What do people use for quota management?

As the title says. I have various accounts on AWS, GCP or other tools where I need to adhere to quotas (hourly/weekly/monthly) for certain services (e.g can’t get over X messages in AWS SNS).

I'm wondering how I can monitor/alert on these before it will have an impact on my pricing.

Thanks for any pointers!

https://redd.it/13apl9h
@r_devops

Quick poll for folks using centralized logging: Do you use encryption to ingest your logs?

I'm looking into setting up proper centralized logging and found that most tutorials either don't mention encrypted (TLS/SSL) ingestion at all or just touch on it barely, so I wanted to see how the real world handles this.

Note: I am not asking whether the web interface is using HTTPs, I am asking about the way the logs are being sent.

View Poll

https://redd.it/13aw20d
@r_devops

Supercharge Your Cloud Productivity with CloudJin - Convert Infrastructure Diagrams to Terraform!


Hey fellow engineers in the DevOps community! We're excited to introduce CloudJin, an AI-powered cloud infrastructure automation tool designed to skyrocket your productivity by seamlessly converting infrastructure diagrams into Terraform code. We're looking for prospective users who'd like to join our waitlist, offer their valuable input and help us build the product alongside.

What sets CloudJin apart?

Experience a massive productivity boost by automating the conversion of infrastructure diagrams to code.
Gain security visibility and stay on top of potential threats by ensuring consistency across deployments.
Improve agility in your cloud management and adapt quickly to changing business requirements.
Streamline collaboration with a visually intuitive approach, empowering teams to work more efficiently.

We'd be thrilled to have you join our waitlist and be among the first to try CloudJin. Your insights and suggestions will play a crucial role in shaping the future of our platform. Check us out at https://www.cloudjin.ai and let's revolutionize the way we manage cloud infrastructure together!


Join our waitlist now, https://www.cloudjin.ai/#join-the-waitlist !

https://redd.it/13axnt7
@r_devops

Learning about Authentication and Authorization - courses/certifications?

Hi,

Do you know any online course or a certification which would cover, in detail, authentication and authorisation protocols? (the main reason why I'm asking about a cert is because they tend to have a well thought-through learning curriculum, like e.g. RHCSA)

In particular, I would like to learn, in a nice, guided fashion, about Kerberos, LDAP (inc Active Directory), OAuth 2.0, RADIUS and JWT.

I know what they are, I have come across them at my various jobs, I have even fiddled with some settings on an ad-hoc basis, but it has always been incidental to my work - it would be nice to have some in-depth coverage (ideally with demos, examples, best practices etc.).

Sure, I can read through the docs, but that's usually not the most engaging thing to read on Saturday :-).

Cheers!

https://redd.it/13b09hh
@r_devops

Dev self service portal

Does your place have a self service portal where developers can self serve deploy resources in the cloud? How do you do self service kubernetes? I’ve seen a couple options such as backstage and Ops level. What is your solution and what features does it provide. What does the developer experience look like (UI, API, gitops flows)?

https://redd.it/13b966p
@r_devops

Is it ok to use the replication parameter with bitnami postgres?

const sequelize = new Sequelize('database', null, null, {
dialect: 'mysql',
port: 3306,
replication: {
read:
{ host: '8.8.8.8', username: 'read-1-username', password: process.env.READ_DB_1_PW },
{ host: '9.9.9.9', username: 'read-2-username', password: process.env.READ_DB_2_PW }
,
write: { host: '1.1.1.1', username: 'write-username', password: process.env.WRITEDBPW }
},
pool: { // If you want to override the options used for the read/write pool you can do so here
max: 20,
idle: 30000
},
})

https://sequelize.org/docs/v6/other-topics/read-replication/

The doc says that

I've read read replica are switched using round-robin, I am guessing it means that the load is distributed using round-robin, but what happens when the write server goes down, does one of the read replica becomes a write db? Secondly, does bitnami postgres helm chart works well with this read replica parameter, or you need to use another library in order to make it work? I am thinking the write db if it shuts down the whole application just crashes on the write side and that's also what bitnami assumes.

https://redd.it/13b9rvc
@r_devops

No operation guys in Devops anymore.

