Datafile Management for tablespaces in Oracle 19c

Hello everyone, I hope all is well.

I've been facing an issue for the past few days where there is a PL/SQL code that is causing temp tablespace issues where sometimes it gets stuck on the parallel execution and the temp tablespace just gets fully consumed. Note that the allocated storage for temp is around 500GB. I am using AWS RDS runing on oracle 19c as my database and I've noticed that RDS allocates 1 datafile of size 500GB for the entire temp tblspace.


So I have the following questions:

1-Does having multiple smaller datafile for a tablespace actually enhance performance? And would creating 10 datafiles of 50GB each actually up performance? And would that be at the level of execution time or would it actually help solving the issue mentioned above?

2-Does AWS RDS have an easy way to set this up as I know that RDS is serverless and we do not actually have access to the backend servers and I couldn't find any detailed documentation regarding this

https://redd.it/1i91p0t
@r_devops

yml.j2 formatters

Are there any formatters for yml.j2 and any other yml files?
Also, how do you format yml files on Linux machines from the cli?

https://redd.it/1i92tv6
@r_devops

Article about the number of pipelines

Hello,

I am building a new product, and I want to challenge my usual takes over CICD pipelines, and I am specifically trying to determine how many pipeline do I want.

So I have been searching for resources talking about this, but I have really hard time finding any. I can find 50 articles explaining what is CICD and IAC, another 50 will talk about the different possible tools. And they are always in a very silo philosophy: "Here the info if you want to do IAC pipeline", "here the info if you want to do app pipeline".

But I can not find anything talking about the more global pipelines.

I am looking for example for thing like the values of 1 pipeline for the IAC + 1 per product versus 1 pipeline for everything, or anything in general that consider. Or anything that can be related to the fact that IAC and App are not leaving in their own silo but are totally mingle and the pipeline will be at some level.

So I was wondering if you had good article talking about this?

https://redd.it/1i93oag
@r_devops

Trigger jenkins build when Pr descrption is edited or changed

Hi,
I need to trigger a jenkins build when the description my PR is updated. Is it possible?

https://redd.it/1i95jwn
@r_devops

Seeking advice on Interview content when switching roles to Systems Engineer / Devops roles.

I have been in professional services for almost longer time around 5 years, now seeking roles and giving interview for Devops / System engineering roles.

How should one even prepare and i am not talking about the tech stack, i know ansible / aws / openstack / linux and terraform / cfn for Infrastructure as code.

The questions that really throws me off in the interview, are something like below ,

1. How do you do version control ? ( thats the only effort interviewer made while explaining ) - My answer was tools like github, git ( turns out is wrong )

2. There is no order, data is store randomly at different places ( How will you bring order or how will you create process to handle large number of data )

Is there some kind of hidden syllabus or learning content that i am missing ? I clear tech with flying colors but shit like above fails me in interviews ? What am i missing ?



https://redd.it/1i9cqin
@r_devops

Google interview tips

Hi guys, give me tips on how to crack the Google DevOps Engineer role. How many roles will there be, and what kind of programming will they ask? Before that, how can I get my resume shortlisted? I’m from a Tier 3 college. will they accept it?

https://redd.it/1i9ei5m
@r_devops

What's the best way to check the package.json version in the origin/master branch and compare it to the current local branch?

What's the best way to check the package.json version in the origin/master branch and compare it to the current local branch? Are there ways to do this? I tried to find some libraries, but I was out of luck.

https://redd.it/1i9kxpm
@r_devops

Can you help me navigate: Workspaces vs Environment Variables?

Hi everyone,

Currently, our project structure looks like this:

/dev/service_name/main.tf  
/prod/service_name/main.tf  
/prod/main.tf  
/modules/shared-resources/resources.tf  

This setup makes it easy to see what exists in each environment. However, as we’re now ready to mature our workflow, I want to DRY up our resources.

The primary difference between dev, prod, and staging environments—aside from which services are deployed (which I’m working to organize)—is the environment variables passed in. We’ve done a good job of centralizing environment variables (with some default configs still inherited from a higher-level /data repository), making them effectively the control plane for the entire infrastructure.

I’m considering moving away from the dev/prod directories to a structure that allows for deploying all services with less code. For example, we could migrate to:

/service_name/main.tf  

One option is to use Terraform Workspaces to handle environment differences. Alternatively, we could pass an $ENV variable to determine the deployment target and handle this via our existing Justfiles (which are like cleaner Makefiles - we have bash scripts that does any heavy lifting that needs doing before we call Terraform, or anything Terraform is not particularly good at). In CI, we’d SSO into the appropriate dev/prod account via the Justfile and apply changes accordingly.

