Learning So Much Info & It's Hard To Juggle It All

Hey all,

Feel free to remove this post for redundancy reasons - it's not much different than what's been asked many times over on this sub.

I've been following the DevOps roadmap (I know some of you find this controversial - I understand), and just to clarify, I graduated last year with a Bachelor's in Computer Science, so I already have a decent development background (although I could definitely improve, which is something I am working on). I find all of this a bit overwhelming, but I have a strong desire to learn, and I spend time (I try to get to 2 hours of concentrated learning after work, I'm 22 years old and no other obligations mostly) after work learning every single day.

I'm currently in more of a Security-based role at a very large company, which does have me integrated with my company's cloud architecture, so I am learning AWS naturally from that, and would like to get the CCP cert in time. I like the work, and I think some of it will definitely translate to my next job, but DevOps has been an attraction ever since I graduated, and I like the idea of touching so many different tools and technologies, and there's a ton to learn (which again, I enjoy).

I guess the point I'm getting at is, how do you (or did you, depending on your current experience level) juggle all of this new information? For example, I was never exposed to Linux in detail in school. Currently it's something that I'm diving into deeper, but it takes time. Another example is Networking - Networking is huge. I want to learn a ton about Networking, it's interesting! But I find myself frustrated sometimes with the knowledge gaps I have on different topics.

I don't want to just be an average engineer, I want to be good. I'm willing to put in the time. I guess I'd just like some advice on how to juggle all of this information. I know I'm young and have a lot to learn, but I don't want that to be an excuse to take things slow (within reason).

Thank you all!

https://redd.it/118bwzu
@r_devops

Am I being ripped off?

Some context: I job switched internally at my company to be an Associate DevOps Engineer. (I work at a small startup >500 employees). They had an opening for a full DevOps engineer and I told HR I was interested in joining, but didn’t have nearly enough experience. HR ended up telling the team and the hiring manager for the role contacted me and basically said he’d love to bring me on but as an associate, and they’d have to hire a senior to help train me. I didn’t know much about DevOps, just knew a ton of networking, Python, and some bash/linux.

Long story short eventually I got brought on the team and I’m about 3 months in. The only downside is I’m making the same as what I made in my previous role (entry level digital forensics) - $55,000. I love the team and it’s so cool learning all the tools (K8s, Docker, Azure, Debian, Terraform, etc) but I’m definitely doing more work than I was before. There was no pay increase as it was seen as a lateral movement.

The upside I see is that I don’t have to spend 10 yrs as a sys admin to break into DevOps. And I get to learn pretty much everything or anything I want.

Am I being ripped off? Should I be making more? Or is this worth it for the experience?

https://redd.it/118d7ax
@r_devops

is practical DevSecOps worth it ?

Hey guys
I'm considering taking this: https://www.practical-devsecops.com/certified-cloud-native-security-expert/

As I want to establish myself in k8s and cloud native security but the price tag is insane (1k USD) considering that I'm in a country with a relatively weak currency compared to the dollar. I don't care about the certs, I'm after the skills I'll gain in the process

If you think it isn't worth it and there are other resources to learn what this cert offers let me know

https://redd.it/118h416
@r_devops

Learning Azure vs AWS

I can't for the life of me understand Azure as much as I do with AWS. I feel as though Azure is more granular than AWS which makes it a tad bit more difficult, at least for me, to learn. Is this normal or just me?

https://redd.it/118hsdv
@r_devops

Has your org started to evaluate the pros/cons of purchasing ChatGPT Premium?

Just curious what all is going on out there related to adoption of AI technologies like this for professional purposes. Are you starting to evaluate it in any serious way? Is the mere suggestion enough to get laughed out of the stand up?

https://redd.it/118e0yz
@r_devops

Updating stuff inside a pod

Our app's SSL certificates will expire in a few weeks. The app is deployed in a K8's cluster and to updates the certificates we're supposed to exex into the pod and upload the certs into a folder, then run s script to "activate' them. I'm still a Kubernetes noob, so I'm confused: would these changes persist when the pod is deleted and a new one created?

https://redd.it/1188khv
@r_devops

How do i setup ci cd pipeline from gitlab to ECS 2 for next js?

