跟读练习: Excellent 2 Years AWS DevOps Engineer Mock Interview with Questions and Answers | Cloud | Docker - 通过视频学习英语口语

hi welcome sarap to the interview

and yeah i'm putting a cv in front of me

and yeah so just to let you know

that this call will be recorded uh for on the social media like youtube and others

and that you are okay with

that yeah yeah i'm okay with it okay all right uh

so yep uh i've got your cv in front of me

so why don't we start up with your career progression and

and then we'll have some follow-up questions yeah like first of

all thanks a lot for this uh opportunity you know i

really appreciate it uh my name is uh i will start with how i got into the devops and cloud

technology domain so i did my bachelor's of engineering in mechanical field but whilst i I was in third year, I got interested into the computer science.

So then I started learning about the Python, basics of Python.

And then later during the COVID period, that's when I met my mentor.

He taught me a lot of DevOps and cloud technologies.

And the most important thing he did was not only taught, but he created my interest in that field genuinely.

So I really followed that.

And whilst I was in fourth year, I got my opportunity in one of the largest consultancy company in india

and since then i've been working there as a devops

and cloud engineer uh for two and two year

and four months as of now

so that's a little bit about my work uh related things

uh about my personal things i live in the town of

maharashya uh my hobbies are like i like to read books i also like to work out

and just stroll in the nature

so yeah that's a little bit about me awesome awesome all right

so i see from your resume with sort of that you have

experience in aws code pipeline cicd python

and lambda also used amazon quicksight for visualizations and dashboard

so how about uh how about docker

and kubernetes yeah i also know a docker

and Kubernetes okay all right all right let's start up with some of the aws questions

so i'll start right from the basics uh

so in terms of aws security

so there's something called as knuckles yeah uh can you tell me more about knuckles

and how does it how does it differ from subnets yeah

so knuckles are the firewalls uh applied at the subnet level

uh there is another thing called a security group in aws

which somehow relates to it secret but there is a difference

that knuckles are stateless so it means

that you have to specify specifically the inbound as well as outbound rule whether as the security groups

which are attached at the individual resource level like instances uh they are stateful

so you just have to mention like inbound and they will automatically gather

that outbound is also allowed so that's what an actor is right

now in terms of your ec2 uh there is something called as volumes which can be attached

right yes you've got elastic black block stores

and volumes okay what are the different types of volumes and

and what they're used for yeah

so like volumes are the like buses are mounted for having the persistent storage to our EC2 servers

and the block storage is the EBS that AWS offers

in terms of different types of volumes so there are like GP2

which are like SSD based then there are IOPS heavy volumes

so where you can have a if you want the have application

that has heavy read io then you can use the iops

type of volumes then there are also some machine learning related uh volumes i guess i don't remember them particularly

but yeah so these are some types of volumes so

i'll just give you a scenario and you can explain me on the on the questions on it

So let's say you've got a three tier or a four tier application.

And you are asked to design the application from scratch.

So you've got the app layer, you've got an API layer, you've got the business logic, you've got web layer, you've got databases, you're using S3.

So these are the same sort of application.

So in terms of cloud security how would you secure an application

and what are the different attributes

that you would have in your application to make it uh consistently secure okay

so let's start with the things that i will include uh

so first is database layer second is application layer

and third is suppose the client accessing the things from internet

so that's 3d

so i would probably use a api gateway application load violancer

in this architecture where i will put just the alb

or api gateway in the public subnet where it has access to the external internet

and i will put my app server in different private subnet

and i will put my database server in different private subnet

and then just connect my api gateway alb to the private web server

and web servers will have access to the database servers privately

and yeah like and the clients can access my application on on the ip address

or domain name of the alb which are mapped to it

so that is how i would uh like architect it okay are there any other security measures

that you can think of yeah like we can have the like

if your web server needs access to the database

and suppose your database is in rds

so you can have a proper im permissions i am in instance profile for your instances in web servers.

Also, you can have proper knuckles, the security groups following the least privileges rule as the AWS suggested.

So those are some of the security things that you can take in consideration.

Anything else that you can think of?

You can also add the AWS WAP or something like that

if you want the extra security if you want ddos protection

so maybe you can use vap and uh

if you want to like check the your instances for like uh security vulnerability then maybe use the aws inspector

or something like

that okay last chance to say something else ah looking for one more point you did well

but one more point there is uh ppc there is subnet it's alb uh security groups is there im is there

uh maybe like storing the database passwords or something like

that you can use the secrets manager for

that yeah how about how about you how would you secure a data tell me about data encryption So data encryption,

if it's in transit, then you can use the obviously SSL for that SSL certificates.

And if it's at the rest, then you can, in terms of AWS, you can use a service like AWS KMS, where you can have different options for managing the keys.

Either AWS can manage it or like you can manage it and you can encrypt the data at rest using both keys.

Okay.

