86
Reddit DevOps. #devops Thanks @reddit2telegram and @r_channels
Gartner thoughts?
Just curious how do you feel the comments and analysis of gartner and other analysis firms take on platform engineering and ai- automation of Devops..
Have seen the leaders and managers take the gartner suggested tools seriously
https://redd.it/1m668tl
@r_devops
Built a tool to stop wasting hours debugging Kubernetes config issues
Spent way too many late nights debugging "mysterious" K8s issues that turned out to be:
* Typos in resource references
* Missing ConfigMaps/Secrets
* Broken service selectors
* Security misconfigurations
* Docker images that don't exist or have wrong architecture
Built Kogaro to catch these **before** they cause incidents. It's like a linter for your running cluster.
**Key insight**: Most validation tools focus on policy compliance. Kogaro focuses on operational reality - what actually breaks in production.
Features:
* 60+ validation types for common failure patterns
* Docker image validation (registry existence, architecture compatibility)
* CI/CD integration with scoped validation (file-only mode)
* Structured error codes (KOGARO-XXX-YYY) for automated handling
* Prometheus metrics for monitoring trends
* Production-ready (HA, leader election, etc.)
**NEW in v0.4.4**: Pre-deployment validation for CI/CD pipelines. Validate your config files before deployment with `--scope=file-only` \- shows only errors for YOUR resources, not the entire cluster.
Takes 5 minutes to deploy, immediately starts catching issues.
Latest release v0.4.4: [https://github.com/topiaruss/kogaro](https://github.com/topiaruss/kogaro)
Website: [https://kogaro.com](https://kogaro.com)
What's your most annoying "silent failure" pattern in K8s?
https://redd.it/1m5ro6l
@r_devops
Joining in as the first "DevOps guy" at a startup. Any ideas on how I could create good impact?
I've worked as a DevOps Engineer at a big company for 3 years. I'm joining a startup now so I'll be expected to hit the ground running. Where do you think I should start from to enforce DevOps principles?
https://redd.it/1m5wofb
@r_devops
Job Opening
Potential job opening for a seasoned devops engineer in the dmv area. Contract to hire. Must reside locally.
https://redd.it/1m5sfje
@r_devops
Just finished setting up automated deployment - lots of things learned. Was yours different?
For last few years I have been part of a team maintaining AWS infra, however we are at the early stages of learning and development. So far we have been running terraform appllies manually.
Now finally I have had time and desire to setup my first automatic pipeline, just out of the rabbit hole. It was not that easy, here is what I had to do...
My task was harder because I have set these requirements to myself: no AWS credentials, use instance profile + IAM, should work cross-accounts. so need cross-account assume role grants.
1. First thing I learned that our superadmin access to AWS is very different from non admin access. It has all the permissions under the sun. But for the CI/CD , I have setup a separate IAM role, and had to grant all the necessary IAM policies, execution roles, all fine grained. I could have just given admin permissions, bu I only needed stuff for docker repository and microservices.
2. WTF is PassRole? ChatGPT kept convincing me that I need it, even AWS docs said that I need it. I could not understand what it is. Finally, I did not need it in my case.
3. Additional IAM hell, like granting assume roles, configs split between various environments.
4. We use internal git repositories, and gitlab/github practice is to use ssh. Easier was to flip to using `git::https...` in terraform modules sources, with token authentication, but had to do git config changes to use ".insteadOf" for rewriting git URLs
5. if that was not enough, our security team slapped us with HTTP proxy instead of NAT gateways.
Maybe there was something else along the way, I cant remember in the spaghetti of the code and issues I had to fix. But it feels like it was supposed to be easier, or maybe I just did it wrong?
The only way I think it would have been easier, and maybe it should have been to some extent, if I was:
a) using AWS access id/key, I could just store them in git repository, and use per environment where I need to deploy. CI/CD needs to run in pre-prod? use pre-prod AWS keys to run directly in that account.
b) store IAM config in the same repository, run terraform manually, because it needs to be done once or rarely.
c) give wider permissions to the CI/CD pipeline, so that I do not discover what IAM policy is needed for each small thing.
