Deployment Monitoring Fast

Rapidly configure real-time monitoring and alerting for application deployments.

When to Use This Command

Run this command when you need to:

• Set up health checks, metrics collection, and alerting for a freshly deployed service
• Create monitoring dashboards that track deployment success and application performance
• Configure automated rollback triggers based on error rate or latency thresholds

Consider alternatives when:

• You already have a mature observability stack and only need to add a single metric
• Your monitoring needs are limited to simple uptime pings without dashboards

Quick Start

Configuration

Copy
name: deployment-monitoring-fast
type: command
category: deployment

Example Invocation

Copy
claude command:run deployment-monitoring-fast --stack prometheus --app api-server

Example Output

Scanning deployment: api-server
Detected endpoints: /health, /metrics, /ready
Infrastructure: Kubernetes (namespace: production)

Configured monitoring:
  [+] Health check probe: /health (interval: 10s)
  [+] Prometheus scrape target: /metrics (port 9090)
  [+] Grafana dashboard: api-server-overview.json
  [+] Alert rules: 4 rules created
       - HighErrorRate (>5% 5xx in 5min)
       - HighLatency (p99 > 2s for 5min)
       - PodRestarts (>3 in 15min)
       - MemoryPressure (>85% for 10min)

Status: Monitoring active. Dashboard URL: http://grafana:3000/d/api-server

Core Concepts

Monitoring Stack Overview

Aspect	Details
Metrics Collection	Prometheus, Datadog, CloudWatch, or StatsD scraping
Visualization	Grafana dashboards with pre-built deployment panels
Alerting Channels	Slack, PagerDuty, email, or webhook integrations
Health Probes	HTTP, TCP, and gRPC liveness and readiness checks
SLO Tracking	Error budget burn rate and availability percentage

Monitoring Pipeline Workflow

  Application
      |
      v
  /metrics endpoint
      |
      v
  +------------------+
  | Prometheus       |---> Scrape & Store
  +------------------+
      |         |
      v         v
  +-------+ +--------+
  |Grafana| |Alertmgr|
  +-------+ +--------+
      |         |
      v         v
  Dashboard  Slack/PagerDuty

Configuration

Parameter	Type	Default	Description
stack	string	prometheus	Monitoring stack: prometheus, datadog, cloudwatch
app	string	required	Application or service name to monitor
namespace	string	default	Kubernetes namespace or environment identifier
alert_channel	string	slack	Where to send alerts: slack, pagerduty, email, webhook
scrape_interval	string	15s	How frequently to collect metrics from the application

Best Practices

1. Define SLOs Before Alerts - Establish service level objectives for latency and error rate first. Derive alert thresholds from SLO burn rates to avoid noisy, meaningless notifications.

2. Use Multi-Signal Detection - Combine error rate, latency, and saturation metrics in alert rules. A single metric can produce false positives, while correlated signals provide high confidence.

3. Separate Deployment Dashboards - Create a dedicated dashboard for deployment events overlaid with key metrics. This makes it immediately visible whether a new release caused a regression.

4. Set Meaningful Severity Levels - Reserve critical/page-worthy alerts for customer-facing outages. Use warning-level alerts for degradation that can wait until business hours.

5. Automate Runbook Links - Attach troubleshooting runbook URLs to every alert rule. When an alert fires at 3 AM, the on-call engineer needs actionable steps, not just a metric name.

Common Issues

1. Alert Fatigue From Noisy Rules - Overly sensitive thresholds fire constantly. Tune alert windows (use 5-minute averages instead of instant values) and add for-duration clauses to suppress transient spikes.

2. Metrics Endpoint Not Scraped - Prometheus cannot reach the /metrics path. Verify the service has the correct port annotation, network policy allows scraper traffic, and the endpoint returns 200.

3. Dashboard Shows No Data - The metric name in the query does not match what the application exposes. Use the Prometheus expression browser to verify available metric names before building panels.