Kubernetes Scenario Interview Questions | Kubernetes Interview Questions and Answers for Experienced
Вставка
- Опубліковано 30 чер 2024
- Kubernetes Scenario Interview Questions | Kubernetes Interview Questions and Answers for Experienced | 001
Hey folks,
Welcome back to another video in the series of DevOps Interviews. This is the 1st video in this series. This is a complete 34-minute-long technical video purely on #kubernetes #devops #devopsinterview etc. Pick a pen and paper, and note down the questions we have covered in this video.
LogicOps Lab - / @logicopslab
LogicOps Lab - Hindi - / @logicopslabhindi
GitHub - bit.ly/3lsUjdE
Twitter - / logicopslab
Instagram - bit.ly/3tJ9Jyf
Facebook - bit.ly/3ziMoVq
Like | Share | Subscribe | Follow
Music Credits:
Link - uppbeat.io/t/soundroll/fashio...
YT - • No Copyright Music Bac...
Name - Soundroll - Fashion Worship
kubernetes interview questions and answers for experienced,kubernetes interview questions,kubernetes interview questions and answers,kubernetes,devops kubernetes,devops kubernetes interview questions,kubernetes scenario based questions,kubernetes scenario interview questions,kubernetes troubleshooting scenarios,kubernetes scenario based interview questions,devops kubernetes project,kubernetes questions and answers,kubernetes questions,kubernetes interview
NOTICE/DISCLAIMER - We take permission/consent from the Candidate/Agency/Organisation before posting these interviews. Without consent/permission, we DO/WILL NOT publish any content in public.
The interviewer asks the following questions in the video:
1. [03:04] "Can you please give me a brief introduction about yourself?"
2. [04:11] "So, your total years of experience is four years, right?"
4. [05:18] "So, I'll start with a scenario-based question. Consider a company built on some kind of Monolithic architecture. How do you think the company should shift from monolithic to microservices and how can they implement Kubernetes in it?"
5. [07:24] "Can you explain the Kubernetes architecture?"
6. [09:27] "So, there is something known as Docker Swarm. Have you heard about it? Why should I choose Kubernetes, and why shouldn't I go with Docker Swarm?"
7. [11:33] "Based on the first question about monolithic and microservices architecture, how can your company or you solve the deployment problem on the devops side?"
8. [13:44] "What are all the services that you have worked on in Kubernetes, and can you explain a few of them?"
9. [15:21] "Consider your company; your manager wants to optimize the distribution of workloads. How can Kubernetes be helpful over here?"
10. [16:32] "You were talking about a load balancer. What do you understand by a load balancer in terms of Kubernetes only?"
11. [17:37] "Have you ever heard about an application known as Quick Ride? How do you think your organization or you can suggest that the company will deal with the servers and their installation using Kubernetes?"
12. [19:21] "What do you understand by a namespace in Kubernetes?"
13. [21:24] "What is the purpose of an operator in Kubernetes?"
1. [22:06] What is Elasticsearch used for, and how is it different from OpenSearch?
2. [23:29] Can you explain the role of operators in deploying OpenSearch in Kubernetes?
3. [24:32] Discuss the various security measures one can implement in Kubernetes.
4. [26:34] Have you integrated your pipelines with SonarQube for security, and what CI tool were you using?
5. [27:46] Explain how you can obtain central logs from any pod using Loki, Prometheus, and Grafana.
6. [28:00] What is the purpose of Ingress default backend in Kubernetes?
7. [29:22] If a junior team member approaches you with a pod not getting scheduled, how would you guide them in troubleshooting?
8. [30:46] Is there a way to provide external network connectivity to Kubernetes, and if yes, how?
9. [31:15] Describe how to forward the port 8080 from a container to a service, Ingress, and then to a browser.
❤️
Thank you, this is really helpful.
Few things I want to add here: 7:22 : Kubernetes have two components; control plane and worker nodes. If you say 'master node' rather than 'control plane', it might not include etcd. Because etcd can be setup external to master nodes.
The component kube-controller-manager doesnot combine 'all' the processes to be precise, but combines various controller processes like node controller, service account controller, etc.
Kube-scheduler not only schedules work for worker nodes, but for all nodes including master nodes. For example, pods inside master node is scheduled by kube-scheduler.