Learned a lot, happy it is working, will do it again.
https://redd.it/1m5lzgz
@r_devops
Anyone from big tech companies??
I need some guidance on how to crack into the big tech companies as a cloud and devops Engineer... Can anyone help me.?
https://redd.it/1m5nb63
@r_devops
Helm charts
I’m a Senior Software Engineer and have recently earned my CKAD certification. Now, I’m looking to deepen my expertise in Helm, as I believe it’s one of the best tools for organizing and managing Kubernetes manifest files efficiently.
Would you recommend investing time in mastering Helm further? Is it truly valuable in real-world environments?
If so, I’d appreciate any guidance on where to start in order to build solid, hands-on experience. Any advice or learning path you can share would be greatly appreciated.
https://redd.it/1m5homt
@r_devops
Project Idea Is there value in an AI (RAG)-powered deployment platform that provisions AWS/Azure infra automatically?
Hello, I am currently in grad school majoring in cs, wanted to work with rag systems and deployment services like aws infra, ci/cd pipelines, would this project solve some of your issues, if I build one would you be willing to use it?
Elaborate idea: An application where you give your repo, or github link or github authorization, and using its rag system it reads context from the repo, and answers your questions like to write a dockerfile, tells you why your deployment failed from logs, even helps with infra, like "solve this problem and push the pr to github" and it does that.
Your feedback would really help me out, otherwise i'll look for some other project to work on.
Thanks
https://redd.it/1m5fzhz
@r_devops
Should I pivot to AI/MLOps or go deeper into platform engineering? (36M, 14 years in tech, feeling stuck)
Hey everyone, throwaway account for obvious reasons. I'm feeling pretty lost about my career direction and could really use some outside perspective.
**Background:**
* 36M, based in Madrid
* \~14 years in tech (started in network/security, transitioned to DevOps \~6 years ago)
* Currently Senior Cloud DevOps Engineer at a mid-size company
* Have experience with the usual stack: AWS/Azure/GCP, Kubernetes, Terraform, CI/CD pipelines, monitoring tools, etc.
* Currently finishing my Master's in AI (should be done by July)
**The problem:** I feel completely stagnated. I've been bouncing between companies every 1-3 years trying to find growth, but I keep ending up in similar roles doing similar work. The pay is decent but not amazing, and I honestly don't know what my next move should be.
Some days I think about:
* Going deeper into platform engineering/SRE
* Leveraging my AI Master's to pivot into MLOps/AI infrastructure
* Moving into management (though I have zero leadership experience)
* Maybe even switching to software development completely
* Looking into remote work for international companies (better pay?)
**What I'm struggling with:**
* I don't have a clear 5-year vision of where I want to be
* Not sure if I should specialize deeper or go broader
* Feel like I'm behind compared to peers who seem to have clearer paths
* Impostor syndrome is real - sometimes feel like I'm just copying configurations without truly innovating
* Market seems super competitive right now, especially in Europe
**Questions:**
1. For those who made it to senior+ levels in DevOps/Platform Engineering - what differentiated you?
2. Is it worth pursuing the AI/MLOps angle given my current background + upcoming Master's?
3. How do you know when it's time to pivot vs. when to stick it out and go deeper?
4. Any specific skills or certifications that actually matter for career progression?
5. Should I be looking internationally or focusing on local market?
I know this is pretty scattered, but I'm genuinely feeling lost and would appreciate any advice from people who've been through similar situations. Thanks in advance!
**TL;DR:** 14+ years in tech, currently DevOps, feeling stuck and unsure about next career moves. Need advice on specialization vs. pivoting, and general career direction.
https://redd.it/1m5ctoq
@r_devops
Serverless Scaling: Deploying Strands + MCP on AWS
Link:
https://redd.it/1m55eqq
@r_devops
Live tech courses and training Platform
Hi DevOps community members. Whether you are swtiching into DevOps from a different career or a different domain in technology, a recent fresh graduate, upskilling your skills for your current work placement or attempting to get certified, Tutrx is launching a live online tutor marketplace.
For Students:
- experienced indsutry led instructors who bring real-world expertise to teach you in order to meet your needs and goals such as landing a job, tranisitoning into a new career, giving support to your current work or someone looking to get certified.
