Question 1

Walk me through the most painful production incident you helped resolve.

Accepted Answer

Two minutes. Symptom, suspect list, what you tried, root cause, and the change you shipped after.

Question 2

What does it mean when df shows 0% free but du shows plenty of space?

Accepted Answer

Tests Linux internals — open file descriptors holding deleted inodes. Senior bar.

Question 3

Walk me through what happens when a Pod fails to start in Kubernetes.

Accepted Answer

Cover image pull, init containers, readiness vs liveness, ImagePullBackOff vs CrashLoopBackOff, kubectl describe.

Question 4

You see latency spike on a service. Walk me through the investigation.

Accepted Answer

Network → DB → app → upstream. Cover dashboards, percentile vs average, recent deploys, capacity, and rollback decision.

Question 5

Explain the difference between rolling updates, blue-green, and canary deploys.

Accepted Answer

Tradeoffs in cost, risk, rollback speed. Tie each to when you would actually use it.

Question 6

Write a shell pipeline that finds the top 10 IPs hitting an Nginx access log.

Accepted Answer

awk, sort -n, head. Tests basic Unix fluency.

Question 7

Write a Terraform module that provisions an S3 bucket with versioning, lifecycle, and a least-privilege IAM role.

Accepted Answer

Tests real IaC habits — variables, outputs, no hardcoded ARNs.

Question 8

Design the deployment pipeline for a service that ships 10x a day across staging and prod, with zero-downtime rollouts.

Accepted Answer

CI stages, gating tests, canary, rollback, observability hooks. Cover the bad-deploy detour.

Question 9

Tell me about a time you said no to a developer asking for prod access.

Accepted Answer

Tests judgment without sounding territorial. Frame as protecting the developer too.

Question 10

Tell me about a postmortem you wrote that changed how the team operates.

Accepted Answer

Concrete change with a date and a metric. Vague answers signal you have not led one.

DevOps / SRE Interview Prep — Linux, Kubernetes, CI/CD — for engineers in Tashkent and Central Asia

Core skills tested

Salary ranges in Tashkent (2026)

What you will actually be asked

Walk me through the most painful production incident you helped resolve.

What does it mean when df shows 0% free but du shows plenty of space?

Walk me through what happens when a Pod fails to start in Kubernetes.

You see latency spike on a service. Walk me through the investigation.

Explain the difference between rolling updates, blue-green, and canary deploys.

Write a shell pipeline that finds the top 10 IPs hitting an Nginx access log.

Write a Terraform module that provisions an S3 bucket with versioning, lifecycle, and a least-privilege IAM role.

Design the deployment pipeline for a service that ships 10x a day across staging and prod, with zero-downtime rollouts.

Tell me about a time you said no to a developer asking for prod access.

Tell me about a postmortem you wrote that changed how the team operates.

Practice these questions out loud — for free

Core skills tested

Salary ranges in Tashkent (2026)

What you will actually be asked

Walk me through the most painful production incident you helped resolve.

What does it mean when df shows 0% free but du shows plenty of space?

Walk me through what happens when a Pod fails to start in Kubernetes.

You see latency spike on a service. Walk me through the investigation.

Explain the difference between rolling updates, blue-green, and canary deploys.

Write a shell pipeline that finds the top 10 IPs hitting an Nginx access log.

Write a Terraform module that provisions an S3 bucket with versioning, lifecycle, and a least-privilege IAM role.

Design the deployment pipeline for a service that ships 10x a day across staging and prod, with zero-downtime rollouts.

Tell me about a time you said no to a developer asking for prod access.

Tell me about a postmortem you wrote that changed how the team operates.

Practice these questions out loud — for free

Related reading