Skip to main content

Application Deployments

Updated Mar 11, 2022 ·

Best Practices​

After selecting the best model and reviewing requirements for monoliths vs. microservices, the next step is implementation. Understanding and following best practices during the release and maintenance phases is essential for building resilient and highly available solutions. While strategies vary by organization, these guidelines help ensure stability.

Health Checks​

Health checks monitor the application’s status to confirm it can handle traffic.

  • Typically done via HTTP endpoints like /health or /status
  • Returns HTTP response codes to show if the app is functioning properly

Metrics​

Metrics measure application performance and usage.

  • Includes data such as logins, active users, requests, CPU usage, and memory
  • Often accessible through an HTTP endpoint like /metrics

Logs​

Aggregated logs provide insight into application operations over time.

  • Useful for debugging and troubleshooting

  • Collected through standard outputs (STDOUT, STDERR) and tools like Splunk

  • Often include timestamps for precise tracking of events

    Logging LevelsDescription
    DEBUGFine-grained application events
    INFOHigh-level operational events
    WARNPotential issues to monitor
    ERRORErrors that don’t stop the app
    FATALCritical errors that halt operation

Tracing​

Tracing maps the journey of a request through the application.

  • Records each function invoked, often using a tracing library
  • Consists of spans (individual operations) that together form a trace

Resource Consumption​

Resource consumption tracks how much CPU, memory, network bandwidth, and requests an application uses.

  • Essential for monitoring resource efficiency