Cloud DevOps mentorship/tutoring needed

Background
I am a msc it security student in Germany and btech computer science graduate from india, with multiple internship experience with full stack web dev. I have completed some course on docker and AWS cloud practitioner.

Expectations
I will complete my first year of msc in 3 more months after which I need to land job with a company to do my master thesis along with the company. I want to do it specifically in the intersection of cloud DevOps and security.

Requirement
I am looking for experienced cloud DevOps engineer (at least 1 years), who can get me interview ready to land a job for such roles. I only have 3 months to land a job so the duration of the contract will also be 3 months. I specifically want to learn in depth about Kubernetes, observability and infrastructure as code (terraform).

Bonus
If someone also can teach me potential security aspects of cloud DevOps and a potential master thesis in this field that would very beneficial for me.

Pay: up to 12 euro per hour

https://redd.it/1l93ej2
@r_devops

What do you use to automate self-healing scripts?

Hey everyone! just asking this to see if I'm missing something or the hereditary blindness already got me.
The thing is, I've been a DevOps engineer for about 5–6 years in two different companies, and in both of them, my main task was creating auto-remediation/self-healing scripts that run automatically when a monitoring tool detects something, like a spike in CPU, swap usage, low disk space, and so.

For that whole pipeline, I've been using a mix of Python/Go/Shell (sensible scripts), orchestrated by Rundeck/Jenkins/n8n/Tower as the executors, and Grafana/Datadog or similar tools for monitoring.

So my question is: is there anything dedicated to this? I mean, a tool that, when a monitoring metric hits a threshold, can automatically trigger something on a machine or group of machines?

https://redd.it/1l956jb
@r_devops

Developer cheat sheet

I created this free cheat sheet for cli commands.

I tend to prefer to invoke commands in my IDE vs GUI.

This is free.

If there is anything you want me to add please let me know.


Https://devcheatsheet.io

https://redd.it/1l95236
@r_devops

Automate adding vCluster to Argo CD using External Secrets Operator - GitOps

A blog post about how to automate provisioning virtual clusters (vCluster) using External Secrets Operator. Basically, when vCluster is created, it will be added automatically to Argo CD using External Secrets PushSecret and ClusterSecretStore.

Automate adding vCluster to Argo CD using External Secrets Operator

Enjoy :-)

https://redd.it/1l973i6
@r_devops

Best way to structure a new Azure DevOps pipeline for Playwright tests?

Hi everyone, I could use some help structuring a test pipeline in Azure DevOps using Playwright. My team used to work with Cypress, but we’re currently migrating to Playwright. The thing is, we never had a dedicated pipeline for automated tests, only build and deploy pipelines for the dev team, which were recently moved to another Azure DevOps project.

Now we want to create a separate pipeline specifically for testing, and I’m unsure of the best approach: should I create a brand-new YAML file just for the Playwright tests? Or try to reuse the old pipeline structure (even though it’s from another project and wasn’t built for testing in the first place)?

I’m looking for advice on what would be the best practice here, especially in terms of long-term organization and maintainability. If anyone has been through a similar migration, I’d really appreciate your insights. Thanks!

*E2E tests

https://redd.it/1l984wd
@r_devops

Secure s3 dashboard/website

Hi everyone. I am loosing my mind over what seems to be a simple problem.

So basically, I created internal dashboard (website stored in private s3). I have internal route53 record to use with it if needed, and internal ALB.
What i can't figure out is how to restrict access to it to only users behind the VPN. I tried CloudFront but the problem is that VPN uses split tunnel and public IP doesn't change, so WAF, lambdas, etc do not work.

What are my options to control access to this dashboard to selected users (preferably ones behind VPN without extra layers to login)

https://redd.it/1l9cd6d
@r_devops

Ode to the sysAdmin

Did the world forget that Systems Administrators existed before heirachical power structures?

- Customer support
- Engineer
- Architect

The architect’s role is to understand the shape of the bridge the customer needs, and the engineer builds the bridge.

If an Architect is expected to play Engineer, asked to build the bridge, whilst others were sabotaging the structure, who’s at fault?

The Architect?
The Engineer?
The 400 other people between,
Or the customer, which isn’t one, but many.

Please, think about that for a second.

A Domain Admin can never be asked to unsee what’s been seen.

We make sure others hold the same responsibility with the same honor, hoping that somewhere along the chain takes up enough of the slack to keep it together.

Systems Engineering isn’t easy.
Complex-Systems Architecture isn’t hard.

