<API_KEY> \
  --set freshservice.domain=company.freshservice.com

Or install from the local chart during development:

helm install freshservice-exporter ./charts/freshservice-exporter \
  --namespace monitoring --create-namespace \
  --set freshservice.apiKey=<API_KEY> \
  --set freshservice.domain=company.freshservice.com

License

MIT