Hi I’ve been looking for tutorials on this and I can’t seem to find one that incorporate gitlab even if I find one they are either not working or outdated. Can someone tell me how to set one up for next js ? Or if I can may I message any of you that are capable I mostly work on front end and backend but never on devops or ci cd now I am confused. Thanks

https://redd.it/118ls12
@r_devops

what do you use shell scripts in your day to day work as a devops practitioner?

I've learnt the syntax and understood shell scripting, I'm finding it hard to understand where exactly do one use in their every day practices as a devops engineer.

https://redd.it/118n051
@r_devops

TF401025 'repoName' is not a valid name for a Git repository - solved

I'm just dumping this here, hoping to help someone googling this issue in the future.

When trying to create a repository named 'Server' you get the error TF401025
'repoName' is not a valid name for a Git repository or
'newRepositoryName' is not a valid name for a Git repository

You just can't name a repository 'Server' in DevOps, as explained here

https://redd.it/117zykd
@r_devops

Should we use OpenTelemetry traces for running tests?

Writing tests that check that side effects across the system are actually happening was always difficult (for me at least).

Then came OpenTelemetry, which basically gives you amazing visibility on what's happening across your system. So if you want to check that a user gets an e-mail everytime they register - theoretically you just need to search for the right spans.

We thought it might be an interesting concept to explore, so we started writing an open source testing framework that does exactly that - with Jest. We've actually managed to get into YCombinator with that so we can work on it full time.

Does it make sense? Would you be using something like that?

(Adding here our repo - https://github.com/traceloop/jest-opentelemetry, although this honestly doesn't mean to be a self-promoting post. I'm actually looking for honest feedback).

https://redd.it/118q0xn
@r_devops

Is anyone great at drawing (have a strong artistic side)?

So I think networking/devops is all very interesting, but I have this strong skill in drawing very realistic art. I'm not trying to brag, but I can really draw some great stuff - professional portrait artist level.

Sometimes I feel like this skill is going to waste and people say "oh you should just do frontend/UI developer" type of work.

Are there any devops or network/systems engineers that have a strong drawing/artistic side? Maybe there is some link here in terms of visual talent and working with computer systems?

Btw, this post is not a joke. I seriously am at odds sometimes in terms of choosing/settling on a career path due to having a strong artistic/drawing side. If someone could chime in or offer advice it might really help settle a lot of internal conflict that I have.

https://redd.it/118qixb
@r_devops

A Chainlink alternative? How API's work in Web3

https://www.thestatuscode.co/p/a-chainlink-alternative-how-apis

https://redd.it/117vpw0
@r_devops

What do you do if there is a strange behavior or falling tests in staging env and you have 100 recent commits?

i.e., git bisect?

https://redd.it/1174ycb
@r_devops

Getting "unknown manifest name" with docker registry behind CloudFront distribution

Howdy Devops community,


I am having an issue trying to pull images from my private docker registry running on an EC2 instance through my CloudFront distribution, unable to fetch it via digest and having a fallback on image tag, which will be soon deprecated. Below the logs on the different services processing the request.


For info:

cdn.example.com is my cloudfront domain

awsdocker.example.com is my docker registry address


On the client-side, getting:
```
docker pull cdn.example.com/python:3.10.8-slim-bullseye
WARNING: ⚠️ Failed to pull manifest by the resolved digest. This registry does not
appear to conform to the distribution registry specification; falling back to
pull by tag. This fallback is DEPRECATED, and will be removed in a future
release. Please contact admins of https://cdn.example.com. ⚠️
3.10.8-slim-bullseye: Pulling from python
025c56f98b67: Already exists
778656c04542: Already exists
85485c9f43dd: Already exists
23b3c91f0de2: Already exists
fd19b936aab8: Already exists
Digest: sha256:6a34910f419a1fdaf2b6d886a1b303eaaa093a828528e0f59b95ccc71feddfa4
Status: Image is up to date for cdn.example.com/python:3.10.8-slim-bullseye
cdn.example.com/python:3.10.8-slim-bullseye
```

Fetching the image directly from the registry works as expected. But...


