86
Reddit DevOps. #devops Thanks @reddit2telegram and @r_channels
What are the self-service tools/CLI automation you have build around AWS
Hello Experts,
I would like to listen What are the self-service tools/CLI/platforms , solutions or process/ automation you have build around AWS which helped in your Organization to solve big head-ache.
https://redd.it/1hknljj
@r_devops
Bootstrapping new EKS cluster with Helm releases in Terraform
I've hit a bit of a frustrating road block trying to sort something out, pretty much what the title says.
I have a client that is using Terraform to build EKS clusters. However, they also want to use the Helm Terraform provider to bootstrap new clusters with some of their tools/apps and open source tools like karpenter.
I don't use Helm with Terraform so I'm not super familiar on the tips and tricks of it. From what I understand, you need to set the kubernetes cluster, the certificate, etc.. in the helm provider block. Which you can't do because the cluster doesn't exist yet.
I've gotten some opposition to any sort of hacky workarounds, like creating a separate directory that the deployment pipeline would read and apply after the initial deployment is done.
I'm curious if anyone has a solution for this or if I'm entirely not understanding something.
https://redd.it/1hkosy0
@r_devops
Learn about Helm charts with open-source contributions. We've just tagged several "good first issues" in the SigNoz(OS alternative to DataDog) Helm charts repository and we would love any help in tackling them.
Hey everyone! We've just tagged several "good first issues" in our Helm charts repository (https://github.com/SigNoz/charts), and we'd love your help tackling them. This is a perfect opportunity to get hands-on experience with Helm chart development and contribute to open-source.
Many of these issues focus on chart improvements, documentation updates, and value file enhancements - perfect for both beginners and those looking to deepen their Kubernetes/Helm expertise. Some areas you'll get to work with:
Helm template functions and best practices
Kubernetes resource definitions
Chart dependency management
Values file structuring
To get started:
1. Check out our issues labeled good-first-issue
2. Comment on any issue you'd like to work on
3. Fork the repo and submit your PR
We strive to provide detailed feedback and support throughout the PR process.
https://redd.it/1hklwpt
@r_devops
Resources provisioned by pipelines should be protected
We want resources provisioned by pipelines should be protected to avoid re-configuration after another pipeline run
to make sure infrastructure resources are not broken.
Two solutions coming to my mind :
1. Identity-based policies that rely on resource tags: the tags are used to identify resources that must not be modified outside the pipeline
2. Resource-based policies that state something like "this resource can only be modified by SpecialRoleX".
How do you designed this for your environment ? Do you have some design/process in place ?
May you put some guideance.
https://redd.it/1hkktfg
@r_devops
Got an internship opportunity, still on the beginning of the roadmap
Hi, I just started my journey on the DevOps roadmap, and I am still learning Linux commands and administration. Today, I was contacted by one of my country's biggest companies for a long internship opportunity. So, basically, I have three days to learn whatever headlines I need to pass this interview. If anyone has any tips for me, that would be delightful. Thank you.
https://redd.it/1hkb3ze
@r_devops
Current state of bare metal provisioning
What is the current state of bare metal provisioning in the context of private cloud environment? Is it yet possible to rack a host and cable the host and automagically have its ipmi/redfish configured and OS installed? Are there solutions in the wild that support this yet or are we still stuck in technology 10 year old where at most the BMC card will get a DHCP ip address and then have to be remotely connected over ipmi or redfish and configured for PXE boot or whatever to get the OS installed?
https://redd.it/1hk92it
@r_devops
Anyone struggling to get interviews if they don't have XP with the company's pref. cloud?
I am struggling to get interviews for devops/sre positions if company uses GCP as I only have AWS experience.
I am not sure if I am actually getting filtered based on keywords or market is tough or they simply don't want to waste their time with someone who never touched GCP.
Other than that I usually use whatever they are using be it ArgoCD,Ansible or monitoring stack.
Any tips on this?
https://redd.it/1hk27bi
@r_devops
Anyone else have this?
I’m a senior member of our DevOps team. The team isn’t very big and most of the other teammates have been here a year or less. Any time I bring up an idea, my boss is quick to say “no.”
