I loved what you said about the knowledge level / experience level conundrum. Reassuring to hear for sure. It's important to recognize that not every engineer, whether SRE or software or other, has the opportunity at the beginning of their career to work on large-scale systems or even interesting & challenging work, for that matter. Plenty of people get stuck at dead-end jobs writing CRUD apps or never have mentors or exposure to anything neat until they seek better employment. Fortunately, I got some decent experience early on but I also worked at a place where I was told they needed sandwich makers, not the people who design sandwich makers, and that I should just copy/paste existing stuff to bang out tasks without any concern for code improvement or good SWE practices. That's a real crappy way to manage talent if you ask me.
few questions to gauge logging experience 1. how do you standardize the logs in distributed environment - Answers expected would be related to mutation , filtering (gork) in order to standardize 2. what tools are available out there for a infrastructure team / person to monitor logs and logging mechanism itself 3. how to you manage log archival 4. how do you control log ingestion at server level and log creation at microservice level -> which again relates to standardization, expect someone talking about log rolling , file limit and at server level queing the logs , forwarding using logstash/filebeat etc.
Nice vid. One example of logging is to push data via AWS Lambda function which can be used to stream events over HTTPS to Splunk HTTP Event Collector . Quite powerful
You just spot on! Thank you for your videos!!! I'm using your videos as part of the reference for the interview preparation because initially, a stack of the tech not going to change much and I will be asked the same questions over and over again until I pass the interview) The thing is that without at leas a baaaaasic hands in to cloud experience all of the theory is worthless...
Hey man, i just want to let you know that your video is helpful. I got a job offering after reviewing the material on this video. More importantly, your word of encouragement at the end give me a huge boost of confidence during interview. Thankyou!! i hope you have a nice day
Had my first DevOps interview this morning! No technical questions at all... 🤦🏻♂️ Prepped all night for it too. Thanks for your content dude, I wouldn’t have got the interview without following your stuff and your course over on Udemy. 🙏🏻
Poojitha Mittapally I was just asked about the tech I’d used previously and how I’d used them (I don’t have a tech background so it’s all self-taught). We touched on whether I knew azure pipelines too. Based on what I’ve read my interview wasn’t typical though. This is also a company I did a Python assessment for just a couple of weeks ago.
@@tutoriaLinux I am going to piggyback off of this. I would love to hear more in-depth conversation about the logging. Specifically, what would YOU do if you were asked about logging?
Well I'm less nervous about my interview on Tuesday. I'm brand new to AWS, and my second interview consists of 2 Senior infrastructure engineers an a sys admin for an hour , then get to meet the CEO and the director of infrastructure. Pretty nervous, never really had an interview like this.
Maybe one logging-related question: What are the pros and cons of "sidecar" solution and "daemonset" solution when it comes to collecting and sending microservice logs (with fluent bit for example)
For a logging / observability I'd lean towards asking about Golden Signals, white box vs black box, what do averages tell you and what don't they, etc. Could Also ask a scenario where the companies having a hard time correlating logs to the question everybody asks what is this user doing. Assuming this is a fresh logging stack, what kind of improvements would you suggest? I think the elephant in the room here is that you always need some kind of cardinality like a UUID or something like that.
Love these videos so much and I'm not planning to have or conduct an interview any time soon . I just have one question: Why do you have, what I can only assume, is a meth lab in the background? :D
Haha, I love scientific lab-glass-looking things. Somewhere in my subconscious I guess I fancy myself a bit of a mad scientist. Those are some flasks that I picked up for $1 each at some point in my life :-D
About logging I've been working with data dog over hybrid cloud and though it has it glitches here and there it work very well with k8 and can do Apm nicely
Thanks for the comment! Yeah, datadog has its warts but it's usable if you put in the time and effort. Thankfully I have a few people on my team who are more interested in making datadog behave than I am -- I've been benefitting from their work :-D. Also I just implemented some tracing from a Go application, making it talk to Datadog APM, and it was a reasonably good experience! It's no stackdriver, but it's good.
I always get nervous and forget how to answer in project mode, The reason I had a job with one company for too long not knowing whats out there, Now I am jobless for past 2 months and market is slow in toronto
Do you have any links you could provide for professional Devops Engineering courses/info. Specifically how to set up & manage a codebase, from a messy set of gits & exe stores to an organised self compiling distro for differing architectures ? Thank you.