Is there a strong reason to favor one approach over the other, or is there an alternative I should consider?

This is my first time leading this kind of effort, so while I’m confident in these ideas, I’d appreciate any advice to navigate this transition effectively.

https://redd.it/1i9lgc0
@r_devops

How would you 80/20 DevOps at a tiny company?

Let's say you're at a very small company: only a few developers, who are doing system administration by SSHing directly into the server; everything running on a Linux machine in a closet or on a couple of rented VMs. You'd like to know about your code's errors and performance issues right when they happen; you'd like to be alerted when the disk or RAM is close to full; you'd also like to be able to change the database schema, upgrade the Python/PHP version, or upgrade PostgreSQL with only a short maintenance window or none at all. However, the team is too small to take on the complexity of the entire modern cloud stack with Kubernetes, Helm, ELK, and so on.

In a situation like this, what would be your approach to get 80% of the value out of DevOps with 20% of the ongoing maintenance effort?

https://redd.it/1i9rhpn
@r_devops

What are some interesting errors you encountered in your pipeline and how did you resolve them?

Title - curious to see what strange issues people have ran into and how you were the hero that day :)

https://redd.it/1i9tala
@r_devops

When you must perform operations outside of CI/CD what do you prefer, CLI or User Interface?

I'm really just curious the proportion of people in this field who love using CLI vs UI systems for setup, management, and cleanup of infrastructure, service, and application resources.

I'd love to get your opinion. I guess I'd have to classify myself as a hybrid. I love the CLI and always have because you can automate it, but I have also found tremendous value from UI interfaces like Kubernetes Dashboard, ArgoCD, Grafana, etc... I'd really like to have a way to describe the usage and benefits of each to business folks, and in my journey I feel like I need outside perspective.

https://redd.it/1i9vnpv
@r_devops

Installing Kong API Gateway on GKE and deploying an application with OIDC authentication.

Comprehensive guide for setting up a GKE cluster with Terraform, installing Kong API Gateway, and deploying an application with OIDC authentication.

Kong API is widely used because it provides a scalable and flexible solution for managing and securing APIs https://medium.com/@rasvihostings/kong-api-gateway-on-gke-8c8d500fe3f3

https://redd.it/1i9wgp4
@r_devops

I’m afraid to change job

Guys, at the moment, I’m working as a DevOps professional at the same company for one year and nine months. During this time, I’ve learned a lot, but I feel like I’m stuck right now. I don’t feel very happy with the current tech stack, and there are some aspects of leadership that I don’t like. I’m a Junior DevOps, but in my role, I’ve never worked exclusively on ‘Junior-level’ tasks—I just do what needs to be done. Do you think I should change jobs? I’d appreciate advice from people with more experience.

https://redd.it/1i9zzb6
@r_devops

i’m building an up-to-date, community-driven list of Awesome European Tech ! Can you suggest some projects or companies to include in the DevOps section that I’ll be creating, please?

i’m building an up-to-date, community-driven list of Awesome European Tech projects! This includes browsers, productivity tools, FinTech, and more, all focusing on privacy, sustainability, and innovation. The goal is to support European startups and projects (from the EU and EFTA that comply with GDPR regulations), helping to strengthen the European tech ecosystem.

If you'd like to show your support, feel free to leave a star on GitHub or contribute to expanding the list it’s greatly appreciated! Keep in mind, this list is just a starting point and will grow and evolve over time.

Disclaimer: This project does not aim to exclude or avoid products from outside Europe. There are many outstanding global tools that are widely used and appreciated. The focus is simply to highlight and support European projects that align with the values of privacy, sustainability, and innovation.

https://redd.it/1ia0qd1
@r_devops

Freelance

I'm searching for some part-time job, but not sure which platform to use. I wanted to subscribe on flexjobs, but then I saw that a lot of people said not to do it. After that I wanted to check remotive, but didn't find that much source.

Any other platforms?

https://redd.it/1ia1ku3
@r_devops

Wonder if customer is ready for DevOps practices

I’m in a bit of a tricky situation. I’m working on a customer project where I need to deliver infrastructure that includes a VPC, subnets, routing, an internet gateway, and a few customized EC2 instances so the customer can upload, install, and run their application code. The customer has requested that the infrastructure code be written in Terraform, which I’m totally fine with. However, they don’t have any CI/CD pipelines defined or implemented, as these will be their first components in the cloud.