In addition, I feel like I’m doing the lion’s share and our Jira board, more or less shows that. Anyone else feel like they’re not heard or feel like they get zero support from “leadership”?
Aside from just picking up and leaving, has anyone else tried to make things better successfully?
https://redd.it/1hk26go
@r_devops
how to prep for AI future?
It looks like we now have a fairly clear direction on AI future at least on dev/DevOps side where we're looking at team sizes to be reduced by 60-70% and the remaining intermediate to senior level folks will be overseeing AI agents do the brunt of the work.
How are you preparing for this future and be marketable so that you can be that top 30% dev/devops who remains and not laid off.
https://redd.it/1hk12si
@r_devops
Devops Resources for a FullStack Engineer
Hi guys,
I am a fullstack engineer with Typescript, Python , Express, Django ,React and PSQL/SQL as my stack looking to improve my knowledge of devops. I can build fully fledged applications and would like to know more about the infrastructure side of things. Im looking for direction on what books I can read as I learn best with them. I dont want something too detailed i.e books that a senior devops engineer would be reading but ones that will teach me the fundamentals and tooling required to deploy my apps, set up ci/cd pipelines and how to use docker/kubernetes to scale an app. Any recommendations will be greatly appreciated. Thanks
https://redd.it/1hjvrmc
@r_devops
What’s Your Biggest DevOps Struggle? Help Me Shape My Next Deep-Dive Blog Post
I'm planning my next blog post and want to make it genuinely valuable by addressing real struggles that DevOps professionals face with emerging technology trends.
Could you share the topics you'd like me to explore in-depth within the DevOps space? Here are a couple of ideas I’ve already shortlisted, but feel free to suggest others:
1. Kubernetes Security: A comprehensive guide to securing your application end-to-end, from CI/CD pipelines to the Kubernetes cluster.
2. OpenTelemetry Best Practices: Practical, real-world examples of using OpenTelemetry for end-to-end tracing and telemetry in Microsoft services, including exporting data to different backends with best practices.
Is there anything specific you’re struggling with or a topic you’d love to see explained in detail?
I promise to thoroughly research the subject, consult with experts if needed, and find real-world use cases at scale to create an informative article. If the content gets lengthy, I’ll break it into a series of posts to cover the topic comprehensively. Let me know your thoughts!
I will write and post on medium so will be accessbile to all free of charge.
promise: I will only write after thoroughly research the subject
https://redd.it/1hjoa3q
@r_devops
Vector embeddings, tokenization, and Vector databases
Understand rasvihostings/vector-embeddings-tokenization-and-vector-databases-041c93049487">vector embeddings, tokenization, and vector databases with a clear explanation and practical examples
A vector database is a certain type of database which is created to store and manipulate vector embeddings, which are the numerical representations of data such as text, images or audio in a dimensional space. These embeddings are usually produced by machine learning algorithms and are employed for various applications such as similarity search, recommendation systems or semantic indexing.
For instance, in a vector database, you can have embeddings of product images and therefore, easily get recommendations for similar products when a user uploads an image.
Now you might get confused what is dementional, you can get clear understanding by reading this article
rasvihostings/vector-embeddings-tokenization-and-vector-databases-041c93049487">**rasvihostings/vector-embeddings-tokenization-and-vector-databases-041c93049487" rel="nofollow">https://medium.com/@rasvihostings/vector-embeddings-tokenization-and-vector-databases-041c93049487**
https://redd.it/1hjjnbr
@r_devops
Need help with apache kafka strimzi ( im crying)
not sure if this is the right place but i really need urgent help with this
My senior ask me to deploy kafka and specifically wants the strimzi (not confluent) and i dont know how to do it.
I read through the website , it does say you can deploy it on docker, or minikube (not for prod use) . I tested both. Since minikube is not for prod, i went with the docker option.
I created a Dockerfile that will pull the image and set the environmental variables, most importantly the internal and external port / listeners, set the healthcheck, then CMD run the bash script required to start it.
I then message my senior for the Pull Request. He took a glance, didnt really read all of it and said... wait why do we need a dockerfile for this? strimzi operator already has an image with all the broker,zookeeper,boostrap in it, you can deploy straight to kubernetes just using that image.