100

Can you tell me the difference between server-side encryption

101

and client-side encryption yeah i would say server-side encryption is something

102

that uh we do like as a app server where like suppose data is address

103

so that's where you are uh encrypting your data

104

so that's i would say server-side encryption and client-side encryption is something

105

that if client is sending the data then you're using the ssl uh key that

106

public key to encrypt the data so i would say that would be client-setting okay

107

all right uh now coming back to tell me tell me why would you why would you need auto scaling

108

and what are the different types of auto scalings available uh

109

so auto scaling is required if you have a varying terms of traffic to your applications

110

and you want to automatically scale your instances

111

or containers whatever uh type of deployment you have done and uh

112

so you don't need manual intervention suppose the traffic is uh exceeding beyond uh what your cp levels can handle

113

so you don't need to manually increase go

114

and increase the number of servers and set up everything

115

so for that we can use something like auto scaling groups where you already have the pre-configured launch templates

116

that uh and you can give it a various metrics

117

so asg can scale based on like a metric base

118

so like uh suppose average cpu utilization it can also you

119

like scale based on maybe number of requests going to per instance

120

so that is another thing

121

or you can create your own custom metric uh whatever you want and then scale based using that metric for your needs.

122

Okay.

123

Now, have you done any sort of monitoring on any applications?

124

I have not done as per se, but I use CloudWatch whenever I like require to like check something if something has happened or something,

125

but I haven't done specifically monitoring.

126

I do created the dashboards using quicksight uh uh

127

which is the native aws services where we have monitored the like things like cost of our accounts

128

and like uh different s3 bucket size

129

and everything else okay all right can you explain me this term connection draining connection draining

130

so basically it means

131

that your connection is going to die like it's it's finishing up all the things

132

and it's draining it's removing the connection so what does it do

133

uh connection training uh you're talking in terms of specific something

134

or just general term connection so so if an application is going to die what does the connection draining do

135

uh so it if there are any pending requests it will finish up those requests

136

and then it will go down it will close

137

that connection okay yep right all right uh now let's talk something about on the docker

138

and kubernetes so we do we do have docker can you explain me the role of kubernetes in docker

139

so kubernetes the docker container orchestration platform what it means is

140

that it comes on top of docker

141

so docker natively does does not provide things like auto scaling auto healing

142

or uh maybe load balancer type of things uh

143

which the any enterprise level of application would require

144

so that's where a kubernetes come it adds a lot of features

145

that an enterprise require in order to operate their application uh docker does have something called docker swarm

146

but it's not i would say uh like uh fulfills the

147

requirement needed for the enterprise level of application you can use it for a small application

148

but i won't say for enterprise okay

149

so now in kubernetes there is something called as image pull policy yeah okay can you tell me about that

150

so image pull policy is uh suppose you have created a docker file

151

and you want to specify your docker file that

152

when do you pull the image so you can mention in image pool policy

153

that if the image is not present locally then you can pull it from the registry

154

okay and what are the different values that it accepts ah i know one

155

because i have used that one that is

156

if not present in current repository

157

but for others maybe i'll need to like look up in documentation yeah that's all right

158

so there are two more always and never yeah yeah all right

159

so what do you understand by service in kubernetes and how many different types of services are there

160

uh so service is used uh to offer features like i said load balancing It also has a feature called auto discovery.

161

So it automatically discovers the new instances coming up in your cluster for that specific application.

162

And it does that by using the thing called as selectors and labels.

163

Our services has three types.

164

The first is the cluster IP, which is default.

165

It means that whatever service you are exposing to the external world, sorry, whatever services you're exposing, it's only available within the cluster itself.

166

Second is the node port, which means that the services are available to the clients

167

or the servers that has access to the node on which the Kubernetes cluster is running.

168

And third is the load balancer where the cloud providers come

169

in picture where you can expose it to the external world

170

by using external application load balances like alb all right now in kubernetes you have got deployments and you've got demon sets

171

yeah can you tell me the difference between both of them

172

i remember deployment i don't remember demon set to be honest the deployment is basically for uh

173

if you want to deploy your application

174

and have the capability as i say auto healing or auto scaling so that's when you can use our deployment In deployment,

175

you can mention different strategies like rolling deployment or maybe blue-green deployment, something like that.

176

So that's what we use the deployment.

177

I can't remember Devon set.

178

I have read about that, but I have not used it.

179

All right, so deployment makes the pod available in any node, but Devon set, it makes sure that the replica runs in each and every node in the custom.

180

Okay, okay.

181

Okay.

182

Now about RDS, can you tell me the different types of SQL and NoSQL databases and what's the difference between them?

183

RDS is the service offered by AWS called Relational Database Services.

184

So relational means the SQL, which means the structured databases.

185

So that has various things like PostgreSQL, then MySQL, MSSQL.

186

like that.

187

And the other type of database service is the NoSQL type, which is things like DynamoDB or like MongoDB.

188

So where your data is not structured

189

and the data is stored in a key value kind of format rather than the structured column-wise format in SQL.