There’s huge demand in the U.S., companies are having serious problems filling cloud engineering roles with competent people. If you use the “devops” keyword you’ll find a bunch of them. It pays better than “linux engineer” positions, too: Juniors are 80-100k/yr, mids around 130-150, seniors 160-200 base.
@@tutoriaLinux thanks for the response. To expound on my initial reply, I am a novice in the cloud world with a little hands on experience and a few AWS certs. The good old saying is it the chicken before the egg or vice versa. Would a company hire someone with limited experience? Why would they? How does someone get experience that counts and land a junior gig!? Not just like hey I do it in my free time and that’s my current experience. Thanks
Not sure what the context is - How many places have you interviewed at? How many have you applied to? Where are the interviews going wrong? If they’re rejecting you, try to find out why. If you’re overthinking this before applying for at least 30-40 devops/infra gigs, then don’t worry too much and just start applying. Companies hire newer engineers all the time, otherwise no one would have ever gotten started. If you’ve been job hunting for a while and you’re not getting interviews, find a tech recruiter in your area and work with them to land your first gig. Meetups are an option too.
Hey @TutoriaLinux, I just got my aws saa cerification. I have no prior experience in IT whatsoever. What other advice would you give me in terms of resume preparation, interview question since I'm so new to this?
Just start interviewing! Think of it less as an attempt to reach a specific goal (one job), and more of a learning experience like setting up a server or troubleshooting a tough problem. Start applying to jobs, see how the first few interviews go, ask for feedback about your strengths and weaknesses, and adjust your learning/practice accordingly. Don't take anything personally (easier said than done) and just focus on the process.
Is this for a senior devops engineer position? I don't see how anybody who has not been in the cloud space before could find a way to answer these questions. How are people going to migrate over from their current on prem roles to the cloud space where people are really needed when they have to answer such in depth questions like this?
Dave, awesome as always. In your last video, you said you'd sit the person at a terminal and ask them to do things so you can see if they know what they're doing. Do you try to get them to do some cloud infrastructure in person, or is that too much of a burden for an interview?
Hi, I am working to get into the IT field after being in it back in the 80s. Now, I would think that is how current interviews are done. "Hey show us how to set up a simple VPC from this laptop". Go! That is what I am striving for in my studies. To be able to present that skill set when asked, and to be able to talk through it. Both from the command line and the console. I know vpc would be difficult from the CLI, but not an S3 bucket and such.
I think surfbug's approach is a good one. Learn and study in a way that gives you the practical skills you need. And I agree with both of you; I think practical "set up this infra" problems are a cool idea. THAT BEING SAID: in 10 years of interviewing for jobs, I've never actually had anyone ask me to do that. I think these things can get pretty time consuming -- hunting down the last aws console checkbox or terraform resource argument that's missing, the last syntax error, etc. Interviews are extremely time-limited and I suspect that's one of the reasons I haven't seen this style of interviewing. FWIW software dev interviews tend to be more like that -- less theory, more practical application. But the complaint *there* is that the toy problems that are suitable for interviews are often very far removed from "real" programming work, etc. etc. etc. lots of long arguments all over the web about this. Short answer: it sounds like a good idea. Maybe it is! I'm not sure.
@@tutoriaLinux thanks for the feedback! Testing for my assoc architect on oct 22.... Let's see how it goes! Next step are those interviews, and thanks for the info! Subbed and shared, you have a new fan....
@@tutoriaLinux My manager always care about the personality and character rather too much digging on technical side. He said a person can learn our system but we can't change someone attitude
splitting the database ?? my gawd my head is spinning already...and I have been in IT for decades but not exposed to that... I hate whiteboarding thank god for remote jobs interviews ..lol
You won't need any special infrastructure for front page of HN, not really a useful example. Splitting off the database to a separate server would just add a lot of latency, not necessary for 99.999% of businesses. With CPUs getting faster, future developers will rarely ever need more than one server. In my opinion, the bigger problem is too many buzzword collectors trying to boost their career/self-importance with technology solutions nobody actually needs.
You're not familiar with the HN/reddit/whatever hug of death? I still see it happen several times a week, and that's after *drastically* cutting down how much HN I read :-D. I hate buzzwords and tech for the sake of tech just as much as the next person, but a lot of this stuff solves real problems for businesses that have a significant online presence.