- will have hands on labs with demo data, quizes and assignments all created and made to improve your skills with practical knowledge and use cases from experienced instructors.
- For people who want to learn something quick like a trouble shooting problem or a quick solution, the platform will have 2 minute or less videos that give quick steps in solving technical problems.
- At Tutrx there are courses for everyone and design to host live classes to match your time schedule and availability around the world from anywhere.
- Courses will be matched to your skill level and how much prrior experience you have, so you will know what's the best pathway to go to meet your end goals.
Tutrx is not only designed for Students but also for Instructors who want to start teaching as a fulltime or part time at your own dedicated time.
For Teachers:
- Creating a course is as simple as 5 steps which you can create any style or format of course that meets your training guide.
- You can host 1 on 1 private sessions or a group classroom session all can be done with our customized video conferencing tool which has unlimited minutes.
- With this dedicated video conferencing tool, students and teachers can take notes, have video to text transcription, summarization with AI, remote control and many more features.
- Our platform gives you the ability to create any learning material and content easily such as Short Video builder to create shorts in minutes, quiz, assignments and lab builders to create practical lab and learning material. You can manage everything in your own dashboard, see metric and progress of your students and revenue generated.
- Get instant payouts with our partnered payment gateway so all you have to worry about is just the enjoyment of teaching.
Go see our platform features and sign up to be registered using the link below.
https://tutrx.org
https://redd.it/1m52bay
@r_devops
Hi guys, need your suggestion and opinion on this project!
I was thinking to build an open source alternative for Control-M. I'm yet to plan this out but need to check whether it's any good of an idea.
I need to do some project for my resume as I'm quitting my job (don't like the work) and i would love if it was an actually useful one. I am not sure if this is the right sub to ask this question, but you guys seem really supportive.
Once again, even though it is a side hustle project I would be happy if it would be actually
Useful.
Please provide your valuable suggestions/inputs.
Thanks in advance,
https://redd.it/1m4uo0a
@r_devops
How Do Big Cloud Providers Like AWS/DigitalOcean Build Their Infrastructure? Want to Learn and Replicate on a Small Scale
Hi all,
I’m really interested in learning how major cloud providers like AWS, GCP, Azure, or DigitalOcean set up their infrastructure from the ground up—starting from physical servers to running a full self-service cloud platform.
My goal is to eventually build my own version on a smaller scale where users can sign up, create VMs or databases, and be billed hourly—similar to what cloud providers offer. But before jumping in, I want to study and understand:
• What kind of software stack do big cloud providers use on bare metal?
• How do they manage virtualization, networking, storage, and tenant isolation?
• Which open-source tools (e.g., OpenStack, Proxmox, Harvester, etc.) are worth exploring?
• How are billing, metering, and provisioning automated?
• Any good resources (books, blogs, courses) to learn all of this from the ground up?
If anyone here has built something like this or works in infrastructure/cloud engineering, I’d love to hear your advice or learning path suggestions.
Thanks in advance!
https://redd.it/1m4qlq9
@r_devops
imo DevOps Market is still Great
Hi Folks,
I recently did only one job interview tbh out of boredom (2 stages) and got the offer (EU). 143k EUR TC (on-site) - it's okay for EU since we have lower salaries here than US, but that's not the point.
They told me they had about 50 candidates, but I have solid fundamentals and have kept my stack reasonably fresh. I do infrastructure and coding for my side project (shameless shoutout to prepare.sh), so it was relatively easy.
I started as full-stack, then worked in finance for 5 years, and moved back to tech in 2019. Compared to finance, this market is still great. Even during the best days in the financial sector, I was looking for months for ANY job, getting maybe 1-2 calls out of 300 applications.
By no means do I consider myself a great coder or architect - I'm okay at best. This makes me think there's either a great mismatch in expectations (e.g., people get heavily misled thinking they can pass a few certs, know "helm install," write basic CI/CD) or there's some other mystery, because every time I read Reddit, I see doom and gloom posts from people.
https://redd.it/1m4lnr0
@r_devops
Europe: Girlfriend finished IT degree with DevOps focus - can't land an entry job. Any advice?