ETCD not only stores usernames and passwords, rather it stores state of cluster data, including states of pods, rs etc.
Also kubelet, kube-proxy and Container runtime (3 main components) runs not only in worker nodes but also in all the master nodes.
9:12 : Docker swarm has concept called docker service which is similar to deployments in kubernetes, so we can easily make sure the replicas count is correct. Also, we can perform rolling updates or rollback using docker service. The main difference between docker swarm and Kubernetes is a). docker swarm is tightly coupled with docker tools and ecosystem, but kubernetes can use other container runtimes including sandbox containers. K8s have large community and is highly extensible and customizable.
13:12 If we delete a pod or is evicted automatically, it will not always get recreated. If the pod is part of an replicaset or deployment, it will comeup.
13:14 Nodeport exposes a higher range IP on every nodes which is available external to the cluster. All the services provide load balancing. But load balancer service is used to provision an external loadbalancer if the cluster runs on cloud.
18:46 Namespace is used for logically divide and isolate resources within a cluster (not just users). For user per namespace, we can use rbac (namespaced rolebinding)
23:57: Important security measures in kubernetes: mainly four levels of security; cloud, cluster, container and code hardening (4 Cs). Apart from RBAC and proper service accounts, use admission controllers like pod standard policy, use network policies, secrets management using external secrets manager like vault or AWS secrets manager, use security capabilities and other steps related to image security.
Appreciate your feedback.
can you share you please opinion on how to shift your architecture from monolith to micro service?
Thanks for the free cka session 😂, do you also have a UA-cam channel? Because I am willing to subscribe to learn more.
Thanks
thanks you for clarifying, a good teacher you can make
One small correction here..scheduler doesnt actually schedule any control plan pods but it's a job of a Kubelet to pick up the configuration files and run it as Static Pods
You’re so Awesome! Thank you so much for doing this for us. This is very helpful
You are so welcome!
By the time I didn't knew about Kubernetes, I thought this girl is so much knowledgeable but now as I also have some knowledge I can see that she is just always using the architecture of Kubernetes in every scenario based question.
Thanks for the DM feedback. To be honest, this was a Kubernetes based interview, so, she was being judged on the same.
The architecture of Kubernetes is not a scenario-based question bro, The Interviewer needs to how basic you are aware of Kubernetes.
I agreed since she has 4 yrs experience in k8s the question should be hard.
@@00000201 afterall, she is a girl.
Excellent. Interview with lot of information on K8S real time based Scenarios. Really Enjoyed Watching :)
Thanks for the support, Azeez!
Thanks a lot for your all videos, keep continue this series, it's very helpful ❤️❤️
Sure, I am already looking for mock interviews and permissions from the people I have interviewed in the last year. Without their permission I can't post the videos.
Wow....... just amazing bro 🙏🙏
Good to see you again bro.congrats for your hardwork
Much appreciated!
He's back kkk with improved way. Thank you.
Thank you!
Thanks a lot bro for ur comeback 🙏🙏
Means a lot 🙏🏻
Excellent comeback
Glad you liked it 🙏🏻
Such a cool interviewer, no matter candidate gets selected or not but they will have pleasant experience.
This means a lot, Govind. Thank you.
Thank you for doing this, it really helps for a bunch of people preparing for interviews.
Means a lot, do share it with people, please 🙏🏻
Amazing , Excellent interview session
Glat to hear that!
Amazing content, keep it up❤🎉
Thank you, Samridhi. 🙏🏻🤝🏻
Thank you Bhai. Extremely helpful
Happy to help
Thank you very much bro
This is one of my favourite interview
I hope everyone give you permission and all the interviews will get back to channel
We hope so, too.
The way you are questioning and the way she is giving answers that really awesome and very informative Bhai....
Means a lot!
Good stuff, thanks 😊
Glad you liked it!
Thank you for this session. it was very helpful.
You're very welcome!
Finally, the wait is over
Yesss!!
really good work, good luck
Thanks a lot!
This is very helpful 👌
Glad you think so!
Hello,
thanks for sharing this video.
Pleasure
I'm glad to see you back with more improvement. One thing I would suggest is that whatever questions you are taking in your interview, you should record them in a Google Doc and add them to the video below.