Meet me in the middle; or help me build the bridge.

https://redd.it/1l9f98n
@r_devops

Need a config management solution for structured per-item folders

I’m building a Python service that monitors various IoT devices (e.g., industrial motors, cold storage units).
Each monitored device has its own folder with all of its configuration inside:

A `.config` file with runtime parameters
A schema.json file describing the expected sensor input
A `description.txt` file that explains what this device does and how it's monitored

Here is the simplified folder strucure:

`project/`

`├──` [`main.py`](https://main.py)

`├──` [`loader.py`](https://loader.py)

`├── devices/`

`│ ├── fridge_a/`

`│ │ ├── config.config`

`│ │ ├── schema.json`

`│ │ └── description.txt`

`│ ├── motor_5/`

`│ │ ├── config.config`

`│ │ ├── schema.json`

`│ │ └── description.txt`

`│ └── ...`


What I’m Looking For:


A web interface to create/edit/delete these device folders
Ability to store and manage `.config`, `schema.json`, and `description.txt`
A backend (self-hosted or cloud) my Python service can query to fetch this config at runtime



https://redd.it/1l9hr4a
@r_devops

How to trigger AWS CodeBuild only once after multiple S3 uploads (instead of per file)?

I'm trying to achieve the same functionality as discussed in this AWS Re:Post thread:
https://repost.aws/questions/QUgL-q5oT2TFOlY6tJJr4nSQ/multiple-uploads-to-s3-trigger-the-lambda-multiple-times

However, the article referenced in that thread either no longer works or doesn't provide enough detail to implement a working solution. Does anyone know of a good article, AWS blog, or official documentation that explains how to handle this scenario properly?

P.S. Here's my exact use case:

I'm working on a project where an AWS CodeBuild project scans files in an S3 bucket using ClamAV. If an infected file is detected, it's removed from the source bucket and moved to a quarantine bucket.

The problem I'm facing is this:
When multiple files (say, 10 files) are uploaded at once to the S3 bucket, I don’t want to trigger the scanning process (via CodeBuild) 10 separate times—just once when all the files are fully uploaded.

As far as I understand, S3 does not directly trigger CodeBuild. So the plan is:

S3 triggers a Lambda function (possibly via SQS),
Lambda then triggers the CodeBuild project after determining that all required files are uploaded.

But I’d love suggestions or working patterns that others have implemented successfully in production for similar "batch upload detection" problems.

https://redd.it/1l9j2gg
@r_devops

Projects for resume

Hi folks.
I have 2 yoe in IT and I want to proceed in devops. Now I have theory and a little hands on on devops tools like jenkins, ansible, docker, k8s. I have also taken some random codes from chatgpt and built their docker images using jenkins and applied k8s deployment in them.
So now I wanted to know if I can add these in my project or not?
Also if I want to contribute in open source then how to search regarding same?
Would also love to know if you can help me to know about some other project ideas.

https://redd.it/1l9ke1a
@r_devops

Anyone switch from Python to Golang for most of their day-to-day tasks?

I'm in a situation where there's a lot of teams that each use different Linux distributions and dealing with Python dependencies, venvs, etc... is becoming a royal PITA.

https://redd.it/1l9lqdm
@r_devops

How can I create a clear SBOM output for my applications?

I am new to this community and currently looking for a way to creating a SBOM on my Windows systems and then scanning for security vulnerabilities. My goal is to get a consolidated block per application in the terminal, so not one line per CVE, but all the information (similiar like a winget view) grouped together per application. This way, you can quickly see which application needs to be updated instead of having to search around. Additionally, this should also be displayed as a list in the terminal.

So far I have tried syft + grype

Maybe someone can help me here, thanks in advance :)

https://redd.it/1l9nrvq
@r_devops

Opsgenie shutting down, looking for replacement. Suggestions?

Opsgenie will be ending its service in 2027. We want to find a good replacement soon so we have enough time to choose carefully and not rush last minute. Does anyone have recommendations for other tools we should consider?

Here's what we mainly use Opsgenie for:

* Checking who is on call and directing calls from our VOIP system to the right person, using a webhook from our VOIP provider. We’d prefer a tool that has built-in on-call scheduling and works well with 3CX. If it doesn’t support 3CX, options like Twilio or other providers are okay.
* Sending alerts to people when they are on call.
* Notifying team members if a service goes down, based on alerts from tools like Pingdom or other monitoring services.
* Creating and managing work schedules.
* Temporarily changing schedules (for example, if someone is taking time off or is sick).

So far, I’ve checked out Incident.io, Pagertree.com, and Firehydrant (which is way too costly). Do you have any other suggestions we should look into? Right now, our team is small—just four people handling on-call duties and standby SLA —but we might grow in the future.

https://redd.it/1l9o0e6
@r_devops

Just spent 2 hours looking for feature specs that were 'somewhere'... again

Been working on the same web service for 3 years. Today I needed to update a feature and literally spent 2 hours searching for the latest API documentation. Went through Google Drive, Notion, GitHub, Slack threads, old emails...