Hey all,
My girlfriend moved to Europe (Austria) with me and recently finished a Bachelor’s in IT here to get her foot in the door. She came from a music education background (which she didn't enjoy doing at all) but switched to IT after getting inspired by my work and me (regretfully) saying that IT would always be a strong market (boy, was I wrong). I'm a senior software developer, but not in DevOps specifically.
She leaned toward DevOps during her studies (CI/CD, cloud, automation, etc.). She's not into programming-heavy roles but really liked the infrastructure/ops side of things.
Now she’s struggling to find a job. Even junior roles ask for 2–3 years of experience, or companies just end up hiring seniors instead. She has no internships or formal work experience, and the market seems brutal right now for beginners. I am specifically refering to the EU market here, as I assume that most people here are from the US.
Any advice?
* Are there real entry points into DevOps right now?
* Would cloud certs (AWS, Docker, etc.) help?
* Do self-built projects matter, or do companies only care about professional experience?
* Should she aim for sysadmin or cloud support roles instead?
* Is there any sign of the situation improving?
Thanks in advance. We’d appreciate any input or real-world advice!
https://redd.it/1m4kiai
@r_devops
Looking for Advice (Please reply don't skip)
Hi
Everyone,
I have 3.5 years of experience in SEO, however I want to switch it into devops because of various reasons including personal, finance and professional reasons.
My education background is from commerce.
I chose tech because i already interact with websites, so I know little about technicalities.
And, I felt I may be good for more tech instead of marketing.
That's why I started preparing for the same since March month.
I completed:
Basic overview of theory concepts
Linux commands
Git and GitHub
Python (from Hello world to oops and then python scripting)
Bash scripting
CI and CD pipeline (GitHub actions)
And , Just started AWS.
And, all this I did through my friend course instead of purchasing my own.
But, from a job perspective i needed a certificate, that's why thinking of purchasing a devops course from PW skills (same purchased by my friend).
So, what are your thoughts on this
Am I going on the right path
Or, any mistakes or suggestions?
Note: i know devops is not for entry level and also I don't have a tech degree like btech. That's why It will be difficult for me to get a job. But, i will give my best because I have back up (my current job).
So, please give me just realistic and practice advice in a positive manner.
https://redd.it/1m6465g
@r_devops
SecretSpec: Declarative Secrets Management
We've recently released secretspec.dev, I wonder what's the opinion of the folks here on a tool that unifies the interface between secrets providers and applications? See the announcement post at https://devenv.sh/blog/2025/07/21/announcing-secretspec-declarative-secrets-management/
https://redd.it/1m5shwf
@r_devops
How do you handle security tool spam without ignoring real threats?
Our security people just dumped another 5000 "critical" findings on us. Half of them are like "S3 bucket allows public read access" for our fucking marketing site that's literally supposed to be public.
Meanwhile last month we had an actual data leak from a misconfigured RDS instance that somehow wasn't flagged as important.
I get that they need to cover their ass but jesus christ, when everything is critical nothing is critical. Anyone else dealing with this? How do you separate signal from noise without just ignoring security completely?
Starting to think we need something that actually looks at what's running vs just scanning every possible config issue.
https://redd.it/1m5uezj
@r_devops
Are the titles merging?
Hey folks,
Trying to get my head around the titles we are given vs what we do.
Although I’m a Cloud Engineer by title, I’m completely in control of the CICD, software release and deployments.
I’ve also been tasked with the secure code pipelines. This is outside of my day to day AWS operations, cost analysis etc etc.
When does Cloud Engineer become SRE / DevOps / Platform engineer and so on?
https://redd.it/1m5rndc
@r_devops
Struggle with the fundamentals?
I joined as a graduate at one of the FAANGs and immediately started working on projects. I have worked as a DevOps engineer for 4 years but I feel I still struggle with the fundamentals. For e.g. I did an interview recently and they asked me about how ssl certificates work, no biggie but I struggled with an answer since I had forgotten the theory. I really want to get to a stage on where I don’t have to struggle with the fundamentals and theory anymore. I have been advised to be able to crack interviews better, you need to be good at the fundamentals and I really want to get to that stage!
https://redd.it/1m5p1ga
@r_devops
Livy alternartives
Hi we are deploying apache spark and wondered what altervatives people are using to Livy.
https://redd.it/1m5l9ss
@r_devops
Need your help for my cloud learning journey and help me decide on a instructor ?