We should try voice to text tool
You can use Show Transcript function in UA-cam
Can you please let me know one?
I think that's on. If not, I will check.
@@LogicOpsLab Hello sir , Show Transcript function in UA-cam in not on video sir can u enabled sir this help us.
It's going to be very informative
Thank you, Sagar!
Good to see you back, kindly do a video on how microservices expose in dns or route 53 little confussion on this pls do a video
Can you explain me the exact problem statement, please?
@@LogicOpsLab i mean i am unable understand different pods have different load balancing services with unique ips, i mean how all of them included in route 53 or any other hosting domains
Hey Bro,
This interview really increase my interest on K8s, Her communication skills very attractive to every one and her reply of every question very clearly explained. Thank you very much to bring her into your videos bucket, I watch your videos very helpful to us. I hope you provide more videos to us those of are help to increase my knowledge.
Thanks for all the love. The candidate was indeed a great one. There are 50 more videos in this playlist. Do subscribe and like to support. Much love ❤️
such an informative video. Kindly post the answers for those questions which are un answered/guesses/may be u feel incomplete during interview.
Almost every answer was correct except one if I remember correctly.
Anyone who is struggling to answer the questions .
Study for CKA and then CKS, all these topics are mentioned in syllabus and you need to study them in detail, you will do a lot of hands on also which will boost your confidence
Correct.
My CKA expired.. had an aspiration to do CKS.
Thank you... Finally this series is back🙌🔥🔥
Yes, more to come.
thank u sir its very helpful for us
Appreciate it, Suraj.
few things i have noticed in this interview:
1. The person is using words pods and nodes interchangeably. That is a bit wrong and confusing given the experience. We should try to be precise on what we are saying.
2. on every start of an answer the person says k8s is the solution for the problem that might irritate some interviewers. bcz rather than answering the question directly we are playing with words.
3. we should try not to get off topic too much. As when question about crd was asked the person went on to opensearch vs elasticsearch. And the info provided was a bit wrong also, so if the interviewer knew about the products in details they can trap you with questions which might give bad image.
other than this the candidate was very good. definate hire.
thanks a lot for the video.
Thanks for the feedback.
Exactly same what I also felt. It was too irritating to see for every scenario the same answer that answer is Kubernetes
Port forwarding can be enable using the below cmd from local port no 80
kubectl port-forward ingress 80:8080
Thank you Sir 🙂
Glad you liked it.
bro i thought that there would be really advanced questions but im happy if floks with 4 years of expertise got these kinna questions then I hope the internship interview will be a bit easier for me. Thanks a lot for informative content.
Glad I could help. Yes, internship level questions are not that difficult. Just basic concepts with one or two scenarios.
Best Kubernetes interview ever! Waiting for the other ones.
Thanks a bunch 🙏🏻
@@LogicOpsLab plz upload azure admin interview questions also plz...
When she said locki my thoughts shifted to the Marvel universe 😂.
Jokes apart very good interview 😄
Hahaha! Thank you. Loki is just another tool, I haven't heard about it too much either.
Honestly this is L1 level interview. All theories.
Thanks for the feedback
Helpful
Thank you
thanks alot ..
You're welcome, Shiva.
I know a little bit about Kubernetes and have been interviewed once for Kubernetes. From what I she is spot on
Indeed. Thanks for the feedback.
Good Job buddy..you have covered entire Kubernets..
Can you answer port forwarding exactly how can we do 8080 to 80 from browser?
Port forwarding from port 8080 to port 80 in a Kubernetes cluster can be achieved using a Kubernetes service and the kubectl port-forward command. This allows you to access a service running on port 80 in a pod by connecting to port 8080 on your local machine.
kubectl port-forward service/my-service 8080:80
Remember that this is useful for development and debugging purposes, but it's not suitable for production traffic. In production, you'd typically use a load balancer or an Ingress controller to route traffic to the appropriate service and port.
@@LogicOpsLab Thanks you for response buddy
thank you sir for all the videos....please how do we migrate from a monolith environment to a micro service environment ?
That is a very broad question. What exactly is the problem statement?
Answer for the last question:
How to external traffic will flow to the container inside the pod?