Have you guys noticed that operations/ app engineering part of DevOps is dissipating? Soon, there will be only developers.

https://redd.it/13bfefu
@r_devops

Argo Ecosystem: Argo CD, Argo Workflows, Argo Events, Argo Rollouts, Argo Everything

I feel like not everyone realizes, that Argo is a whole ecosystem today, with different tools covering different use cases, but all working together in a very nice combo. I made a short video explaining Argo ecosystem: https://youtu.be/ae0XJGJPLvo

https://redd.it/13bgzdj
@r_devops

When am I ready for a junior role?

I have been spending most of my free time over the last few months studying DevOps related things so that I can leave my Help Desk job. I have been using the roadmap(https://roadmap.sh/devops) to help guide my studying and now I'm wondering how far down the list I should get before I should think about applying? I know mastering all of it is ideal but in your experience, how much of this do you expect a junior devops candidate to know or have experience with?

https://redd.it/13bgso9
@r_devops

OpenTelemetry Tracing - What are the options?

So OTel is all over the place, but many developers are still paying tons to their APM and they lean on logs and metrics w/o using tracing. But in microservices, you must understand the full context... services, APIs, dependencies, etc. Otherwise you are in a troubleshooting nightmare.

The thing is that OTel is "just" to instrument the data, you need other tools to create visualization and insights.

So Helios created this article that shares everything including the tools that can be used and examples on how to use them :

https://gethelios.dev/blog/opentelemetry-tracing/

https://redd.it/13bjamr
@r_devops

A resilient Redis cluster on kubernetes

I would like to share my experience of deploying a Redis cluster on Kubernetes with different Helm charts. While Bitnami is widely used for Redis clusters, I had doubts about its ability to handle production workloads following a chaos engineering test. After thorough research, I found an extremely robust Helm chart that operates without any problems. In this post, I will introduce this new Helm chart and detail the reasons why it's the superior choice for production deployments.

https://medium.com/@mallakimahdi/most-resilient-redis-cluster-helm-chart-e04632ec7403

https://redd.it/13bl1w2
@r_devops

Whenever i applied the scaledobject file of keda of rabbitmq, the pods start creating and immediately terminated

I am working on minikube cluster of kubernetes. I have deployed prometheus using kube prometheus stack helm chart and deployed rabbitmq using bitnami helm chart, than i have deployed rabbitmq exporter and keda using helm charts.

Issue: Whenever i applied the scaledobject file of keda of rabbitmq, the pods start creating and immediately terminated.

What I want to achieve: I want to do hpa of rabbitmq using keda in minikube cluster and GKE

https://redd.it/13blgel
@r_devops

eliminating saas bills

Hey all, I'm getting sick of paying SaaS bills. Whenever we go with a vendor, it always feel a bit like signing a deal with the devil. The feature set is never quite as promised, the pricing is difficult to predict, we end up getting locked-in to that vendor as our codebases start reflecting that vendor's way of doing things specifically, and the load to switch to someone else becomes too high.

I'd much rather go with open-source solutions for most infrastructure/devops things but that's always met with:

1. We need to put in work to architect, deploy, and integrate the OSS solutions

2. We need to put in work to continuously update, upgrade, and support the OSS solutions

3. "How do we guarantee it's secure?"

To me, the first is a one-time cost we can easily put the work in to figure out. The second becomes negligible with the proper lifecycle automation (i.e. tie your infrastructure deployments to the rest of your deployment scheme). The third is silly, if you can secure your code, you can secure someone else's code, and popular open-source projects have many eyes looking at the security of the code. Not to mention...using SaaS for cicd/observability/identity/etc means trusting a third-party with ALL your proprietary code/data, how is that MORE secure?

I feel like I could easily put together a package of open-source solutions that deploys easily on k8s and solves 90% of the above issues. The last 10%, primarily involving bug-fixing/being a subject-matter expert when one of the tools inevitably breaks, could be reduced down to one or two people, or maybe even out-sourced.

Has anyone found success in replacing ALL their SaaS vendors with open-source solutions and fully self-hosting/supporting?

https://redd.it/13bjyv7
@r_devops

help with my job title.

How can you name a specialist DevOps/cloud architect?
I know this is a silly question, but at this point i don't know who am I anymore. And my colleagues don't know too.