190

Okay.

191

Any more differences that you can think of?

192

You can like, NoSQL is very scalable, like DynamoDB, you can store like millions and billions of record in it.

193

And it's very resilient in terms of it, whereas your MySQL is not meant for that purpose.

194

all right

195

so now coming back to your experience can you tell me

196

about a about a scenario where you really struggled in some of the cases

197

or something that was very hard to do and then what happened to

198

that finally did you achieve it or not okay um

199

so yeah um like recently there was a once as i might have mentioned in my resume

200

that i've also worked on security things so there was recently one finding

201

that came that our clients had lots of lambda functions running on the aws lambda

202

and all of them were using the runtimes

203

which were deprecated by the aws so

204

that means there were no security patches nothing

205

and it could be vulnerable to the security vulnerabilities

206

and they were around 1300 functions which our clients had on lambda

207

and uh no one like i took the responsibility to upgrade it then uh

208

and 850 were around pythons and rest of the were of node.js

209

so for node.js i took help of someone else in my

210

team who uh has the experience on node.js whether as python i myself have uh the experience

211

so i did them on my own so the challenges i faced in

212

that was their lambda so there are two times the types of functions

213

so one that uses the layers and one

214

that does not use this layer so the one

215

that does not use this layer it was easy to upgrade them

216

so i just created a simple python script and

217

because the python syntax for 3.7 and 3.9 was almost similar there were no differences

218

or nothing that needs to be changed so

219

that i completed like in just few days

220

and then there were the next set of functions

221

that were the layers one saying

222

that i had to like done do some manual things also where like i automated the process of creating

223

and updating layers by using the aws cli so

224

layers are updated

225

but then i had to contact to the various team those were handling those applications uh those functions

226

and then yeah so it was like a big process

227

and at first i thought like 1300 functions that's a lot

228

and like how can i complete this but it needed to do

229

that so yeah i completed it and i'm almost at the end

230

so i have around 150 functions remaining out of those 1300

231

and yeah i will be finishing it in just few days okay that.

232

All right, I think I'm done with the questions and answers.

233

So my feedback to you that you did pretty well.

234

You answered almost all the questions.

235

And yep, so I'll be closing up this recording now.

236

And I'll be giving you more feedback after the recording is done.

237

Thanks for your time.

238

Yeah, yeah, thank you.

239

Thank you.

✨ 推荐视频

154 句

Running AI Agents in Secure Sandboxes with E2B & Docker MCP | Docker’s AI Guide to the Galaxy

University StudentsBusiness English+1

405 句

Netflix product manager mock interview and tips (with ex-Netflix Lead PM)

University StudentsBusiness English+2

11 句

full audio online video cutter com

University StudentsBusiness English+2

392 句

Uber Data Engineering Mock Interview - Ride-Sharing Data Warehouse Schema

University StudentsBusiness English+2

198 句

2025 DevOps, SRE, Cloud Engineer Interview Questions: Ace Your Interview with This Guide

University StudentsBusiness English+2

什么是跟读法？

跟读法 (Shadowing) 是一种有科学依据的语言学习技巧，最初开发用于专业口译员的培训，并由多语言者Alexander Arguelles博士普及。这个方法简单而强大：您在听英语母语原声的同时立即大声重复——就像是一个延迟1-2秒紧跟说话者的影子。与被动听力或语法练习不同，跟读法强迫您的大脑和口腔肌肉同时处理并模仿真实的讲话模式。研究表明它能显着提高发音准确性，语调，节奏，连读，听力理解和口语流利度——使其成为雅思口语备考和真实英语交流最有效的方法之一。

影子跟读法: 阅读完整分步指南 →

跟读练习: Excellent 2 Years AWS DevOps Engineer Mock Interview with Questions and Answers | Cloud | Docker - 通过视频学习英语口语

Manual Testing: QA Interview Preparation Questions and Answers

How to ACTUALLY Get a DevOps Job in 2026

Next JS Interview Questions and Answers 2026 | Top 20 Most Asked Questions

Data Analyst Interview 2024 | SQL | POWER BI | Data Science | Mock Interview

Terraform Scenario Based Interview Questions and Answers | DevOps Interview

Distributed Task Scheduler: System Design Interview (Stripe & Amazon Offers)

Job Interview English 🇬🇧 | Real Conversation Practice (B1–B2 Level)

Running AI Agents in Secure Sandboxes with E2B & Docker MCP | Docker’s AI Guide to the Galaxy

Netflix product manager mock interview and tips (with ex-Netflix Lead PM)

full audio online video cutter com

Uber Data Engineering Mock Interview - Ride-Sharing Data Warehouse Schema

2025 DevOps, SRE, Cloud Engineer Interview Questions: Ace Your Interview with This Guide

为何这个视频是绝佳的英语口语素材？

拆解自然表达，学习地道口语

影子跟读练习法，快速提升口语

什么是跟读法？