Ans: First, we will have, for example, traffic that comes from was AWS load balancer and is distributed across different worker nodes(e.g. ec2) from there, it will go into the ingress controller, and then the ingress pod will redirect the traffic to the proper service which is configured ingress.yaml from there service will talk to kube-proxy, and then kube-proxy will intelligently distribute traffic to pods. Then from the pods, it will go to the container.
AWS Loadbalancer > Node > Ingress > Service > Kube-proxy(select pod) > Pod > container
Note: We need to open a port for ingress in the security group of the worker node.
Also, ravish bhaiya, if there is any internship/job opportunity for me, do let me know, Mostly around "Kubernetes and DevOps".
Please DM on LinkedIn, I will see what I can do.
One suggestion. The car pooling app can scale their apps using the autoscaling features in K8's .
Horizontal and vertical scaling.
Please correct me if I am wrong
Here's how the carpooling app can use Kubernetes to scale:
1. Containerization: The first step is to containerize the carpooling app's services and components. This involves packaging each component into lightweight, portable containers using technologies like Docker.
2. Cluster setup: Set up a Kubernetes cluster, which consists of multiple nodes (servers) to run and manage containers. Kubernetes can be deployed on-premises or on cloud platforms like Amazon Web Services (AWS) or Google Cloud Platform (GCP).
3. Service deployment: Use Kubernetes Deployments or StatefulSets to define the desired state of the carpooling app's services. These resources specify the number of replicas (instances) to be running at any given time.
4. Autoscaling: Kubernetes offers Horizontal Pod Autoscaler (HPA) functionality, which automatically scales the number of replicas based on specified metrics (e.g., CPU utilization, request rate). This ensures that the app can handle increased demand without manual intervention.
5. Load balancing: Kubernetes provides built-in load balancing through the use of Services. Services expose the carpooling app's components internally within the cluster and externally to users. The service load balancer distributes traffic evenly among the available replicas.
6. Rolling updates: Kubernetes supports rolling updates, allowing new versions of the app to be deployed without downtime. This process ensures that the carpooling app is always available to users during updates.
7. Node management: Kubernetes can automatically manage the allocation of containers across different nodes in the cluster. It helps optimize resource utilization and ensures that the carpooling app scales efficiently.
8. Monitoring and logging: Utilize Kubernetes monitoring tools and log aggregation mechanisms to gain insights into the app's performance and identify any issues that may arise during scaling.
9. Resilience and high availability: Kubernetes enables the deployment of replicas across multiple nodes and availability zones, ensuring high availability and fault tolerance for the carpooling app.
10. Continuous integration and deployment: Implement CI/CD pipelines to automate the process of building, testing, and deploying updates to the app in the Kubernetes cluster.
what is meaning of external network connectivity ? you mean between nodes ? if between nodes we can use weavenets, flannel, cillium, etc
Kubernetes services like Node port, Cluster IP, and Load balancer provide networking between different components within the cluster and with the outside world.
Weavenet provides a network to connect all the pods, it creates a virtual network that creates them.
Flannel manages New York between multiple nodes in the cluster, it doesn't control how containers are networked to the host, only how the traffic is transported between hosts.
Calico enable Kubernetes workloads and non Kubernetes or legacy workloads to communicate seamlessly and securely
3 different services - 3 different work roles
Glad you posted this interview !!! But she doesn't know the actual concepts of kubernetes.
Hi, Sunil. Thanks for the feedback.
Feel free to add all the answers you think or feel is not correct. That'd be really helpful for the community.
bhai sanjay dahya ka bhai he kya tu? 😂
Node Port service will enable the port and port number will be in range of 30000 to 32767 and this port number will be mapped with the cluster IP address !!
Thanks for the feedback.
Can you please upload another DevOps real time interviews
We will, taking permission from everyone.
Brother we are still waiting for other videos as well ...please
We are in a process of getting permissions from the people. It's is taking some time. I am uploading it one by one. Thanks for your patience. 🙏🏻
Can we expect this type of questions for experience devops professional interview on kubernetes?
Yes, possible. But, there are other topics, as well.
I dont intent to criticize but how does the candindate say she has used kubernetes at the beginning then later says she has not used any kubernetes services she only knows about them?How? how now?
Not sure how I missed that part. Can you please point me to both timestamps?
Hello How the Kubernetes Rollback Automatically we need to roll back right if there is any issue.I am Taking about Application upgrade.
