Katkat
Back

Debug Kubernetes crashes and review Python ingestion script

IXL LearningSoftware Engineer · Technical Screen
EasySoftware EngineerTechnical ScreenSoftware Engineering Fundamentals

Last updated

This question evaluates operational debugging and system reliability skills — specifically Kubernetes pod/container troubleshooting, health-probe configuration, and practical Python ETL code review, focusing on observability, dependency management, and production-readiness.

Part A — Kubernetes: Pod/Container Crash Debugging

You are on call and a service deployed to Kubernetes starts failing. The Pod shows CrashLoopBackOff (or the container exits repeatedly).

  1. Walk through a systematic, step-by-step process to debug the issue.
  2. What are the most common root causes you would check for first?
  3. What Kubernetes commands and signals (events, logs, status fields) would you use?

Part B — Kubernetes Probes

Explain the following probe types and when to use each:

  • Liveness probe
  • Readiness probe
  • Startup probe

Include:

  • What problem each probe solves
  • Typical failure modes and pitfalls (e.g., causing restart loops)
  • Example probe choices for (i) a fast-starting stateless API, (ii) a slow-starting JVM app, (iii) an app that depends on a downstream database

Part C — Python ETL/Script Code Review

You are given a small Python script that:

  1. reads a local file, 2) performs simple processing, and 3) writes rows into a SQL database.

A simplified version looks like:

  • Open file and read entire content into memory
  • For each line: parse/split fields and do minimal transformation
  • For each record: build an SQL string and execute an INSERT
  • Commit at the end

List the major issues you would call out in a code review (correctness, reliability, security, performance, operations). Propose concrete improvements suitable for production.