Monitoring and Alerting with Prometheus

Explore Prometheus, an open-source monitoring solution for metrics and alerting based on time-series data and labels. Scrape HTTP endpoints to monitor cloud-native systems.

Install Grafana to enrich Prometheus dashboards, set up a Prometheus data source, and build dashboards with graphs, invite colleagues, and connect with LDAP or GitHub.

Explore how Prometheus stores data as time series identified by a metric name and labels, with samples carrying float64 values and millisecond timestamps, using notation like metric{label=value}.

Configure the Prometheus yaml file, reload with sighup, and start with --config.file. Explore default global settings, 15-second scrape and evaluation intervals, and targets such as localhost:9090/metrics.

Explore Prometheus configuration by examining the prometheus.yml file, scrape_configs, and targets, view metrics and dashboards at localhost:9090, and learn how to graph and alert from scraped data.

Install the node-exporter to monitor linux or unix nodes and expose cpu and memory metrics for Prometheus to pull, with Windows using a WMI exporter for similar metrics.

Install and configure Linux node_exporter for Prometheus, set up a systemd service and Prometheus job, then load /metrics and start graphing node metrics across hosts.

Demonstrates installing and running the WMI exporter on a Windows server, exposing metrics at localhost:9182, and configuring Prometheus to scrape and monitor Windows metrics such as disk free bytes.

Explore the Prometheus architecture, including the Prometheus server, service discovery for targets, PromQL queries, and dashboards in Grafana that visualize metrics and trigger alerts.

Explore how Prometheus client libraries in Python and Go push metrics, query with PromQL, and understand use cases of pushing metrics, service discovery, and exporters.

Instrument your code with Prometheus client libraries across languages, using counters, gauges, histograms, and summaries. Expose metrics in the chosen exposition format with labeled series to monitor latency and throughput.

Prometheus mainly uses pull-based collection, but pushes via the pushgateway for batch jobs when NAT or firewall blocks scraping, using grouping keys and push_to_gateway and delete_from_gateway.

Explore PromQL to query Prometheus metrics using instant and range vectors, scalars, and strings with arithmetic, rate over five minutes, and aggregation operators, including http_request_total and up.

Prometheus service discovery enables automatic detection of devices and services across AWS, Azure, Google, Kubernetes, DNS, Consul, Zookeeper, and file discovery, enabling dynamic scraping without manual configuration.

Learn how Prometheus exporters collect metrics from systems like Linux, MySQL, and HAProxy, and how to install and configure the node exporter to expose data in Prometheus format.

Discover how to set up Prometheus alerting with alerting rules and Alertmanager, route alerts to email and Slack, and manage deduplication, grouping, and silences. Configure alert rules and receivers.

Set up Prometheus alerts by installing Alertmanager, configuring mail and Slack integrations, and adding a cpu usage alert that fires on sustained load and notifies via email and Slack.

Prometheus storage uses local on-disk two-hour blocks with a write-ahead log and optional remote adapters for long-term backups, plus horizontal partitioning and an inverted index for efficient queries.

Explore Prometheus security by examining the absence of authentication and TLS on server components, and learn to secure Prometheus with a reverse proxy and TLS config for targets.

Demonstrates enabling tls and basic authentication for a Prometheus server via a reverse proxy with nginx, using self-signed or real certificates and htpasswd authentication.

Demonstrates configuring mutual TLS for Prometheus targets by encrypting with TLS, authenticating with client certificates, and using a reverse proxy (Nginx) to secure node exporter metrics.

Learn to monitor a Flask web app with Prometheus, calculate the Apdex score, and set alerts for performance targets like 95% of requests under 1 second.

Install and configure the Cloudwatch exporter for Prometheus, add aws elb metrics such as latency, healthy host count, and request count, and monitor with cost-aware query practices.

Provision Grafana with YAML and JSON files to manage data sources and dashboards as code, keeping state in a database and enabling version-controlled deployments.

Install Grafana to improve Prometheus visualizations and build dashboards. Log in to Grafana on port 3000, add a Prometheus data source at localhost:9090, and create graphs via queries.

Learn to deploy Prometheus on Kubernetes using Helm, expose metrics from the API server, nodes, and cadvisor, and configure relabeling and annotations to scrape service endpoints.

Learn how Prometheus integrates with Consul to scrape metrics, enable service discovery, and auto-register services as targets, with a focus on Prometheus integration rather than implementing Consul.

Demonstrates integrating consul with Prometheus using docker-based deployment, registering services, and exporting health metrics through a consul exporter, with automatic target discovery and relabeling.

Explore how Prometheus EC2 auto discovery automatically scrapes EC2 instances, with IAM roles, security groups, and node exporter, and visualize the results in Grafana.

Explore more courses on DevOps, Kubernetes, and Spinnaker after finishing this course, and learn how Prometheus enables monitoring and alerting across infrastructure.