TIG Stack + Jenkins

Monitor Jenkins CI metrics using the TIG pipeline:

Jenkins (CI server) → Telegraf (scrapes metrics) → InfluxDB 2.x (stores) → Grafana (visualizes)

Architecture

Service	Image	Port	Purpose
InfluxDB	influxdb:2	8086	Time-series storage (Flux)
Jenkins	jenkins/jenkins:lts	8080	CI server (metric source)
Telegraf	telegraf:latest	—	Scrapes Jenkins → writes InfluxDB
Grafana	grafana/grafana:latest	3000	Dashboards and visualization

Telegraf waits for both InfluxDB and Jenkins to pass health checks before starting. Grafana's InfluxDB datasource and a Jenkins Overview dashboard are auto-provisioned at startup.

Prerequisites

• Podman with podman compose (or Docker with docker compose)

Quick Start

1. Create the .env file

Copy the example below and adjust values as needed:

# InfluxDB
INFLUXDB_ORG=tig-org
INFLUXDB_BUCKET=jenkins
INFLUXDB_ADMIN_USER=admin
INFLUXDB_ADMIN_PASSWORD=admin12345
INFLUXDB_ADMIN_TOKEN=my-super-secret-token

# Grafana
GRAFANA_ADMIN_USER=admin
GRAFANA_ADMIN_PASSWORD=admin

2. Start the stack

podman compose up -d

First run pulls images and may take a few minutes. Telegraf will start automatically once InfluxDB and Jenkins are healthy.

3. Verify everything is running

podman compose ps

All 4 containers should show Up:

NAME       IMAGE                              STATUS
grafana    docker.io/grafana/grafana:latest   Up
influxdb   docker.io/library/influxdb:2       Up
jenkins    docker.io/jenkins/jenkins:lts      Up
telegraf   docker.io/library/telegraf:latest  Up

Service UIs

Service	URL	Credentials
Jenkins	http://localhost:8080	No auth (wizard disabled)
InfluxDB	http://localhost:8086	admin / admin12345
Grafana	http://localhost:3000	admin / admin

Running a Test Jenkins Job

Jenkins has the setup wizard disabled, so you can create jobs immediately via the REST API.

Fetch a CSRF crumb (required for all POST requests)

CRUMB=$(curl -sf -c /tmp/jenkins-cookies \
  'http://localhost:8080/crumbIssuer/api/json')
CRUMB_FIELD=$(echo "$CRUMB" | python3 -c "import sys,json; print(json.load(sys.stdin)['crumbRequestField'])")
CRUMB_VALUE=$(echo "$CRUMB" | python3 -c "import sys,json; print(json.load(sys.stdin)['crumb'])")

Create the job

curl -s -b /tmp/jenkins-cookies \
  -H "Content-Type: application/xml" \
  -H "$CRUMB_FIELD: $CRUMB_VALUE" \
  -X POST 'http://localhost:8080/createItem?name=test-job' \
  -d '<?xml version="1.0" encoding="UTF-8"?>
<project>
  <description>Test job for TIG stack verification</description>
  <builders>
    <hudson.tasks.Shell>
      <command>echo "Hello from Jenkins!" &amp;&amp; sleep 5 &amp;&amp; echo "Done."</command>
    </hudson.tasks.Shell>
  </builders>
</project>'

Trigger a build

Re-fetch the crumb (they are single-use), then trigger:

CRUMB=$(curl -sf -c /tmp/jenkins-cookies \
  'http://localhost:8080/crumbIssuer/api/json')
CRUMB_FIELD=$(echo "$CRUMB" | python3 -c "import sys,json; print(json.load(sys.stdin)['crumbRequestField'])")
CRUMB_VALUE=$(echo "$CRUMB" | python3 -c "import sys,json; print(json.load(sys.stdin)['crumb'])")

curl -s -b /tmp/jenkins-cookies \
  -H "$CRUMB_FIELD: $CRUMB_VALUE" \
  -X POST 'http://localhost:8080/job/test-job/build'

Check build result

Wait a few seconds for the build to finish, then:

curl -sf 'http://localhost:8080/job/test-job/lastBuild/api/json?pretty=true' \
  | python3 -c "
import sys, json
d = json.load(sys.stdin)
print(f\"Build #{d['number']}  Result: {d['result']}  Duration: {d['duration']}ms\")
"

View console output

curl -sf 'http://localhost:8080/job/test-job/lastBuild/consoleText'

Viewing Metrics in Grafana

1. Open http://localhost:3000 and log in (admin / admin)
2. Go to Dashboards → Jenkins Overview
3. Telegraf scrapes Jenkins every 30 seconds — wait up to a minute after a build for data to appear

The dashboard includes 6 panels:

• Executor Usage — busy vs total executors over time
• Job Build Duration — build time per job
• Job Build Result — last result color-coded (green=success, red=failure)
• Build Number Over Time — build numbers plotted chronologically
• Node Status — table of nodes with online/offline status
• Node Response Time — node response latency over time

Stopping the Stack

podman compose down

To also remove stored data (volumes):

podman compose down -v

File Structure

tig-stack/
├── docker-compose.yml
├── .env
├── .gitignore
├── telegraf/
│   └── telegraf.conf
└── grafana/
    └── provisioning/
        ├── datasources/
        │   └── influxdb.yml
        └── dashboards/
            ├── dashboard-provider.yml
            └── jenkins-overview.json