You get paged: the site is down. You have no other information. What are your first five minutes?
What they are really testing: Whether you have a repeatable opening rather than a panic reflex. They want to see you establish scope and recency before touching anything, because those two facts eliminate most of the search space.
A real interview question
You get paged: the site is down. You have no other information. What are your first five minutes?
What most people say
drag me
“I would start looking through the application logs to find the error.”
Logs are enormous and unfiltered when you do not yet know the scope. Without knowing whether one endpoint or everything is failing, you are reading a haystack, and you have skipped the two questions that would have cut it by 90%.
The follow-ups they ask next
Scope is one endpoint failing for everyone, no recent deploy. Where next?
A dependency of that specific path: a downstream service, a cache, a database index, or a third party. Trace one failing request end to end rather than reading logs broadly.
Everything is down and the status page says the provider is healthy. What now?
Check the layers you own that are shared: DNS, certificates, the load balancer, and the network path. Total-everything failures usually sit at a shared edge component rather than in application code.
What the interviewer is listening for
- Establishes scope before touching anything
- Asks what changed recently
- Mitigates on a plausible cause rather than waiting for proof
What sinks the answer
- Opens logs immediately
- No communication
- Investigates to certainty while users are down
If you genuinely do not know
Say this instead of freezing. Reasoning out loud from what you do know beats silence every single time, and a good interviewer is listening for exactly that.
“First 2 questions: [what is the scope: one endpoint or all, one region or all] and [what changed in the last few hours: deploys, config, flags, certs]. Meanwhile [check dependencies and the provider status page] and [post in the channel]. Then [mitigate on a plausible cause rather than waiting for certainty].”
Keep going with troubleshooting
Junior
A developer says the build passes locally but fails in CI. How do you debug that?
Junior
A production server is failing writes. Disk is 100% full. Walk me through what you do.
Junior
At midnight everything started failing with TLS errors. What happened and what do you do?
Mid
A pod is in CrashLoopBackOff after a deploy. Walk me through your debugging, command by command.
Mid
The deploy reports success, pods are Running, but users get 502s. Where do you look?
Mid
A Kubernetes node goes NotReady and pods are stuck Terminating. What is happening and what do you do?
Knowing the answer is not the same as recalling it under pressure
Sign in to send the questions you fumble to spaced recall, so they come back right before you would forget them, and learn the concepts behind them with hands-on labs.
Start free