Vonage India Site Reliability Engineer Interview: Rounds, Questions and Preparation

What the job involves

Site Reliability Engineers keep large-scale production systems fast, available, and reliable. At global tech companies, they split time between writing code to automate operations, responding to incidents, and improving monitoring and capacity planning. Freshers usually start by owning small services, building dashboards, and joining on-call rotations with mentorship.

Role and technical knowledge

1. 1

   What does an SRE do differently from a traditional sysadmin or DevOps engineer?

   Explain that SREs apply software engineering to operations, use error budgets to balance reliability and feature velocity, and automate repetitive work instead of doing it manually.

2. 2

   Explain SLI, SLO, and error budget with a simple example for a customer-facing API.

   Define SLI as a measured metric like request latency or success rate, SLO as the target like 99.9 percent availability, and error budget as the allowed failure. Use a concrete API example.

3. 3

   How would you monitor a web service and decide what alerts to create?

   Cover the four golden signals: latency, traffic, errors, and saturation. Mention alerting on symptoms users feel, not every internal metric, and avoiding alert fatigue.

4. 4

   What happens during an incident, and what is a blameless postmortem?

   Describe detection, triage, mitigation, communication, and resolution. Explain that a blameless postmortem focuses on systemic causes and action items, not punishing individuals.

Core technical deep dive

1. 5

   A Linux server is running out of memory and the application is slow. How do you investigate?

   Walk through checking free, top, vmstat, and dmesg for OOM killer logs. Look at process memory, swap usage, and recent deployments. Suggest immediate mitigation and a long-term fix.

2. 6

   Explain how DNS resolution works and what you would check if a service is unreachable by hostname but works by IP.

   Describe recursive and authoritative lookups, caching, and TTL. Mention checking resolv.conf, nslookup or dig, and whether the DNS record or load balancer is misconfigured.

3. 7

   How do containers differ from virtual machines, and why does Kubernetes matter for reliability?

   Containers share the host kernel and are lighter; VMs include a full OS. Kubernetes provides self-healing, rolling updates, and scaling, which improve availability when configured well.

4. 8

   Write a small script or describe logic to check if a service is healthy and restart it if not.

   Show a simple loop with an HTTP health check, a threshold for failures, and a restart command. Mention logging, avoiding restart storms, and using systemd or Kubernetes probes instead of ad hoc scripts.

Practical scenarios

1. 9

   Users report that a checkout page is slow only in one region. How do you debug this?

   Check regional latency dashboards, CDN and DNS routing, database replication lag, and recent changes in that region. Compare with a healthy region and isolate network versus application causes.

2. 10

   A deployment caused a spike in 500 errors. What do you do in the first 10 minutes?

   Prioritize mitigation: roll back or feature-flag off, then communicate status. After stability, investigate logs and traces, and add a guardrail like canary deployment or automated rollback.

3. 11

   How would you reduce alert noise for a service that pages the on-call engineer every night?

   Group related alerts, raise thresholds, remove alerts without action, and add suppression during maintenance. Tie alerts to SLOs and review pages in postmortems to fix root causes.

4. 12

   You need to plan capacity for a service expecting 3x traffic during a sale. What do you check?

   Review current resource usage, bottlenecks like database connections, and autoscaling limits. Load test, pre-scale critical components, and prepare a rollback and communication plan.

Behavioural and HR

1. 13

   Tell me about a time you debugged a technical problem under time pressure.

   Use STAR format: situation, task, action, result. Highlight how you stayed calm, prioritized, and communicated, and what you learned to prevent recurrence.

2. 14

   On-call work can include nights and weekends. How do you feel about that?

   Show willingness and maturity: mention healthy habits, escalation paths, and the importance of reducing toil so on-call is sustainable. Avoid saying you are fine with no boundaries.

3. 15

   Describe a time you disagreed with a teammate or lead. How did you handle it?

   Focus on listening, data-driven discussion, and committing to the team decision even if you disagreed. Show respect and a bias for the best outcome over being right.

4. 16

   Why do you want to work as an SRE, and how do you keep learning?

   Connect personal interest in reliability, automation, and systems to the role. Mention concrete learning habits like labs, open-source, certifications, or personal projects.