Hello Everyone,
Hope you are having a great day and enjoying the sunny days :)
I have recently started my journey into AWS Cloud and would love to know which course should I move forward with ?
I've have 4 popular instructors ->
Neal Davis (Digital Cloud Training)
Stephane Maarek (Udemy)
Adrian Cantrill
GPS (Learn to cloud)
Questions:
1. How do these instructors compare in terms of theoretical knowledge gained vs applied knowledge (any other factor that I may have missed) ?
2. Is it worth combining two of them ? If so, which one ?
3. Any underrated resources I should be considering ?
I don't want to run behind certifications I would like to develop a fundamental understanding in the cloud domain.
Your advice and experience would help me during my cloud learning journey !
https://redd.it/1m5gpo8
@r_devops
what should i know before deployment full stack system
i am talented at building spring boot java and angular/react systems with a database (relational/nonrealtional) but my problem is i dont have the skills or knowledge to deploy the systems for real use by users in addition i have dockerized systems before i know that helps
now i want to know how to deploy please help me what should i look for and know before deployment
https://redd.it/1m5dxxo
@r_devops
Node.js project deploying in Hostgator Shared Server?
I build a small node.js project, can i deploy it in hostgator shared server?
https://redd.it/1m5c9dl
@r_devops
Automated before?
Has anyone faced numerous blockers when automating something, such as agent issues, firewall, logging, bamboo build fails etc etc? I've been working on something for almost 2 months and have a deadline but no matter what I do nothing is working after a developer added environment tests to the code and everything started skipping tests or more issues were created.
I use Amazon Q for code understanding and chatgpt if needed, but prefer one on one with my seniors to guide me. I've just been extremely stressed and worried about this it feels I'm getting nowhere and nobody understands that and I have everything documented but they don't want to read it.
https://redd.it/1m53uta
@r_devops
What do people use for monitoring/o11y? Why did you pick that provider?
Title says it all. I've tried most of them but I feel like I'm missing something-- most of these providers are painful to implement.
Super curious what people use, why you use it, and how you make it suck less
thanks all
https://redd.it/1m4tw26
@r_devops
Idempotency in System Design: Full example
https://lukasniessen.medium.com/idempotency-in-system-design-full-example-80e9027e7bea
https://redd.it/1m4qh1x
@r_devops
Production support to Devops Switch
Hi All,
I have around 11 years of experience in production support, currently I am working in partial SRE role but I want to completely switch to a Devops role. Could you please guide me.
https://redd.it/1m4q70l
@r_devops
Dynamic Reverse Wireguard
Hello DevOps folks!
I want to share with you my exciting project which I had to develop because I live in Iran.
It all started after Israel and Iran war. Our internet was super slow for the first few days, and got worse everyday until we almost had 0 internet connection to outside.
I was trying my best to setup a working VPN but everything would be blocked withing a couple of hours.
But I saw something weird. For a Wiretuard setup, it was possible to have a working VPN, but only in a reverse setup, meaning server MUST have sent the handshake. The other way around (Handshakes from Iran to outside) was being blocked.
I've developed a simple python script which reverses the handshake process.
I've posted on this subreddit because this project was so exciting for me, I figured you guys would like it too.
It's kinda a dynamic reverse Wireguard VPN.
Github repo
https://redd.it/1m4l0q7
@r_devops
How many infra engineers you have for how many developers?
Hey all,
Wondering about scaling the infrastructure org in connection with how many product developers they serve.
When I say the infrastructure org, I mean SRE, Platform, devops, Tooling, Ops and every other team that takes care of stuff for the Product teams.
So how many people and team do you have in your company and how many product team and engineers are they servicing?
Of course I'm aware some companies are more infra intensive, happy to get more specific answers.
https://redd.it/1m4ilm5
@r_devops