He is busy and didnt really said much but told me to get it done by monday. I spent all night reading up and i dont understand how this can be done in any other way than using a dockerfile. Dockerfile is where we write all the instructions on how to deploy isnt it? I cant really think of another way of deploying it to our kubernetes (EKS).
Our pipeline right now for everything else is Dockerfile -> github actions trigger -> create image in ECR -> deploy to EKS based on .kube deployment.yaml, service.yaml files
Any one knows how to do this?
https://redd.it/1hjde83
@r_devops
Developers who moved to Devops could ever come back ?
I mean you were a good developer and organization put you in a situation you couldn't say no to taking extra responsibility of Devops and struggle now to come back to development
I know some who felt it overwhelming to go back to full fledged development and there are some who become an IT dept in themselves, one man army to the market outside
How about you
https://redd.it/1hjcnss
@r_devops
How do you tell your security you can't keep your container images to be 0 vulnerability?
I am using Debian and the Debian maintainer don't actively patch every packages immediately.
https://redd.it/1hjb5ld
@r_devops
Question about nfs-kernel-server / nfs-utils
Hello everyone,
I'm quite new to all devops concept so this question might be a little noob. I'm working on a WordPress website that has many servers grouped together under a load balancer. We need to find a way to share the media files because WordPress stores all images under a directory called `uploads` by default and it would take a lot of efforts to rewrite this behavior. Therefore, I was thinking about setting up another server, turn it into a "host" server to contain all the asset files under a folder. This folder will then be mounted on my web servers (the ones that are grouped together by a load balancer) so I could simply create a symbolic link to replace the default `uploads` directory with the mounted directory.
Here are some of my questions:
1) Is this `nfs-kernel-server` or `nfs-utils` package production ready?
2) Would it be capable of handling high requests? My site doesn't have that much of traffic but we can have around \~50 to \~100 requests at once, depends on the time of the day.
I know there are other solutions with rich documentation such as JuiceFS or GlusterFS, however I was trying to try out if a native package of Linux could handle this task or not and also, I think the installation process of `nfs-utils` is quite simple, my website is only a WordPress service and it doesn't make sense if the setup is too complicated for just a WordPress site. However, since I'm quite a newbie to devops, I'm very open to hear your opinions on how should I handle this case. Thank you all.
https://redd.it/1hkpmxp
@r_devops
Mounting to a volume makes empty directory
So i had been trying to self host erpnext, since couple of days. I was using Azure file shares and container apps to create container named backend.
Now whenever i run the container without volume it gets running and via console i can verify that the container has a folder named sites and it has all the files required and was added from the start.
Now when i do deployment with volume(Azure file store) and look through container via console, the sites folder is there but it's empty, i had absolutely new volume! now if i create a file say hello.txt in that folder it's reflected in the azure files and vice versa.
how to stop this thing. i mean i need the initial files of that folder and need to have that in volume too, if anyone has idea around this things please help me with your findings!
Thanks
One of my findings is that named volumes and bind volumes behaves in different ways, like what we have in local machine which works as expected copies the container data to volume if that's empty, but bind volumes just binds that directory doesn't matter there was data or not, not 100% sure but azure files might use bind volumes?? then how to overcome this!!
https://redd.it/1hknluy
@r_devops
Requirements of AWS, GCP, Azure everywhere. But what is it that employers want?
So I am looking to get into the Cloud Platforms, having experience with On-Premise, Edge Computing and Hardware specific software and containerization. I get that most job postings will add in Cloud-Platforms as a requirement however, no one seems to be clear as to what it is they want to do with these platforms.
It is always Good Knowledge of Cloud-platforms like GCP, AWS, Azure.
This poses as a learning challenge for me. If it were merely stopping starting VMs can I just add these platforms to the CV? Nobody mentions any Cloud-DBs or any specific cloud-specific offerings in the postings.
So if I were to proceed with getting into Cloud Platforms what am I supposed to learn?
https://redd.it/1hklgkf
@r_devops
What should an Infrastructure Director learn?
