Thoughts on Aiven vs cloud vendor services?

Aiven offers managed versions of open-source tools (Postgres, MySQL, Kafka, Redis, ClickHouse, etc.) across multiple clouds. They pitch it as avoiding vendor lock-in while still getting fully managed infra. Anyone here using it in production? Worth it vs the native AWS/GCP/Azure databases?

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

Curious how's the market right now in North America.

Who's looking for a DevOps position? How do you feel the market is right now? About remote, seems like companies want back to office even more.

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

Are DevOps services actually worth it for small teams?

Hey all, not usually in this sub (I’m a product person, not ops), but I figured you’d know best.

Our small SaaS team is drowning in infra... suff like broken deploys, weird billing spikes and no one who really wants to own “ops.” I keep seeing DevOps services advertised as a way to fix this. They claim they’ll handle pipelines, monitoring, scaling, etc. so your devs can stay focused on product.

On paper it sounds amazing, but I’ve never talked to anyone who’s actually used them... guide me please

https://redd.it/1nwiu55
@r_devops

Dev team & operations team but no devops team.

My company are in the process of replacing all of our saas with in-house apps.

I work in the operations team and have been operating as a sort of translator between the devs and the rest of IT

I’d like to move into devops and I’m wondering the best way to position myself to do this given the opportunity.

We operate exclusively in azure.

I’m not sure any of the work iv done so far is what you would call real devops work, things like setting up SSO, recommending we setup defender for cloud so the security team has visibility into any vulnerabilities inside the code, configuring service principals for the applications to access different parts of our environment, iv recommended moving to azure devops and want to moving into more devops related work, so my question is, what can I do at this point to provide value and maybe gain some experience with working in devops?


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

Need a devops partner

Hey I am a beginner in devops i need a partner to study devops u should be active atleast 1hr a day we can learn and grow together if anyone interested dm me

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

Team culture, whinging

I’m in a team that has a culture of whinging, mostly other parts of the business being incompetent (which aren’t actually too bad and pay the bills), also external parties, but also other team members’ work, when those team member aren’t present. Additionally, a focus on technical aspects as opposed to business outcomes.

Have you ever seen such culture turn around and how?

https://redd.it/1nwz0uq
@r_devops

What I learned from hiring a software development partner for our startup

Our startup recently partnered with a custom software development team to build a scalable app. At first, I wasn’t sure how to pick the right vendor there are so many options out there.

What really made a difference was working with a team that communicated clearly, had proven experience across industries, and remained flexible as our requirements changed during development.

The structured approach saved us a lot of time and headaches, and it was a relief to see the project move smoothly. I’d love to hear how others choose development partners for their projects.

https://redd.it/1nwysm5
@r_devops

Git CI/CD Integration Testing

I’d like to get some opinions and advice on how to set up the basic structure of a test pipeline and repository structure in gitlab.

At my company, we’re starting a new project that integrates multiple components. Some of these components already exist and just provide Docker images. But several other components are being developed from scratch specifically for this project. My task is to write a test pipeline that brings all of these components together and runs tests.

My initial idea was to create a separate repository for each new component so we can version them properly. Then, have one dedicated repository for integration, which would only be responsible for deploying the different component images (for example, via Kubernetes) and running integration tests.

However, a colleague who has been with the company for many years suggested a different approach: a single project repository, with each component in its own folder, and one big pipeline that builds everything from source, runs unit tests and coverage checks for each component, and then also runs the integration tests.

Personally, I think it makes much more sense to separate the components. The downside I see, though, is that some components might need dependencies from others just to test themselves properly.

So my questions are:

What’s considered best practice here?

How do you usually structure something like this in a clean and maintainable way?

What are the pros and cons of each approach?


I’m open to hearing different strategies and experiences.

https://redd.it/1nx0zly
@r_devops

How to deal with a coworker who is almost never available and is un-fireable?

In our department, we essentially handle the entire stack from native app development to deploying our product into the cloud. I work with 3 platform engineers with the infrastructure architecture and deployment as well. One of the senior guys who’s the most knowledgeable one is barely ever around and does not do his portion of the work to get new features in for qa to test and deploy into a couple of the staging environments. So I and another engineer have to pick up his slack and get it done before the next release deadline.

That senior engineer in question is the son of the CTO of the company. So telling management about him goes nowhere. We’ve tried. I know we should leave, but job market seems pretty bad even for seniors. With that being said, I still love working here. I’m just trying to get some advice on what to do here with him in particular.

https://redd.it/1nx39k6
@r_devops

Octopus Deploy Pricing & Use Cases.. Feedback…

For those of you running Octopus Deploy day-to-day in the enterprise.. How are you finding it? Specifically:

Are you finding the value in audit trails, approvals, and environment management worth the premium?

If you’re using it for Kubernetes or multi-cloud, how does it compare to alternatives like ArgoCD or Flux… Would love to hear from other teams (especially mid-sized orgs or regulated industries) on how you’re using it and what’s been working.

https://redd.it/1nx2bv5
@r_devops

Need help setting up Clickhouse DC DR Setup

# What I already have

* Two Kubernetes clusters: **DC** and **DR**.
* Each cluster runs ClickHouse via the **Altinity Operator** using `ClickHouseInstallation` (CHI). Example names: `prod-dc` and `prod-dr`.
* Each cluster currently runs its own **ClickHouse Keeper** ensemble (StatefulSet + Service): e.g. `chk-clickhouse-keeper-dc` in DC and `chk-clickhouse-keeper-dr` in DR.
* ClickHouse server pods in DC point to the DC keeper; ClickHouse pods in DR point to the DR keeper.
* Networking: there is *flat networking* between clusters and FQDNs resolve (e.g. `pod.clickhouse.svc.cluster.local`), DNS resolution has been verified.

Tables use `ReplicatedMergeTree` engine with the usual ZooKeeper/keeper paths, e.g.:

CREATE TABLE db.table_local (
id UInt64,
ts DateTime,
...
) ENGINE = ReplicatedMergeTree('/clickhouse/tables/{shard}/table', '{replica}')
PARTITION BY toYYYYMM(ts)
ORDER BY (id);

# My goal / Question

I want **real-time replication** of data between DC and DR — i.e., writes in DC should be replicated to DR replicas with minimal replication lag and without manual sync steps. How can I achieve this with Altinity Operator + ClickHouse Keeper? Specifically:

* If separate keepers are kept in each cluster, how do I make `ReplicatedMergeTree` replicas in both clusters use the same replication / coordination store?
* Any recommended Altinity CHI config patterns, DNS / service setups, or example CRDs for a DC–DR setup that others use in production?

Any help is really appreciated. Thanking in advance.

https://redd.it/1nx4ssh
@r_devops

devops/SRE tasks with MCP server?

hey folks, I am a maintainer at SigNoz. We recently open sourced our MCP server ( https://github.com/signoz/signoz-mcp-server )

We have got some community members using it but wanted to get feedback from the community here also on what type of devops/SRE tasks do you expect to do with your MCP server of o11y tools? what are you doing currently

We have basic tools like getting metrics, logs, dashboards, alerts etc. Looking for feedback from the community before deciding on what tools to build next.

Also, if anyone uses AI SRE tools like Resolve AI, Traversal - would love to learn what you use it for? What things you expect to do using MCP servers and what you would do with a more full fledged AI SRE product?

https://redd.it/1nx36sa
@r_devops

Vpc and Networking

Practicing devops projects we can do it but I have a doubt or confusion. Suppose I need to setup ips and network how should I do i have seen many videos but i dont understand this concept of subnets /32 /16 and ip hashing and how can i allocate custome network for a projects and vpcs any resources.W

Tl:Dr
I need resources to learn about cloud and networking vpc, subnets from scratch

https://redd.it/1nx0xms
@r_devops

Dipping my toes in to DevOps/DevSecOps

Hey there everyone!

A few months ago I started my journey in IT.

I got a job as a SOC Analyst/System Engineer in Microsoft 365 environments.

It's been pretty great and I've been learning a lot but I'm starting to want to deepen my understanding of the full IT landscape.

My company deals with a lot of DevOps related stuff as well and out of curiosity I asked to be put inside a huge Cloud Migration project involving Azure and to be honest it's been kind of hard following what everyone is saying inside these meetings.

Nobody (rightfully so) will take time out of their day to explain to me what everything is and I'm trying to do my best to understand what is going on.

I've learned a few things and concepts like what a Gantt diagram is or what "lift & shift" means but I'm still having a hard time in understanding the full picture.

I'd appreciate if anyone could link some resources so that I can begin getting into this world.


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

AWS WAF rules visualizer

Hey there,

Has anyone else noticed that the AWS WAF visual editor just stops working once your rules get a bit complex ( nested statements / 5 or more statements) ? You get stuck in JSON view with the “cannot switch to visual editor” error, which makes it painful to understand or explain what’s going on.

I've built WAFViz to help with this, add your JSON and verify the diagram

You could also share the config with others

https://wafviz.ardd.cloud

Feedback is appreciated!

https://redd.it/1nxdtb5
@r_devops

Setting up VPN vs Zero Trust Network Access (ZTNA)

I have built the architecture of Pritunl VPN for our IoT devices and works great. Love Pritunl VPN where it is more manageable and cheaper compared to other vendors. Now when it comes to accessing our Gitlab server to other hosted services, my CTO has tasked me into utilizing ZTNA rather than VPN. First thing that pops in my mind is Twingate but would setting up ZTNA be the right decision?