Trying to set up a Docker Registry behind an AWS CloudFront Distribution, I'm getting the following error message on the registry:
```
docker-registry-registry-1 | OBFUSCATED_IP - - [22/Feb/2023:08:07:00 +0000\] "HEAD /v2/python/manifests/3.10.8-slim-bullseye HTTP/1.0" 200 13227 "" "Amazon CloudFront"
docker-registry-registry-1 | time="2023-02-22T08:07:00.575898015Z" level=error msg="response completed with error" err.code="manifest unknown" err.detail="unknown manifest name=python revision=sha256:6a34910f419a1fdaf2b6d886a1b303eaaa093a828528e0f59b95ccc71feddfa4" err.message="manifest unknown" go.version=go1.16.15 http.request.host=awsdocker.example.com http.request.id=4fec8c41-9afe-40b7-9af2-6023a9518efb http.request.iss.onethod=GET http.request.remoteaddr=OBFUSCATED_IP http.request.uri="/v2/python/manifests/sha256:6a34910f419a1fdaf2b6d886a1b303eaaa093a828528e0f59b95ccc71feddfa4" http.request.useragent="Amazon CloudFront" http.response.contenttype="application/json; charset=utf-8" http.response.duration=1.775224ms http.response.status=404 http.response.written=182 vars.name=python vars.reference="sha256:6a34910f419a1fdaf2b6d886a1b303eaaa093a828528e0f59b95ccc71feddfa4"
docker-registry-registry-1 | OBFUSCATED_IP - - [22/Feb/2023:08:07:00 +0000\] "GET /v2/python/manifests/sha256:6a34910f419a1fdaf2b6d886a1b303eaaa093a828528e0f59b95ccc71feddfa4 HTTP/1.0" 404 182 "" "Amazon CloudFront"
docker-registry-registry-1 | time="2023-02-22T08:07:00.714835837Z" level=info msg="rewriting manifest sha256:49749648f4426b31b20fca55ad854caa55ff59dc604f2f76b57d814e0a47c181 in schema1 format to support old client" go.version=go1.16.15 http.request.host=awsdocker.example.com http.request.id=1e59d24b-24c5-4f1f-a968-e80fd2982f18 http.request.iss.onethod=GET http.request.remoteaddr=OBFUSCATED_IP http.request.uri="/v2/python/manifests/3.10.8-slim-bullseye" http.request.useragent="Amazon CloudFront" vars.name=python vars.reference=3.10.8-slim-bullseye
docker-registry-registry-1 | time="2023-02-22T08:07:00.718986265Z" level=info msg="response completed" go.version=go1.16.15 http.request.host=awsdocker.example.com http.request.id=1e59d24b-24c5-4f1f-a968-e80fd2982f18 http.request.iss.onethod=GET http.request.remoteaddr=OBFUSCATED_IP http.request.uri="/v2/python/manifests/3.10.8-slim-bullseye" http.request.useragent="Amazon CloudFront" http.response.contenttype="application/vnd.docker.distribution.manifest.v1+prettyjws" http.response.duration=6.514434ms http.response.status=200 http.response.written=13227

docker-registry-registry-1 | OBFUSCATED_IP - - [22/Feb/2023:08:07:00 +0000\] "GET /v2/python/manifests/3.10.8-slim-bullseye HTTP/1.0" 200 13227 "" "Amazon CloudFront"
```


On the nginx reverse proxy serving my docker registry on the same EC2 instance:
```
OBFUSCATED_IP - - [22/Feb/2023:08:06:59 +0000\] "GET /v2/ HTTP/1.1" 200 2 "-" "Amazon CloudFront"
OBFUSCATED_IP - - [22/Feb/2023:08:07:00 +0000\] "HEAD /v2/python/manifests/3.10.8-slim-bullseye HTTP/1.1" 200 0 "-" "Amazon CloudFront"
OBFUSCATED_IP - - [22/Feb/2023:08:07:00 +0000\] "GET /v2/python/manifests/sha256:6a34910f419a1fdaf2b6d886a1b303eaaa093a828528e0f59b95ccc71feddfa4 HTTP/1.1" 404 182 "-" "Amazon CloudFront"
OBFUSCATED_IP - - [22/Feb/2023:08:07:00 +0000\] "GET /v2/python/manifests/3.10.8-slim-bullseye HTTP/1.1" 200 13227 "-" "Amazon CloudFront"
```
I wonder why this fallback happens when going through cloudfront.
Guessing some header mix up with CloudFront so that my registry cannot find the request digest?