Finally found it in a spreadsheet linked in a 6-month-old Slack message. The "official" documentation in Notion was created 3 years ago when the feature was first built and hasn't been updated since - none of the recent changes were documented.

Anyone else dealing with this documentation chaos? When teams use different tools and nobody knows who has what information. Documents get created and then abandoned, and no one can tell what's current anymore. How do you find the right information in situations like this:

Dev team uses GitHub and Notion
PMs use spreadsheets and Google Docs
Customer support uses spreadsheets and Google Docs
Design team uses Figma comments

https://redd.it/1l9mjdl
@r_devops

Is CPU utilisation the only thing it matters when it comes to performance?

I work with a lot of dev teams and we keep getting told to scale up when the CPU (or some other hardware metrics) utilisation is approaching 100%.

I can't help but keep thinking back then when I used to game a lot, having a better hardware meant higher performance in terms of FPS, and that older hardware could have utilisation not reaching 100% but still has low FPS.

I can't understand why they don't focus on the end result metrics rather than hardware metrics.

Or did I get all of this wrong? I don't deal with app teams directly, so I have no idea about their apps, I just deploy it and maintain the infra around it.

https://redd.it/1l9qedh
@r_devops

Stop the madness: DevOps trends that are ruining teams in 2025

Okay I need to vent. Been doing DevOps for 10 years and I'm losing my mind watching teams chase every shiny new trend.

Just consulted with a startup that has TWELVE microservices for a todo app. Twelve! They have more services than active users. Their deployment process is longer than my morning commute and fails about as often.

And don't get me started on the team that spent half a year setting up Kubernetes to run 3 PHP apps that get maybe 100 requests per day. The operational overhead costs more than just running the damn things on a single EC2 instance.

But the thing that broke me? Production database running out of space, one-line config fix needed, but had to wait 45 minutes for the GitOps workflow. Database died after 20 minutes.

Sometimes you just need to SSH into the server and change a value. I said it. Fight me.

Hot take: most of the "successful" teams I work with are actually pretty boring. They pick proven tech, keep architectures simple, and spend time building features instead of rebuilding their infrastructure every quarter.

Anyway, wrote a whole rant about this stuff: https://medium.com/@heinancabouly/devops-trends-that-need-to-die-in-2025-please-for-the-love-of-all-that-is-holy-22cbbadf2db3?source=friends\_link&sk=3f2bbe0844a62291eefd787da978ef53

Anyone else tired of this madness or is it just me getting old?

https://redd.it/1l9t7mb
@r_devops

Anyone here transitioned from QA to Devops? Do you feel rewarded? Is it a wise move?

I’m a QA based in the US and considering a change to Devops .. looking for connecting with people with similar background as me and willing to move to devops

https://redd.it/1l9y5ez
@r_devops

Transition to developer, potentially fullstack

After about 8 years in DevOps I have realized I always incline more towards development and architecture of the solutions which is a valuable skill to have as a DevOps. But I would rather have the roles swap and become developer with the experience and positive approach to DevOps practices.

The issue is my experience in development is mostly just doing minor code reviews and discussions with devs in context of operation and automation. I am familiar with .NET ecosystem and can easily understand code bases, yet I have not finished a single project in .NET myself. I have made few running websites in Vue or Svelte, doesn't really matter which framework I would use but that's an option for me too.

So the issue is I'm not sure how to improve and advertise myself? Had anyone made transition from DevOps to more Dev work?

https://redd.it/1l9ynml
@r_devops

Deciding between two offers

I’m currently deciding between two job offers and I’d like to hear some advice.

Company A: mostly writing CI/CD pipelines with on-prem deployments. They are trying to modernize their stack.

Company B: 30k USD less than company A’s offer. Cloud based, modern stack with applications deployed globally with proper monitoring. Growth and learning opportunities, especially where I’d like to be: Orchestration, Cloud, SRE… more senior team members who will help me learn and up skill.

Both seem like very healthy environments and cool people to work with.





https://redd.it/1la249k
@r_devops

Wrote this guide on explaining CI costs to CFOs

Work at a CI company, wrote this guide after customers kept asking. Figured others might find it useful.

Guide here

https://redd.it/1la3bip
@r_devops

Should I be worried that you seem to speak chinese for me ?

So I (23) am an engineering student in data science and I will graduate after 6 or 7 months. All I know is some cute data engineering ( cleaning , transforming , etc..) , predicting things with models , do some API services based on RAG , Work with some object detection models and build some Spring boot projects. But you guys seem on a different level that makes me anxious about my capabilities. Please tell me that most of you here are seniors or that I still have time ahead of me to understand what I might need for work .

https://redd.it/1la2kqw
@r_devops