@@tutoriaLinux if somebody's blog freezes up like that, most likely they're on a budget 'shared hosting' account (even WP Engine is shared now) and that's going to be thousands of blogs sharing a single MySQL database... And they get too aggressive with caching which leads to a lot of other problems, like for example Pantheon.io strips out all your useful _SERVER values from the backend, which is going to break a lot of software. But a good size VPS should be fine with Reddit or HN without any special caching. In fact, a sudden cache reset (poorly designed cache software) could be the cause of a database crash, because the server thinks wow this is super easy everything is cached and then boom, the cache expires and all your pages have to be regenerated at once and your server dies. It's ironic that people think more cache is the solution, when it's often the root problem, and that can be very difficult to detect because it's intermittent.
I've witnessed HN induced application death just a few weeks ago. This guy knows exactly what he's talking about. As for buzzword collectors being a problem, yes and no. Technology is developing faster than ever, it is very difficult for anybody to keep up with this rate of progress. Yes there are people like what you described, but for most of us, those people are no problem at all. The real problem is that we expect the technology we learned to serve us our entire career. "Hey, I learned some Wordpress and MySQL - yup, seems like I'm set for life!! Caching layer? Container? Load balancer? Whaat? GTFO with those buzzwords maan."
I loved what you said about the knowledge level / experience level conundrum. Reassuring to hear for sure. It's important to recognize that not every engineer, whether SRE or software or other, has the opportunity at the beginning of their career to work on large-scale systems or even interesting & challenging work, for that matter. Plenty of people get stuck at dead-end jobs writing CRUD apps or never have mentors or exposure to anything neat until they seek better employment. Fortunately, I got some decent experience early on but I also worked at a place where I was told they needed sandwich makers, not the people who design sandwich makers, and that I should just copy/paste existing stuff to bang out tasks without any concern for code improvement or good SWE practices. That's a real crappy way to manage talent if you ask me.
03:30 split database out load balancer 04:20 state 05:11 cache, CDN 06:20 cap theorem
On point with work experience vs knowledge so far I got 5 AWS certs but I would still rate myself as beginner.
This is EXACTLY what I'm looking for as I have an interview for a DevOps and Cloud position coming up shortly. Thanks TutoriaLinux !
few questions to gauge logging experience
1. how do you standardize the logs in distributed environment - Answers expected would be related to mutation , filtering (gork) in order to standardize
2. what tools are available out there for a infrastructure team / person to monitor logs and logging mechanism itself
3. how to you manage log archival
4. how do you control log ingestion at server level and log creation at microservice level -> which again relates to standardization, expect someone talking about log rolling , file limit and at server level queing the logs , forwarding using logstash/filebeat etc.
how to actually make logging useful (extracting alerts in kibana OR connecting cloudtrial )
Love these, thank you for adding them!
@@tutoriaLinux really brilliant series btw, wanted to ask you are these interview questions for devop engineers or junior devops ?
Nice vid. One example of logging is to push data via AWS Lambda function which can be used to stream events over HTTPS to Splunk HTTP Event Collector . Quite powerful
And very expensive. Might as well set up a logging sidecar container at that point if you're already using cheaper compute power.
You just spot on! Thank you for your videos!!!
I'm using your videos as part of the reference for the interview preparation because initially, a stack of the tech not going to change much and I will be asked the same questions over and over again until I pass the interview)
The thing is that without at leas a baaaaasic hands in to cloud experience all of the theory is worthless...
Hey man, i just want to let you know that your video is helpful. I got a job offering after reviewing the material on this video. More importantly, your word of encouragement at the end give me a huge boost of confidence during interview. Thankyou!! i hope you have a nice day
Thank you for your job. This is very helpful and I constantly ask myself similar questions throughout my work.
Had my first DevOps interview this morning! No technical questions at all... 🤦🏻♂️
Prepped all night for it too.
Thanks for your content dude, I wouldn’t have got the interview without following your stuff and your course over on Udemy. 🙏🏻
Hi Smith, I am preparing for my first interview now, can you tell me what type of questions were asked? and what I should focus on ?
Poojitha Mittapally I was just asked about the tech I’d used previously and how I’d used them (I don’t have a tech background so it’s all self-taught). We touched on whether I knew azure pipelines too.
Based on what I’ve read my interview wasn’t typical though. This is also a company I did a Python assessment for just a couple of weeks ago.
Mark Smith Got it.
Thanks for sharing. And good luck 👍🏻
Lucky boy 💙
Would definitely appreciate a more in depth answer set to some of these questions for the noobs ;)
Sure, what kind of depth are you thinking?
@@tutoriaLinux I am going to piggyback off of this. I would love to hear more in-depth conversation about the logging. Specifically, what would YOU do if you were asked about logging?
Dude you are doing great !!!
