neuroforgede/promswarm

Modernized swarmprom

promswarm

Modernized version of swarmprom packaged as a nothelm chart.

How to use this repository

Requirements

• Traefik set up and running with an attachable network traefik-public (configurable name)
• A recent version of docker-stack-deploy https://github.com/neuroforgede/docker-stack-deploy for automatic secret rotation
• A recent version of nothelm.py https://github.com/neuroforgede/nothelm.py/ for the deployment templating

Install in your Swarm

Setup your workdirectory with the following commands (please verify all the variables in the yaml files this generates):

export BASE_DIR="."
export WORKDIR="$(pwd)/test"
export CHART_DIR="../chart"
export START_PWD=$(pwd)

mkdir -p $WORKDIR/secrets

echo "exec nothelm run deploy --project-dir $CHART_DIR -f values.yaml -f secrets/secret_values.yaml" > $WORKDIR/setup.sh

cat > $WORKDIR/secrets/secret_values.yaml << EOF
promswarm_prometheus_teams_webhook_url: "http://some.teams.webhook.url"
promswarm_alertmanager_admin_password: "s3cr3t"
promswarm_prometheus_admin_password: "moars3cr3t"
promswarm_grafana_admin_password: "holys3cr3tyay"
EOF

cat > $WORKDIR/values.yaml << EOF
stack_name: monitoring

promswarm_monitoring_base_domain: some.base.domain

promswarm_traefik_public_network: 'traefik-public'
promswarm_monitoring_net_network_mtu: '1350'
promswarm_external_monitors_network_mtu: '1350'

promswarm_prometheus_volume_config: {}
promswarm_grafana_volume_config: {}
promswarm_alertmanager_volume_config: {}
EOF

After you have verified everything looks good, you can deploy by running

cd $WORKDIR
bash setup.sh

Sample Screenshots

Grafana:

Prometheus:

Alertmanager: