out a way that instead of mocking the function if i mock the db mocker.patch('app.resources',return_value = mock_data) then it returned the expected result. but this beats the purpose of testing using mock

https://redd.it/1j9n7ld
@r_devops

Need advice as I am struggling!


Hey guys, long story short. I accepted a full time job from a a contracting company, and the company I am contracted to is a fortune 500. This is my first career job out of college I had no experience, first two years as integration dev went so slow. Low workload ended up learning a lot. beginning of 3rd year they switch me to a DevOps engineer role. Workload is 10X I’m not shitting u, I start at 6am and don’t finish until 7 8pm but Im only allowed to work 40hrs as I get salary but realistically, I work close to 55hrs or the job wont be done . They pay me 65k/year didn’t have a raise in the last three years. I asked for one but the literally said no or u can seek other opportunities, I love the team and this new role I learned a shit ton in the first 3 months than my last two years. Should I just stick with it for another year or look for another job? Most of my college friends got a full time role within the company and get $100k+, raises and bounces yearly. While Im stock!
Financially Im not doing okay as school loans and inflation, rent ect.


https://redd.it/1j9qaw5
@r_devops

Reproducable Server without Nix/NixOS?

Hi! I've been maintaining servers on bare metal for a while now, and so far I've rolled most of them manually, and for some of them I used NixOS.

I've enjoyed using NixOS. I like it because it allows me to recreate my server very easily when moving hosting providers. I don't want to bind myself to a hosting provider because it's an instance of vendor lock-in (since it takes significant time and effort to move to another service provider).

However, when using NixOS, I've often experienced that support for certain newer services (e.g. Dendrite) was not good (and writing Nix unfortunately feels very inaccessible and unintuitive to me). Also, there was no way to make sure I wasn't using compromised packages (since vulnix was discontinued), making my server vulnerable to CVEs and supply chain attacks.

Guix' Scheme language feels very verbose and cumbersome to read to me, so I'm not sure I want to go that route either.

Therefore, my question is: Can I get the reliable reproducability of NixOS with a different tool or set of tools as well? Ideally without the cons mentioned above, of course. I'm currently already considering using podman, but that still leaves me with the base OS not being reproducable... right? Maybe a tool like Pulumi is what I should be using here? Looking forward to your recommendations, pointers, suggestions and ideas! And questions, of course :)

Thank you for your time! 💜

Addendum: I'm intending to rent a single server to host some self-hosted services on (stuff like a Mastodon server, a Minecraft server, a CryptPad server, maybe Excalidraw). Ideally I will be able to move the services I host from one hosting provider to another with minimum effort.

https://redd.it/1j9ql32
@r_devops

awk pod "observability"

(I'm a noob and I'm making this post just to ask for some ideas before actually go in depth).

I have some pods on my learning awk environment and i would like to be "notified", or somehow be aware, when they fall on a "Not Ready" status.

I know that their restart could be managed through probes but i was thinking if there is a different approach.

So basically in my mind i go to an organized page or something and i see just the pods that are stuck on "not ready" state and possibly i get some notifications.

https://redd.it/1j9s2pd
@r_devops

Stackql - The New Approach To Querying And Provisioning Cloud Services

If you know of Steampipe with which you use SQL to query Cloud services, this is a similar but more modern tool

https://www.i-programmer.info/news/84-database/17885-stackql-the-new-approach-to-querying-and-provisioning-cloud-services-.html

https://redd.it/1j9v3oi
@r_devops

How do I safely practice with cloud services like AWS, GCP, Azure etc. for learning by putting a hard capping of maximum bill?

I am a frontend developer and it seems like every employer still wants cloud experience. I want to make a learning project using cloud service which I do not delete or tear down hourly or daily but actually keep it live for few months.
I would prefer AWS because I have had a little bit of exposure but any of the big 3 cloud services is fine.

What is the best and safest way to put a hard cap on AWS bill and charges? Like if I do not want to spend more than $2 per month how would I ensure the bill never goes about $2?

From what I got to know billing itself is not immediate and billing alerts/notifications could also be delayed. And also we may miss an alarm because of any reason like we may be sleeping at the time, or sick at the time.

If not in AWS, can we put hard caps in Azure or GCP?

https://redd.it/1j9uhmx
@r_devops

DevOps Employees Well-Being

I read this article about DevOps employees' burn-out -- https://itrevolution.com/articles/addressing-burnout-in-our-devops-community-through-demings-lens/

If you are given the power to change one thing in your job to mitigate burn out, what would you do?

https://redd.it/1j9xu73
@r_devops

Am I going through burnout, and/or just dealing with how life is?

The short of it is that I've put more effort than I likely should've over the last 2 years, hoping for a decent salary rise and/or promotion, but ended up getting a metaphorical slap in the face instead.

I'm now dealing with pretty severe mental and physical fatigue to the point I can barely leave my bed until later in the day (thank god for remote work); I've completely lost any motivation to work where I feel physical strain when performing even simple tasks, and I kind of just dread having to wake up every day. Job hunting under these circumstances also feels impossible.

I'm 90% certain I could've done the absolute bare minimum and ended up in the exact same spot I am in now, where my progression appears to be based entirely off of mystical vibes rather than any sort of merit.

I just want to give up and scream, but can't really afford to do so, but now I just feel stuck with the difficulties on moving on from my current role. I don't really know what to even do at this point, so I'm just going day-by-day until something magically happens/gets better. I can't tell if my expectations were just unrealistic, or if I'm right to feel the way I do.

https://redd.it/1j9yxuo
@r_devops

Notemod: Free note-taking and task app

Hello friends. I wanted to share with you my free and open source note and task creation application that I created using only HTML JS and CSS. I published the whole project as a single HTML file on Github.

I'm looking for your feedback, especially on the functionality and visual design.

For those who want to contribute or use it offline on their computer:

https://github.com/orayemre/Notemod

For those who want to examine directly online:

https://app-notemod.blogspot.com/

https://redd.it/1ja0kb8
@r_devops

Monolithic repo or multiple repos?

Which Git based repository model do you think works best for managing source code for Terraform modules?

View Poll

https://redd.it/1j9xbm0
@r_devops

Pathway to become DevOps Engineer

Hello, I am currently working as a Software Engineer and I have got 3+ years of experience in the field. My goal is to lean towards DevOps. I currently work for a company that I believe hasn’t got much to do with DevOps (this is long to explain, so don’t ask me how/why). In the next two years, I would like to see myself as a DevOps Engineer. So, what’s the best way to become DevOps Engineer?

The following I have got in my mind.

1. Do certifications (eg: Azure DevOps expert, AWS DevOps). Can do with the help of my organisation.
2. Although certifications can boost LinkedIn profile and activity, I am aware that’s not enough. So, based on my learnings through certifications and open source materials, have some hobby projects that showcase my skills related to DevOps.
3. Try to impose the skills acquired through these learnings into a read world project within my organisation.

Any suggestions and advice welcome.

Thanks.


https://redd.it/1j9pqkl
@r_devops

Help me understand IOPs

For the longest time I've just buried my head in the sand when it comes to IOPs.
I believe I understand it conceptually..
We have Input Output, and depending on the block size, you can have a set amount of Inputs per second, and a set amount of Output per second.

But how does this translate in the real world? When you're creating an application, how do you determine how many IOPs you will need? How do you measure it?

Sorry if this is a very novice question, but it's something I've just always struggled to fully grasp.

https://redd.it/1ja6src
@r_devops

Blog: Ingress in Kubernetes with Nginx

Hi All,
I've seen several people that are confused between Ingress and Ingress Controller so, wrote this blog that gives a clarification on a high level on what they are and to better understand the scenarios.

https://medium.com/@kedarnath93/ingress-in-kubernetes-with-nginx-ed31607fa339

https://redd.it/1ja89xj
@r_devops

Join Online Webinar: SCA or SAST - How They Complement Each Other for Stronger Security?

𝐑𝐞𝐠𝐢𝐬𝐭𝐞𝐫 𝐍𝐨𝐰 𝐟𝐨𝐫 𝐎𝐮𝐫 𝐍𝐞𝐱𝐭 𝐒𝐚𝐟𝐞𝐃𝐞𝐯 𝐓𝐚𝐥𝐤 𝐒𝐂𝐀 𝐨𝐫 𝐒𝐀𝐒𝐓 \- 𝐇𝐨𝐰 𝐓𝐡𝐞𝐲 𝐂𝐨𝐦𝐩𝐥𝐞𝐦𝐞𝐧𝐭 𝐄𝐚𝐜𝐡 𝐎𝐭𝐡𝐞𝐫 𝐟𝐨𝐫 𝐒𝐭𝐫𝐨𝐧𝐠𝐞𝐫 𝐒𝐞𝐜𝐮𝐫𝐢𝐭𝐲? Most security teams use SCA and SAST separately, which can lead to alert fatigue, fragmented insights, and missed risks. Instead of choosing one over the other, the real question is: How can they work together to create a more effective security strategy. Do you want to find out?

📅 Date: 𝐌𝐚𝐫𝐜𝐡 𝟐𝟕𝐭𝐡

⌛ Time: 𝟏𝟕:𝟎𝟎 (𝐂𝐄𝐒𝐓) / 𝟏𝟐:𝟎𝟎 (𝐄𝐃𝐓)

You can register here - https://www.linkedin.com/events/7305883546043215873/

https://redd.it/1ja9by5
@r_devops

Free AI diagram generator - compatible with drawio

We are offering a free version of draft1 here: https://app.draft1.ai/tryfree

https://redd.it/1ja9tpx
@r_devops

I'm looking for some recurrent advice/mentoring

Hey there!

I'd like to get into devops and sysadmin. I have some knowledge in web development with the JS stack and a bit of C# for desktop apps but I'm not that keen on pursuing a career doing CRUDs for a living so I'm thinking devops might be an interesting path to follow.

So far I'm almost finishing an associate degree and I'm continuing with a full software engineer degree and I find myself looking for a job next year so I can afford my studies later.

That being said I'd love some guidance and someone who really knows about the field and can guide me through my learning process. Of course I'm not asking for a full time teacher, but someone who I can talk frequently (maybe twice a month?) so my process can be tracked and be better oriented. Would anyone be interested in that?

And yes, I know there's tools such as roadmap.sh and others, but I think having someone guiding me and calling me out if I didn't do what he/she suggested and I agree to would make my commitment skyrocket

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

Platform Engineering should be more than DevOps

I've been thinking about the transition from DevOps to Platform Engineering. (Hence the questions.) DevOps was meant to reduce silos, but my personal opinion is it doesn't scale to have everyone be both Dev and Ops. Platform Engineering emerged as the next logical step, but I think it needs a clear center for it to be truly valuable. It needs to be more than just specialized teams handling CI, infrastructure, or Kubernetes setup.

That center should be developer experience. The customer of the platform is the the developers building applications and services. This gives pe a much broader scope than just devops - it's about removing friction everywhere.

I got this idea from Spotify but, this means focusing on various aspects of the developer journey:

Conduct regular developer surveys to identify specific friction points, then prioritize solutions for the most common obstacles.
Fix the problems identified and repeat

So, is platform engineering primarily a developer experience discipline, or is it mainly focused on simplifying operations and deployment? What specific metrics best capture platform success?

I want it be about DevEx and I've written a blog post arguing this. PE should concentrate on the larger mission of eliminating all friction and toil across the entire development lifecycle. Now i just ahve to convince you, my coworkers and the rest of the world.

Edit:
Here are the principles I am attributing to Pia Nilsson:

"Platform Takes the Pain": Platform teams should own migration difficulties, not feature teams
Drive Adoption: Be accountable for teams actually using your platform tools
Measure: Track metrics like "Time to First Commit", "Time to Production" and do dev survey's to quantify improvement
Standards Enable Speed: Well-implemented standards actually accelerate development. Design systems that don't depend on individual "hero" engineers

https://redd.it/1jae9fv
@r_devops

Google Monorepo pipeline build times

I read that Google uses large monorepo but how do they manage their pipeline builds. Do they also run build for each merge to their main branch? How much time does it take on average for them? Despite using effective caching strategies and determining and building only affected projects, with the google's scale that we are talking about, it's still going to take hell lot of time for a build when a project that's being used in multiple places is changed. What are some strategies they use to reduce build times at Google?

https://redd.it/1jadq6s
@r_devops

Jobnik: Open Source K8S jobs managing tool

Hello good folks! So happy to share with you a tool I developed working at Wix that will allow you an easy, Rest API based interface to trigger and monitor your Kubernetes Jobs.

The tool was designed for offloading long lasting processes from our microservices and allowed a cleaner and more focused business logic.

Suggestions, bugs and contributions are more than welcome!

https://github.com/wix-incubator/jobnik

https://redd.it/1jafzf8
@r_devops

CI/CD with TypeScript Instead of YAML (Open-source)

I've always struggled with the various declarative syntaxes other CI/CD platforms use, especially when I just want to focus on shipping my projects.

The goal of PandaCI is enable you to code advanced workflows with little more than a quick example. I've found that by just having a few functions (job, exec, etc), everything else can be done natively in the language. A few such examples are:

* Conditional jobs — Use standard if statements
* Matrix jobs — Just write a for loop
* Code reuse — Create functions or import shared code
* Parallel steps — Use Promise.all
* API integrations — Native fetch or import npm packages

I'd love to hear any feedback (harsh or otherwise)! Personally, It's been a big upgrade for my projects and I'm excited to see what the community makes of it.

🔗 Repo: [https://github.com/pandaci-com/pandaci](https://github.com/pandaci-com/pandaci)

🔗 Site: [https://pandaci.com](https://pandaci.com)

P.S. The core is language agnostic so let me know if there are any other languages you'd like to see

https://redd.it/1jagd5v
@r_devops

AWS Certificate Free Vouchers valid until August 2025


AWS is offering 100% free certification vouchers for select exams, valid until August 2025!

This is a great opportunity to expand your cloud expertise and earn industry-recognized certifications—at zero cost.

Eligible Certifications:

✅ Foundational: Cloud Practitioner, AI Practitioner

✅ Associate: Solutions Architect, SysOps Administrator, Developer, Data Engineer, Machine Learning Engineer

https://community.aws/content/2tm12rQPFomu2bKOP1rIWWtsAAx/opportunity-to-earn-free-aws-certification-vouchers

https://redd.it/1jadmd2
@r_devops