C - Consistency
A - Availability
P - Partition Tolerance
pick 2 AP or CP
Well I'm less nervous about my interview on Tuesday. I'm brand new to AWS, and my second interview consists of 2 Senior infrastructure engineers an a sys admin for an hour , then get to meet the CEO and the director of infrastructure. Pretty nervous, never really had an interview like this.
How was it?
Dude that's heavy. How was that process like?
How did it go, if you don’t mind me asking.
Man, this is gold. Thanks!
Maybe one logging-related question: What are the pros and cons of "sidecar" solution and "daemonset" solution when it comes to collecting and sending microservice logs (with fluent bit for example)
This is a great video. Thanks for posting this!
Good Work. These are questions askend in almost every DevOps Interview. You could have added configuration management tool and IaS solutions also
For a logging / observability I'd lean towards asking about Golden Signals, white box vs black box, what do averages tell you and what don't they, etc.
Could Also ask a scenario where the companies having a hard time correlating logs to the question everybody asks what is this user doing. Assuming this is a fresh logging stack, what kind of improvements would you suggest? I think the elephant in the room here is that you always need some kind of cardinality like a UUID or something like that.
This is amazing.. thanks alot!
Love these videos so much and I'm not planning to have or conduct an interview any time soon .
I just have one question: Why do you have, what I can only assume, is a meth lab in the background? :D
Haha, I love scientific lab-glass-looking things. Somewhere in my subconscious I guess I fancy myself a bit of a mad scientist. Those are some flasks that I picked up for $1 each at some point in my life :-D
a meth lab in the background. :D
I have 8 burnt kettles as background.
About logging I've been working with data dog over hybrid cloud and though it has it glitches here and there it work very well with k8 and can do Apm nicely
Thanks for the comment! Yeah, datadog has its warts but it's usable if you put in the time and effort. Thankfully I have a few people on my team who are more interested in making datadog behave than I am -- I've been benefitting from their work :-D. Also I just implemented some tracing from a Go application, making it talk to Datadog APM, and it was a reasonably good experience! It's no stackdriver, but it's good.
Very good video!
I always get nervous and forget how to answer in project mode, The reason I had a job with one company for too long not knowing whats out there, Now I am jobless for past 2 months and market is slow in toronto
Awesome
Great content. unlike software interview. What is the Technical part of the interview for Cloud computing.
Do you have any links you could provide for professional Devops Engineering courses/info. Specifically how to set up & manage a codebase, from a messy set of gits & exe stores to an organised self compiling distro for differing architectures ? Thank you.
will be best to explain these question with detailed answers
combination of ELK and sumo for the last place I was at.
Ok... I see I need to study a lot harder :D
It seems difficult to even get into a cloud role!? How do you get in with an employer if you have background in sys admin, infrastructure, etc?
There’s huge demand in the U.S., companies are having serious problems filling cloud engineering roles with competent people. If you use the “devops” keyword you’ll find a bunch of them. It pays better than “linux engineer” positions, too: Juniors are 80-100k/yr, mids around 130-150, seniors 160-200 base.
@@tutoriaLinux thanks for the response. To expound on my initial reply, I am a novice in the cloud world with a little hands on experience and a few AWS certs. The good old saying is it the chicken before the egg or vice versa. Would a company hire someone with limited experience? Why would they? How does someone get experience that counts and land a junior gig!? Not just like hey I do it in my free time and that’s my current experience. Thanks
Not sure what the context is - How many places have you interviewed at? How many have you applied to? Where are the interviews going wrong? If they’re rejecting you, try to find out why. If you’re overthinking this before applying for at least 30-40 devops/infra gigs, then don’t worry too much and just start applying. Companies hire newer engineers all the time, otherwise no one would have ever gotten started. If you’ve been job hunting for a while and you’re not getting interviews, find a tech recruiter in your area and work with them to land your first gig. Meetups are an option too.
Hey @TutoriaLinux,
I just got my aws saa cerification. I have no prior experience in IT whatsoever. What other advice would you give me in terms of resume preparation, interview question since I'm so new to this?
Just start interviewing! Think of it less as an attempt to reach a specific goal (one job), and more of a learning experience like setting up a server or troubleshooting a tough problem. Start applying to jobs, see how the first few interviews go, ask for feedback about your strengths and weaknesses, and adjust your learning/practice accordingly. Don't take anything personally (easier said than done) and just focus on the process.
Hey @augustine, good to know, did u find a job with your certification ?