Hi all, new here. I am currently an Infra Director in a relatively conservative government environment, overseeing the traditional infra (OS, network, etc.) and Azure landing zone. I am looking to move to a new external role with exposure to more modern technologies, and I noticed many of these roles also include overseeing the DevOps functions. Any advice on what kind of DevOps knowledge I should acquire in order to prepare myself for the move? I have no problem learning new things quickly and also had developed codes during university.
https://redd.it/1hkidt6
@r_devops
With the year ending, how much of a raise did you get prior to your last year’s salary?
It was a wild year for me, i got a new job earlier this year with a 23% raise only to get laid off few months later. Got a new job with a -23% change compared to my last. So 0% for me.
https://redd.it/1hk9ca6
@r_devops
Bridging the Gap Between Cloud Providers and Users
We’re developing a platform that helps enterprises select the best cloud providers based on their unique needs. Here’s what we’re focusing on:
Tailored Cloud Provider Mapping: We’ll create a complete map for each enterprise, ensuring they get the ideal cloud choice based on optimization, compliance, ecosystem, performance etc.
Unbiased, Independent Data: Instead of just benchmarking, we’ll gather unbiased data by working directly with cloud providers to offer real insights.
Network of Providers: Starting with 5 providers, expanding to 10 by mid-2025, offering the most efficient cloud options.
Tier 1 & Tier 2 Providers: A combination of both to ensure the best fit for performance and cost.
Do you think developers, enterprises value a combination of Tier 1 and Tier 2 providers, or would they prefer to focus on Tier 2 providers for cost-efficiency? We’d love to hear your thoughts on the approach!
Looking forward to your feedback!
https://redd.it/1hk3jtq
@r_devops
Dashboard for tenant management
Hey everyone,
I have a multi-tenant application and I'd like to have a separate app which manages the tenants. Is there any existing, free and self-hosted solution on the market? Honestly I don't even know what keywords should I search for, I want to explore my possibilities before starting to code my own one.
What I need is a simple dashboard/UI to create and manage tenants. I want to customize them (company logo, different theme, pricing plan) in a centralized space and configure where to deploy.
https://redd.it/1hk3n5x
@r_devops
Can I use an external monitor just for the screen in cka exam
Can I give my cka exam by connecting my ThinkPad lenovo laptop to an external monitor (just for the screen) since I will be using laptop camera and keyboard because monitor doesn't have a webcam in it.
Please help my exam is next week.
Or should I just give it on ThinkPad lenovo laptop?
https://redd.it/1hk2c8n
@r_devops
PAID HELP NEEDED Totally bricked my Hetzner VPS due to being impatient
Hey everyone, I’ve had a frustrating experience with my VPS, and I’d love some advice—or even better, a helping hand! I was trying to set up SSL certificates for two domains using Nginx Proxy Manager. As soon as I added SSL for the second domain, everything went downhill.
The main site broke, connections were refused, and Proxy Manager kept throwing errors about missing certificates. To be honest, I only have basic knowledge of servers. My developer, who usually handles this, is currently ill, so I tried to take things into my own hands to get a client’s site live. Big mistake—it’s a total mess now. I’ve spent hours troubleshooting, checking symbolic links, DNS records, Docker logs, and more, but I feel like I’m just poking in the dark.
If anyone has 30 minutes to help untangle this, I’ll gladly buy you a coffee (virtually, of course!). And any advice on how to prevent this kind of disaster in the future would also be greatly appreciated!
https://redd.it/1hjy1ac
@r_devops
Affordable monitoring and log aggregation service for startup (cloudwatch sucks)
Hey everyone, Front-end / full stack dev here just joined a very early startup as the first engineer practically.
Company uses AWS Lambda, Serverless.com, Vercel.
In my previous company we used Datadog. I liked it a lot. Gave a lot of confidence and useful metrics. Nice UI.
I understand it's expensive for a startup. But honestly navigating cloudwatch is such a pain. Purely the fact that I'm unable to find one single stream of logs (groups? Wtf??) or that I can't figure out how to filter by severity or generally..
I've seen Grafana?
I can't possibly ask our founder to pay for something like Datadog, given how expensive it can get..
Serverless has some logging and monitoring built in but it won't log stuff from our Front-end repo / BFF...
Any recommendations from you experts tremendously appreciated.