If there's any issue then there would be a manual roll back.
@@LogicOpsLab But She mentioned Automatically Rollback right how is that possible.
@@akashan6811 Time stamp?
@@LogicOpsLab 10:20 Onwards when differentiating Docker Swarm and Kubernetes.
@@akashan6811 If we want automatic roll back it requires running kubectl commands from within the pod. You can use a script to continuously monitor your deployments and when an error occurs you can run "kubectl rollout undo deployment DEPLOYMENT_NAME"
sir can you please explain operator question which you asked?
Timestamp?
@@LogicOpsLab 21:13
Hi bro . We are not getting new videos from you .. can you pls make it interview point of view
Working on it, mate 😊
🎯 Key points for quick navigation:
00:00 *🎥 Introduction to Kubernetes interview series*
- This section introduces a series focused on Kubernetes interview scenarios and insights.
- Emphasis on gathering permissions for interview recordings and filtering content.
01:10 *🌟 Highlights of a Successful Kubernetes Interview*
- Detailed overview of a successful Kubernetes interview session.
- Focus on candidate expertise and comprehensive Kubernetes knowledge.
03:04 *📚 Candidate's Professional Background*
- Discussion on the candidate's educational and professional journey.
- Experience timeline from a product-based company to Cloud operations.
05:30 *🛠️ Transition from Monolithic to Microservices*
- Insights into transitioning architecture from monolithic to microservices.
- Role of Kubernetes in facilitating this architectural shift.
07:24 *🏛️ Overview of Kubernetes Architecture*
- Detailed breakdown of Kubernetes architecture components.
- Explanation of master node and worker node roles.
09:27 *🔄 Kubernetes vs. Docker Swarm*
- Comparison between Kubernetes and Docker Swarm for container orchestration.
- Advantages of Kubernetes over Docker Swarm in terms of scalability and management.
12:02 *📦 Deployment Challenges in Microservices*
- Addressing deployment challenges in microservices architectures.
- Role of Kubernetes in automating deployment processes.
15:21 *🌐 Load Balancing with Kubernetes*
- Implementing load balancing solutions within Kubernetes.
- Internal and external load balancer functionalities explained.
17:58 *🚀 Scaling Infrastructure with Kubernetes*
- Strategies for scaling infrastructure using Kubernetes.
- Integration of monitoring tools like Prometheus for capacity planning.
19:21 *📂 Namespace Management in Kubernetes*
- Importance of namespaces in Kubernetes for resource partitioning.
- Practical use cases and benefits of using namespaces.
21:09 *🛠️ Kubernetes Operators*
- Role and capabilities of Kubernetes operators in managing applications.
- Custom resource management with operators for enhanced automation.
22:06 *🛡️ OpenSearch vs. Elasticsearch Overview*
- Comparison between Elasticsearch and OpenSearch,
- AWS's creation of OpenSearch and its relation to Elasticsearch,
- Usage of Helm charts for OpenSearch deployment and the introduction of Kubernetes operators.
23:29 *🛠️ Kubernetes Operators for OpenSearch Deployment*
- Explanation of Kubernetes operators,
- Their role in automating OpenSearch deployments,
- Contrast with manual Helm chart deployments.
24:32 *🔒 Enhancing Kubernetes Security*
- Implementing Network Policies to restrict pod communication,
- Role-Based Access Control (RBAC) usage in Kubernetes,
- Leveraging namespaces for segregation and access control.
26:48 *📊 Centralized Logging in Kubernetes*
- Methods for centralizing logs including node-level logging agents,
- Integration with Grafana for visualization and analysis,
- Exporting logs directly from Kubernetes pods using sidecar containers.
28:00 *🌐 Kubernetes Ingress and Default Backend*
- Explanation of Kubernetes Ingress and its role in routing incoming requests,
- Importance of defining default backends for unhandled requests.
29:22 *🔍 Troubleshooting Kubernetes Pod Scheduling Issues*
- Steps to troubleshoot pod scheduling issues,
- Using `kubectl describe` to diagnose pod startup failures,
- Adjusting resource allocations to resolve scheduling problems.
31:15 *🔄 Port Forwarding in Kubernetes*
- Forwarding port 8080 from a container to a service, Ingress, and finally to a browser,
- Using Ingress controllers to expose ports externally,
- Configuring selectors to manage traffic forwarding effectively.