Is this for a senior devops engineer position? I don't see how anybody who has not been in the cloud space before could find a way to answer these questions. How are people going to migrate over from their current on prem roles to the cloud space where people are really needed when they have to answer such in depth questions like this?
Its so expensive ( the training ) that your "current" employer will not pay for it because they anticipate you leaving for greener pastures :)
Bro i have studied basics, i want a devops job
Dave, awesome as always.
In your last video, you said you'd sit the person at a terminal and ask them to do things so you can see if they know what they're doing.
Do you try to get them to do some cloud infrastructure in person, or is that too much of a burden for an interview?
Hi, I am working to get into the IT field after being in it back in the 80s. Now, I would think that is how current interviews are done. "Hey show us how to set up a simple VPC from this laptop". Go! That is what I am striving for in my studies. To be able to present that skill set when asked, and to be able to talk through it. Both from the command line and the console. I know vpc would be difficult from the CLI, but not an S3 bucket and such.
I think surfbug's approach is a good one. Learn and study in a way that gives you the practical skills you need. And I agree with both of you; I think practical "set up this infra" problems are a cool idea. THAT BEING SAID: in 10 years of interviewing for jobs, I've never actually had anyone ask me to do that. I think these things can get pretty time consuming -- hunting down the last aws console checkbox or terraform resource argument that's missing, the last syntax error, etc. Interviews are extremely time-limited and I suspect that's one of the reasons I haven't seen this style of interviewing. FWIW software dev interviews tend to be more like that -- less theory, more practical application. But the complaint *there* is that the toy problems that are suitable for interviews are often very far removed from "real" programming work, etc. etc. etc. lots of long arguments all over the web about this. Short answer: it sounds like a good idea. Maybe it is! I'm not sure.
@@tutoriaLinux thanks for the feedback! Testing for my assoc architect on oct 22.... Let's see how it goes! Next step are those interviews, and thanks for the info! Subbed and shared, you have a new fan....
@@tutoriaLinux My manager always care about the personality and character rather too much digging on technical side. He said a person can learn our system but we can't change someone attitude
Use datadog it is fire
If I wanted to know the right answers to these questions what should I do?
Please let me know if you get the answers
Anyone else here because the song "Gotta Be a Reason" No? Just me? Okay.
splitting the database ?? my gawd my head is spinning already...and I have been in IT for decades but not exposed to that... I hate whiteboarding thank god for remote jobs interviews ..lol
One answer, Amazon Serverless Architecture!
Hahahaha +1, love it.
If anyone mentioned PHP during an interview I’d walk the fuck out.
I can tell you really know your shit.
You won't need any special infrastructure for front page of HN, not really a useful example. Splitting off the database to a separate server would just add a lot of latency, not necessary for 99.999% of businesses. With CPUs getting faster, future developers will rarely ever need more than one server. In my opinion, the bigger problem is too many buzzword collectors trying to boost their career/self-importance with technology solutions nobody actually needs.
You're not familiar with the HN/reddit/whatever hug of death? I still see it happen several times a week, and that's after *drastically* cutting down how much HN I read :-D. I hate buzzwords and tech for the sake of tech just as much as the next person, but a lot of this stuff solves real problems for businesses that have a significant online presence.
@@tutoriaLinux if somebody's blog freezes up like that, most likely they're on a budget 'shared hosting' account (even WP Engine is shared now) and that's going to be thousands of blogs sharing a single MySQL database... And they get too aggressive with caching which leads to a lot of other problems, like for example Pantheon.io strips out all your useful _SERVER values from the backend, which is going to break a lot of software. But a good size VPS should be fine with Reddit or HN without any special caching. In fact, a sudden cache reset (poorly designed cache software) could be the cause of a database crash, because the server thinks wow this is super easy everything is cached and then boom, the cache expires and all your pages have to be regenerated at once and your server dies. It's ironic that people think more cache is the solution, when it's often the root problem, and that can be very difficult to detect because it's intermittent.
I've witnessed HN induced application death just a few weeks ago. This guy knows exactly what he's talking about. As for buzzword collectors being a problem, yes and no. Technology is developing faster than ever, it is very difficult for anybody to keep up with this rate of progress. Yes there are people like what you described, but for most of us, those people are no problem at all. The real problem is that we expect the technology we learned to serve us our entire career.
"Hey, I learned some Wordpress and MySQL - yup, seems like I'm set for life!! Caching layer? Container? Load balancer? Whaat? GTFO with those buzzwords maan."
@@andreigaspar8669 Sorry HN crashed your website, keep learning and someday you'll get it.
Very good video!