My question is: Does it still make sense to deliver Infrastructure as Code (IaC) if they don’t have pipelines configured?

https://redd.it/1ia13bx
@r_devops

Why Am I Not Getting Any Responses? Please Roast My Resume!

Hi Reddit,

I’m looking for some honest feedback on my resume and job search strategy because I’m completely stuck. Here’s my situation:

* **Experience**: I’ve been working as a DevOps engineer for 4 years.
* **Education**: I’m currently pursuing my master’s at the University of Washington, graduating in March.
* **Applications**: I’ve been applying to DevOps/SRE roles requiring up to 5 years of experience and entry-level backend positions. So far, I haven’t heard back from anything, even though I feel the roles align closely with my work experience and most of the “keywords” are already in my resume.

This lack of responses has led me to two possible conclusions:

1. I’m missing key skills that are now must-haves for DevOps/SRE roles in this market.
2. My resume is the problem—it’s not presenting my experience effectively or isn’t optimized for ATS.

Resume: [https://imgur.com/e0bfb5q](https://imgur.com/e0bfb5q)

**The Resume Problem**: I’ve been told over and over to keep my resume to 1 page, but that’s caused issues. To fit everything, I’ve had to leave things out or combine them in ways that don’t fully capture the impact of my work. Some parts even feel incoherent. Is it acceptable to have a resume that overflows to 2 pages? How does a 2-page resume work with current ATS systems?

**What I Need**:

* Brutal feedback on my resume: What’s wrong with it? What should I change?
* Advice on navigating the job market in 2025 for DevOps/SRE roles or backend positions: Am I missing something big?

I’ve attached my resume, and I’d really appreciate any critiques or suggestions, no matter how small. Is my resume totally broken, or am I approaching the job market the wrong way?

I would really really appreciate any help!


https://redd.it/1ia7k38
@r_devops

Orchestration Tooling

What tooling do you all recommend for small teams to automate repetitive tasks across various platforms? Looking at tools like Octopus deploy but need something open source or very inexpensive. We use terraform / prometheus / all the usual stuff.

https://redd.it/1ia7ym0
@r_devops

Devops syllabus in my college. How good is it?

https://imgur.com/a/0VVO6am

Devops is an Honors subject in my department. This the syllabus of my previous semester. There are still 3 more semesters. How good is it?

https://redd.it/1iaaa3w
@r_devops

What branching strategies are best practice?

I've worked as a Devops Engineer for a small company for three years and for the most part it's always been just me working on projects.
I tend to have a main branch which is what is deployed to production.
I also have a branch called 'uat-testing'. Which in our CiCd just points to a different Kubernetes cluster with Argocd apps.
Whenever I do development, I tend to do this in a feature branch, or a development branch.

When I'm ready to deploy to UAT, I just checkout to uat and merge the chains in, push and Argo deploys.
Our QA team tests, then when happy, I checkout to main, merge, push, and Argo deploys.

I've just moved jobs, and I've been told that my git branch strategy is horrendous.
And I should be using tags. This is all new to me, so I'm looking for resources and advice.
What is the best practice for git branching strategies?
Is it completely dependent on your application, what you are deploying?
The example above was for deploying manifests into K8s

https://redd.it/1iaao9j
@r_devops

Managing PG roles and grants as IaC

Hey team,

So we use AWS RDS Postgres and we have Terraform using this provider: https://registry.terraform.io/providers/cyrilgdn/postgresql/latest

We have a monolithic setup, with a module for the whole infra including these roles and perms.

We’ve been executing everything locally, which means our devs have to port-forward to the DB and have the DB credentials locally.

We want to move to a CI/CD approach. However our self-hosted runners, which reside in a dedicated CI/CD account need network connectivity to the prod DB - which is sub optimal. They also have to have access to highly privileged credentials.

Broadly speaking I think our options are:
- Ditch Terraform for this particular use case in favour of another tool or pure SQL?
- Extract the Postgres logic from the monolith infra deployment into its own module
- Use private links or a VPN like Tailscale to connect between runners and the DB. Store the secrets in Secrets manager.
- Manage grants and permits as part of application migration?

There must be a better way! What do you do or have done in this situation?

https://redd.it/1iaau3f
@r_devops