I have looked into Pritunl Zero and looks promising but would like to get your opinions on this methodology. I'm used to just setting up OpenVPN and giving developers a profile to access into any server in a private IP.

Thanks for reading my post.

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

" People don’t quit because of bad products they quit because of bad loading times "

I used to think my SaaS had a " value prop problem " Visitors weren’t converting, So I blamed pricing, copy, features - all the usual suspects. I tweaked everything and Nothing worked.

Then I ran a speed test. Turns out my landing page was taking nearly 5 seconds on mobile. Literally FIVE. SECONDS. That’s an eternity online. Basically, People weren’t leaving because they hated the product - they were leaving because they never even got to see it.

After fixing the basics ( Images, Scripts, Caching ), Bounce Rates dropped instantly. Signups actually started climbing without me changing a single word of copy.

It was one of those " Holy Crap " moments that completely reframed how I look at growth. Sometimes the biggest barrier isn’t price or features. It’s just the fact that people don’t want to wait.

Makes me wonder how many companies are wasting money on ads, design, or funnels, when the real problem is just that their page takes too long to load.

What do you think do businesses underestimate performance, Or do they just ignore it because it feels " too technical " ?

https://redd.it/1nxjsv3
@r_devops

" I thought my SaaS problem was pricing and copy turned out it was just speed. "

For months, I kept blaming my low trial signups on all the " Usual Suspects " - Pricing, CTA buttons, Copywriting. I even redesigned the landing page twice. Nothing worked. Out of frustration, I ran a speed test. And wow my homepage was taking nearly 5 seconds to load on mobile. That was the real issue. People weren’t rejecting the product. They weren’t even waiting long enough to see it.

Here’s what happened after I fixed performance -

Bounce rate dropped almost immediately.

Signups started climbing, with no other changes.

Google rewarded the faster site better ad scores and cheaper clicks.

SEO improved because of Core Web Vitals.

It was one of those " duh " moments. We spend so much time chasing growth hacks - new funnels, viral tricks, A/B testing CTAs. But sometimes the simplest fix moves the needle more than all of that.

That’s actually why I started working on Website Speedy. It basically strips away the junk that slows down sites - render-blocking noscripts, bloated code, uncompressed assets and makes pages load fast, even on older devices. It saved me a lot of trial-and-error pain. I’m not here to hard-sell it (mods hate that, I get it). But honestly, fixing speed was the single best " Growth Hack ". I stumbled into.

Curious - How do you all think about performance? Do you treat speed as part of growth/marketing, or Do you leave it buried in the dev backlog until customers start complaining?

https://redd.it/1nxmcc8
@r_devops

Final Year Project on Cloud & DevOps - Need a real-world problem to solve

Hey everyone, I’m a CS student heading into my final year and I want my project to be more than just something for grades. My focus is on **Cloud & DevOps** (AWS, Kubernetes, CI/CD, monitoring, automation), and I’ve got a whole year to dedicate.

I don’t want a toy demo - I want to build something that:

* Solves a **real daily-life problem**.
* Runs on a **scalable, cloud-native setup**.
* Can be a **solid portfolio piece** to prove I can design, build, and deploy end-to-end.

I have some directions in mind, but I’d really value outside perspective.
If you were in my place, what **everyday problem** would you try solving with tech?

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

Platform Engineer Intern. Is ansible worth learning?

I will be having an interview somewhere next week for a platform engineer internship role. The technologies that will be touched on include VMs, Python, bash, and Ansible.

I have always been wanting to break into devops and have studied many of the different technologies required in Kodekloud(k8, docker, CICD etc)

Have seen a lot of comments where people say Ansible is not used often because of K8 and containerization etc. So just wondering, will this internship still be useful if i want to pursue a career in devops?

https://redd.it/1nxqx8u
@r_devops

Migrating Domains from AWS Route 53 to GCP DNS (with SSL) – Step by Step Guide

Hey everyone,

I recently wrote a step-by-step walkthrough on how I migrated domains from AWS Route 53 to Google Cloud DNS, and also set up SSL along the way. I tried to make it practical, with screenshots and explanations, so that anyone attempting the same can follow along without much hassle.

If you’re interested in cloud infra, DNS management, or just want a quick guide for moving domains between AWS and GCP, I’d really appreciate it if you could give it a read and share your thoughts/feedback:

Read here: Migrating Domains from AWS Route 53 to GCP DNS (Step-by-Step with SSL Setup)

Would love to hear if you’ve done something similar, and if there are optimizations or gotchas I might have missed!

https://redd.it/1nxr8bx
@r_devops