Made with HARPA AI
K8s has master nodes and Worker nodes? and this candidate is an outstanding candidate? Okay...Understood
Back again so soon? Who has your laptop? What's your excuse this time?
If telling you the truth and saying sorry looks like an "Excuse" to you then I feel pitty for you. This is me...:) I have my laptop this time and I am commenting now. All the other interviews you have uploaded are just awesome and the answers are so on point.I mean they are sso accurate that I suggested your channel for interview preparation to more than 10 of my friends. Even the interview on terraform helped me to crack first round in one of the MAANG companies. The answers were so accurate that everyone can follow that. But this interview was not at all upto the mark that someome can follow before their interviews. Standard of your videos are really high.
Again, that was not an excuse. I told the truth that what exactly happend and who made such bad comments that day. Anyway, thanks for replying.
can you please upload resume? Appreciated.
Checkout this video
ua-cam.com/video/hAFGfvc82x4/v-deo.html&ab_channel=LogicOpsLab
Sir I have gone through whole interview, And I have gone through rest of your interviews as well, questions are great no doubt, but this interviewee is most techincally incorrect in most of her answers, I dont know why you consider this one of your best interviews
Hi Jagmohan
Thanks for the valuable feedback. There are many other factors on which the candidate is judged.
I do value the opinion of the folks watching, thus, please do take time to let me know the list of questions or answers which she answered incorrectly (with timestamp). Also, what do you think is the right answer to the question. Please do let us know that, as well.
This will help a lot to improve the content and fellow community members, as well. Appreciate your help in advance.
@@LogicOpsLab Yes Sir, I checked the comment of @hassimnaseef , he has answered most of the wrong answered questions greatly, I will also definitely add answer if they are some left .
Please do. Thanks in advance.
can you share resume ?
Resume of whom?
Good interview..but questions was not so in-depth as 4 year’s experience should have...also she told wrong command for creating namespace…its not “-n” flag, instead its “kubectl create ns
Overall interview was good
Thanks for the feedback, Abhijeet.
Can you share rest of the interview rounds of her?
Thanks in Advanced
Some basic questions are answered wrong.. better check once if you can find them atleast
Thanks, Lohith for the feedback. How about helping the community by answering them?
Or letting us know which of them are answered wrong?
Please write the timestamp and the correct answer(s). This will help others improve, as well.
Thanks!
@@LogicOpsLab 14:17, NodePort is publicly exposed.
I would like to add one point, there in one question you asked about operator in Kubernetes and the girl started telling about OpenSearch in Aws which do not have any relation to the question.
Rbac or Namespaces do not limit pod communication. It can only be done by network policies
Bhai I need end to end devops project bro
Please check Real Time Project series, you will find the projects there, all end to end.
Still she did not meet the ans of of first question why microservices and not monolithic.. 😂
Thanks for the feedback.
With limited knowledge she is elaborating things which is good.
Thanks for the feedback, mate!
First of all thanks for sharing the video but I assumed there were too many repeated questions, please try to avoid this. As an interviewer, this is my suggestion.
One correction related to the different components of the control plane and worker Nodes?
The correct answer is below.
Components running on the control plane node
-----------------------------------
API server
Controller manager (Node controller, job controller, endpoint controller, and Service account & token controller)
etcd
kube-scheduler
Components running on all the nodes
------------------------------------
kube proxy
kubelet
container run time.
That means kube proxy, kubelet, and container run time running on the control plane as well as the worker nodes.
Now question why kube proxy, kubelet, and container run time running on the control plane. Are any pods running on the control plane node?
The answer is yes system defines pods running on the control plane node. Using this command easily identifies
""kubectl get pods -n kube-system"" .
2nd thing is not scenario-based questions?
You are asking these questions for your own satisfaction and try to avoid this. I understand scenario-based questions mean problem-solving questions. I don not find anything in this video.
Finally best of luck.
Regards,
Pinaki Mukherjee
Appreciate the feedback, Pinaki. It's constructive criticism and I take it whole heartedly. Though, I have tried improving it in the future videos. Kindly take a look, any feedback would be appreciated.
Wish you the best, too. 😊
Ravish