Questions d'entretiens - Senior site reliability engineer


Questions d'entretien pour Senior Site Reliability Engineer partagées par les candidats

Principales questions d'entretien

Trier: Pertinence|Populaires|Date
On a demandé à Senior Infrastructure/Site Reliability Engineer...5 février 2018

What resources do linux cgroups manage?

2 réponses

resources. cpu, etc. thats how docker implements limits.

In these sorts of interviews you really need to drill down and understand what the interviewer is looking for. A good way to simulate a real interview experience is to do a mock with one of the Docker Senior Infrastructure/Site Reliability Engineer experts on Prepfully, rated super strongly on TrustPilot... Moins


How is managing infrastructure in public cloud, like AWS, different than managing physical infrastructure?

2 réponses

Managed host vs unmanaged hosts. i still wonder how you was able to get the job x'''DDD Moins

Besides not having actual hardware to contend with and monitor, things like availability, backups and restoration and LUN management are very different. Moins

Bloomberg L.P.

Which package contains the Hashmap class

2 réponses

I don't know, I haven't memorized Java package namspaces



Describe in detail what occurs after you enter a website in a browser?

2 réponses

I hope this detail is enough. 1. Get the IP of hostname - If its not in DNS cache, lookup local hosts, if not then do a DNS lookup on DNS server configured from you network setting. 2. Once you got the IP make TCP handshake to make TCP connection. (going through one or more Autonomous systems) 3. If it is http over SSL/TLS then make a SSL handshake using the TCP connection and then send encrypted http data to server. (So, far you may be making connection to reverse proxy like a loadbalancer or something). 4. The server will present HTTP response to your request as a html page (which may call other java scripts, images, css files and make other API calls if need, etc ) and the browser will present you the page as per the definition in html response. Moins

open socket to a server, process HTML code. the code may include JS that alter the HTML itself. src links are being downloaded and display as well as CSS being interpreted. Moins


If a system was having sporadic issues what would be a way to find out what was wrong, other than using monitoring tools.

2 réponses

I would talk to the customers using the system and see what issues they were noticing. Moins

There's a lot of elements to typically cover in these questions, clarifications, scoping, making sure you're answering the actual question the interviewer is looking for you to answer, etc. Could be worth doing a mock interview with one of the Prepfully DuckDuckGo Senior Site Reliability Engineer experts... they've worked in the role so they clearly know how to get through the interview. Moins

Moka (Indonesia)

What is your daily task in previous company

1 réponses

I manage team, deployment, maintenance, etc.

Credit Karma

Explained about the current project, Professional experience with K8, Cloud, Network TCP/IP

1 réponses

Went Okay, But didn't receive offer as expected.


There's a lot of culture and value questions, which I liked.

1 réponses

Just be yourself, answer sincerely. I commented that I like the diversity in the organization, and they replied that they need to do more. This impressed me. Moins


Já trabalhou com CI/CD? Descreva um fluxo que você acha necessário.

1 réponses

Já criei diversas pipelines de CI e CD, juntas ou separadas. Utilizei ferramentas como Gitlab, Drone CI, Harness (git ops), ArgoCD (git ops), Github actions e CircleCI. O fluxo depende muito do estilo da empresa, uma empresa híbrida por exemplo precisaria de etapas de aprovação de GMUD, isolamento entre CI e CD. Uma empresa ágil pode utilizar gitops com testes automatizados. Mas a idéia principal é pegar o código de uma versão aprovada, seja ela uma branch ou uma release no Github e criar a imagem (Docker) ou o artefato e rodar testes (unitários, coverage, vulnerabilidades) antes de salvá-lo em nuvem. Após essa etapa onde o produto é aprovado, deverá ser deployado automaticamente ou sob demanda para o ambiente de testes onde um time de QA irá validar a solução e no mínimo rodar um load test para manter a qualidade de performance. Após essa aprovação, a release está preparada para o ambiente produtivo e seu deploy deve ser coordenado de acordo com o risco da mudança, como uma migation que deve acontecer em momentos de pouco fluxo, um deploy gradual observando a performance da aplicação em ambiente produtivo, podendo possuir uma etapa de rollback automática caso a performance não seja boa. Moins

Arkose Labs

What's important to lead the team?

1 réponses

To be able to relate to the engineers at a level that makes feel safe and secure. Moins

1 - 10 sur 180 Questions d'entretien