I was hired as DevOps, but now I am in charge of all AWS infrastructure too. I am doing everything: CI/CD, runners, K8s, serverless apps together with trams, governance in AWS, monitoring. You name it.
I am doing AWS Cloud solution architect cert now.

https://redd.it/13booz5
@r_devops

Syllabus for learning DevOps coming from network eng. position

For experienced DevOps professionals, if you were to write a syllabus for an experienced network engineer who wanted to move into DevOps, what would the subject matter outline look like?

https://redd.it/13bl5xa
@r_devops

How important are CSP certificates in today’s job market? (Q2, 2023)

Just for context;
BEng in Computer Engineering, MSc in Digital Networking, 8YOE in total, with 5 as a backend SWE.
Worked at startups most of my career, that never seemed to have DevOps engineers. So backend engineers typically handled the servers and infrastructure as part of our role.

I’m hoping to move purely into Cloud DevOps, and I’m wondering how important these certifications are to land a job.

My question: If I have 4hrs a day for the next 6 weeks, would you advise me to complete the popular DevOps roadmap?
Or study and try for at least 1 AWS cert?

(I can’t find the link to the roadmap, but you probably know which one I’m on about… it’s been shared 50000 times)

I honestly don’t have more time than this right now, with 2 little kiddos (one infant 🥲). I’m on a time crunch to get a job as one income is no longer sustainable with the new addition, just looking for some guidance so I don’t run in circles forever.

I am currently not working, took a year off for medical reasons.
Advice greatly appreciated from anyone and everyone 🙏


Edit: The cert I would attempt to take would be Developer Associate, since it’s closest to what I’ve been doing.

https://redd.it/13bw587
@r_devops

Pulumi Self-Managed vs Pulumi Cloud

I’m looking at migrating to self managed using s3 to store state and I’m finding that there aren’t a lot of resources or best practices. I was hoping to get some peoples experience in the types of challenges they have faced managing their own Pulumi backend, and the types of solutions you may have found useful.

For example, some big considerations seem to be concurrency, state locking, security, saving additional data such as history, backups.

Also, more of a specific technical question here but, when migrating code I’ve noticed that I need to use aws.lambda.Function and aws.lambda.CallbackFunction creates serialization errors. It only does this with s3 and not in Pulumi cloud. If anyone knows more details on how the backend choice affects serialization, that would be a huge help.

Thanks for any experiences/opinions!

https://redd.it/13byf49
@r_devops

How do you keep timezone data up-to-date?

Sometimes business requirements require converting from UTC to a specific timezone in the backend side for whatever reason.

Typically you would install tzdata in the Dockerfile, which has all the timezone information you need.

However sometimes timezone data changes (e.g. some country drops DST), causing potential issues in your code.

The fix would be to rebuild your containers, but this is a reactive solution. I’m looking for a proactive solution, other than constantly rebuilding your images. Preferably without even restarting your Pods (we’re assuming Kubernetes here)

So far I came across k8tz, which could be a solution.

How would you solve this issue?

https://redd.it/13bkudj
@r_devops

Help me get a job in cloud!?

I'm learning linux and wanna switch from non it job into IT cloud is what I think I want to transition into, I watched youtube series on AWS and I think this is what I want to get my hands on!
Somehow I started linux took a course on udemy and started learning taking baby steps like everyday I learn couple of commands and soon I'll be done with the basics part but then I wanna learn AWS could you guys help me out how do I start and any kind of suggestions which will help me get a job into cloud?
I started with linux then will start AWS is it enough to get a job in cloud or do I need to learn more tools?
Thank you!

https://redd.it/13bxbbx
@r_devops

A resilient Redis cluster on kubernetes

I would like to share my experience of deploying a Redis cluster on Kubernetes with different Helm charts. While Bitnami is widely used for Redis clusters, I had doubts about its ability to handle production workloads following a chaos engineering test. After thorough research, I found an extremely robust Helm chart that operates without any problems. In this post, I will introduce this new Helm chart and detail the reasons why it's the superior choice for production deployments.

https://medium.com/@mallakimahdi/most-resilient-redis-cluster-helm-chart-e04632ec7403

https://redd.it/13bkn48
@r_devops

Have you heard of DevOps, DataOps, DevTestOps, or AIOps?

These are just a few examples of the "Ops" that we encounter in our routine lives. As technology continues to evolve at a rapid pace, businesses are constantly exploring new ways to streamline their operations and optimize their workflows.

So what are your thoughts on the world of "Ops"? Have you encountered any other "Ops" that you find particularly interesting or valuable?


We'd love to hear your thoughts in the comments below! 💬

https://redd.it/13bfozl
@r_devops