Peace and enjoy the winter break / holidays.
https://redd.it/1hjtz2c
@r_devops
Create EKS cluster from git repository
Hi,
I'm new with devops and I want to install eks cluster with cloudformation from git repository.
When a push from main is triggered I want to execute a bash script with aws cli or eksctl (like aws eks ...) to create the cluster or update it if there is a diff. So I need to insall aws cli, eksctl, etc inside the cluster to execute all these command.
In the futur, I want to execute some other cli command like installing helm to install some helm chart like argocd, karpenter, etc. I want to avoid GHA to install helm, and other stuff. I want to put a bash script in the cluster and execute it.
It's possible to do that? I'll have some other repo, for example create a rds database with same approach. Another one to install some message broker, etc
After that other repo (service app) will use argocd to be deploy app but for cloud stuff I need to execute some script to install/update stuff.
I don't know if it's the best way. Or maybe a docker image with some tools like aws cli, eksctl, helm charts available in the image, and run this image to be inside the container and execute commands (install helm charts, install aws eks addon, etc). But I will need a repo to build this docker image, push it to ECR, etc, etc. Different repo like eks-cluster, rds-db will be able to use this image and use binaries to do stuff I want.
https://redd.it/1hjmmzl
@r_devops
Should I take a year off to become an expert (no money issue)
Hello guys,
For personal reasons, I can take a year off to be focus on becoming an expert in several high paying tech in the DeVops/ Cloud space.
For context, I am a DevOps engineer with around 5 years of experience mainly AWS (4 certs), terraform, kubernetes , python etc ..
I am good but not the best.
I do not have issue to found a job, I have a security cushion and I get passive income that cover my monthly expense so if any issue I will be okay.
So why do this ?
I am currently at a crossroads where I always wanted to go very deep and becoming expert for Kubernetes, DevSecOps, FinOps etc.
My goal would be for that year to do freelance and learning same time .
Find a side hustle as well and just be focus with projects , advanced hands labs etc .
I think this year off would skyrocket my career later on, I am only 29 years old.
I am in Europe ( France).
What do you think any ideas ?
https://redd.it/1hjjjov
@r_devops
Tips for Deploying a Laravel App with Docker (Simplified Installation and Updates)
Hello everyone,
I have developed a Laravel application that uses PHP 8.1, Apache, and MySQL, and I would like to distribute this application to my clients in the simplest way possible. The goal is for clients to be able to install it on their own (or with minimal intervention from me) using Docker.
I’d like to know what the best practices or most common solutions are for:
1. Creating a Docker configuration that includes Laravel, Apache, and MySQL, and is easy to use.
2. Automating the initial installation, including steps like creating the .env file, running migrations, etc.
3. Managing future software updates, making the process as simple as possible for both me and the clients.
If you have experience with similar challenges or suggestions on how to approach this, I’d be really grateful!
Thanks in advance.
https://redd.it/1hjcnwr
@r_devops
Setting up VPN to access private VPC on AWS, and general guidance on how to do that.
I am working on a personal project, and by doing it, I am attempting to familiarize myself with working with AWS, and working with IAC tools.
I am using terraform to provision the resources, and serverless framework to deploy my docker images to ECS. That has been working fine.
When I create a database, I connected it to the private VPC, however that means I can test my docker images locally. I did some research, and I understood that most people would set up a VPN that would connect the private network to your network.
To do that, I understood that I need to use EC2, and OpenVPN. Now comes my question, from my understanding terraform is used for provisioning resources, so I can't use only terraform to do this since I would need SSH access to the machine to configure terraform. That means I would need to use Ansible ?
Is my reasoning correct ? And what are the best practices here, and can someone give me some tips ? I haven't found a guide on this online, and I am struggling.
https://redd.it/1hjbjn8
@r_devops
Why do I have this issue with Copilot in VS Code?
VS Code, even after uninstalling, deleting the app data, re-installing, and everything, it still fails to send messages or allow me to select the model. In short, the model drop-down is absent, prompts fail to send, etc. I also have a similar issue with "First parameter has member 'readable' that is not a ReadableStream."
Anyone understand this?
https://redd.it/1hj2ucg
@r_devops