Big thanks in advance for anyone having a clue!

https://redd.it/118t6aq
@r_devops

DevOps: Learning & Development - looking for feedback

My company (start-up, not too big, not too small with 70 tech employees) provides every employee with $2000 in Learning & Development budget.

I have two years of experience as a DevOps Engineer and aim for a senior role in the mid-term.

What L&D resources would you book or recommend?

Fyi: we can spend the budget on every L&D worldwide (books, conferences, E-Learning, coachings, trainings etc.)

https://redd.it/1173qdj
@r_devops

Best tool to deploy a lot of microservices in kuberentes

Hi,

I have a kuberentes running within the cloud (azure) and local (minikube) and I have at the moment around 10 services which I wrote in C# as docker images in a azure docker container.

At the moment I am deploying my services manually with kubectl (local) or kubernetes-task in Azure Devops (Azure kuberentes). It feels very mehhhh because I know the amount of services is fast growing.

Is there a tool to deploy all of that in a "good manner" like with a nice UI or something half automaticly?

https://redd.it/11764g9
@r_devops

Suggest some Devops project

Hello I’m working as intern and I want some Devops advance project with all tools and learn kubernetes in deep because my trainer does not have to much knowledge on kubernetes and my compony does not have any Devops project yet so
Where can I learn improve my self as Devops
Suggest some
Intership currently 3month

https://redd.it/116ckyy
@r_devops

Best options for SonarQube

hello, colleagues. Tell me how you can optimize sonarqube, which is hosted on azure web app (p1v2). Unfortunately, I can't increase the characteristics of the web app because of the money. But I would like to maximize the performance of sonar.

https://redd.it/11671hx
@r_devops

Being underpaid, what should I do?

Hey everyone,

I recently found out that a coworker who has the same level of experience and job responsibilities as me is getting paid 50% more than me. I was shocked and honestly feel pretty upset about it. Especially since my manager mentions I'm doing extremely well.

I always knew I was being paid a bit less than everyone else, but now I realize that I am extremely underpaid. I don't want to cause any trouble, but I also don't want to continue being undervalued and underpaid for the work that I do.

So, I'm wondering if anyone has been in a similar situation before, and what did you do about it? Should I bring it up with my boss or HR? Or should I just accept it and move on?

Any advice or insights would be greatly appreciated.

TLDR; I found out my coworker gets paid 50% more than me. What should I do?

https://redd.it/114lzpb
@r_devops

IaC git repo advice

I've started to use gitlab CI for internal service provisioning for my company's private cloud. Each service consists of templated configuration that is applied to the infrastructure using a mixture of python scripts, ansible, and terraform to configure the end-to-end network, compute, storage, and virtualisation stacks.

For the first service I'm testing with, there is a single repo with the python, ansible and terraform templates and a variable file that the templates reference. The CI pipeline is kicked off manually to begin service provisioning and configuring the infrastructure. The current workflow is for the engineer to create a new branch, update the variables file, commit, and then to run the CI pipeline from the new branch. After the service is provisioned the branch is left for future reference of how the service was provisioned.

I realise this is not best practice as branches should really be used for testing and developing the actual templates and code, not for service instances. What are some suggestions on how I could improve this git CI workflow?

Sticking with one repo - could you create a variable file for each service instance to avoid the per-instance branches? So the engineer would create a new branch, create the new variable file, raise a PR, and once approved/merged to master the CI pipeline would have a method of running the python scripts, ansible, and terraform templates using only the new variable file (as existing services shouldn't be touched)?

https://redd.it/1191vm5
@r_devops