Central config repository for ansible, terraform and kubernetes

In order to avoid having to manage the same credentials, host variables etc. in multiple places, I'm currently running everything through ansible and use jinja2 templates for the terraform and kubernetes files, but this makes working with the individual tools a nightmare.

Is there a simpler solution to share variables and secrets between these different tools?

I know that something like Vault could provide with somewhat of a solution, but the overhead and complexity makes me want to see if there isn't a simpler solution out there.

How do you all manage this?

https://redd.it/z885cn
@r_devops

LoadTesting as Code

Hello,

What are you all using to define web (HTTP/1.1, 2, 3, gRPC) load tests?

Stuff that can live in version control, is diff-able

(Bonus points if it can load tear arbitrary protocols)

https://redd.it/z8ak15
@r_devops

How would you set up your work laptop differently if you had to do it again?

After much saving and researching I finally bought a MacBook Pro M2 chip coming in today. I’m starting a devops boot camp in January (I know, I know, devops isn’t a position, it is an environment, process, or whatever you want to call it) but right now I’m learning Python, web development and taking various courses on code academy like learning the command line. I don’t know much else, but I’ve read the phoenix project and to be as specific as I can, I’m aiming towards working in a position similar to Bill. Understandably, after being part of this sub for a while, I know I will end up in a position completely different especially at first.

I know this might seem like a minuscule thing to ask, but do you have any tips or advice on how to set it up and how to use this new lap top after unboxing for someone who has only used laptops for Microsoft office and Netflix before? (What default settings to apply, programs to delete or download, habits to get into, etc.) Being a considerable investment, I’d like it to be optimized and dedicated to the max specifically for what I am going to be using it for right from the get go.

https://redd.it/z8ao7d
@r_devops

New DevOps please learn networking

I know the current meta is f college and lets become DevOps engineer after watching few YouTube videos… but please add some networking videos to your playlist… I interviewed more than 20 “DevOps” engineer in the last few weeks and the lack of basic networking knowledge is nuts…

https://redd.it/z8fanh
@r_devops

S3 to sns when files uploaded

AWS SNS - post message Using a URL

I am trying to figure out how to send a message on to a SNS topic on clicking a url ?

How do I go about creating the url and how do I associate with the SNS topic ? My files are html integration testing results

https://redd.it/z8j43e
@r_devops

What are the main differences between OpenSearch, CloudSearch, and Kendra

All of them are used to build search engine from your data. I could see Kendra uses machine learning, while OpenSearch and CloudSearch does not. What exactly are the main differences between the three?

https://redd.it/z8elqh
@r_devops

Is KodeKloud's annual membership worth it?

Thinking of whether to subscribe. I have passed all Kubernetes cert

https://redd.it/z8m1ij
@r_devops

AWS CloudHSM

Has anybody tried to do code signing with AWS CloudHSM recently?
It seems borderline impossible to use. The documentation is poor at best. At worst it's just plain wrong.

It seems that they've tried to rewrite the CLI utility but it's just half finished but they've already released it.

The funniest part... The user management utility isn't even codesigned!

https://redd.it/z8940z
@r_devops

Need help using cpanel certificate on nginx with digital ocean

So I have a cpanel account with a wildcard domain I have been using for a while, I created a front-end app with would be on the cpanel, but my backend is hosted on digital ocean( I have a paid plan from a company and I would have to pay to build node on my cpanel).

Is there a way to use the wildcard certificate or install it on my digital ocean server

Say my domain is yes.com, since it's wildcard, can I have something like backend.yes.com on digital ocean?? While using same certificates?

Do I need to change the nameservers, need help with this, thanks.

https://redd.it/z8nxhd
@r_devops

What do you guys think about WASM + Docker?

The tech is still in preview, but might that just be the future of containerized apps?

https://redd.it/z8ojev
@r_devops

Terraform and Kubernetes deployment samples

Hello all,

i wanted to share my previously deployed templates for Terraform and Kubernetes. It should be easy to understand at all level developers with a little knowledge in the field.


https://github.com/tanerjn/terraform-deployment


https://github.com/tanerjn/kube-deployment


Feel free to star or fork to check it.

https://redd.it/z8ubn4
@r_devops

Build AWS AMI on GP3 with Packer 1.8.4?

Hello....I'm trying to figure out where the Packer experts hang out. I ask questions on the Hashicorp forums and they go unanswered, seems like there's little activitty there in general.

I'm trying to build AWS AMIs using GP3. GP2 works fine, but I can't get the syntax right for GP3.

This is what I have:



"builders": {
"type": "amazon-ebs",
"launch_block_device_mappings": [{
"volume_type": "gp3",
"volume_size": "3",
"device_name": "/dev/nvme0n1p2" },

But I keep getting an error that /dev/nvme0n1p2 is an invalid device name. /dev/nvme0n1p2 is the device root is being built on. I have also tried /dev/nvme0n1, /dev/nvme0, and even /dev/sda all with the same error.

I've also read that as of 1.7, GP3 is supposed to be the default volume type, but I'm still getting GP2 when I look at the AMI Storage tab.

Any help appreciated.

https://redd.it/z8ujdr
@r_devops

Is hosting Sentry ourselves worth it?

As our usage is growing, we have to pick whether to byte an expensive bill (over 3k/month), use sampling a lot, or self-host. The primary cost driver is the use of transactions.

What has been your experience self-hosting Sentry?

https://redd.it/z8v967
@r_devops

Seeking recommendations for code coverage and documentation presentation from Github actions.

Hello everyone,

**tl;dr** What's the best solution for maintaining unit test coverage HTML output and auto generated from code HTML docs for every open branch in a repository that's triggered by GitHub actions or a similarly automated update?

I'm looking to create a CI/CD pipeline from my teams growing set of GitHub repositories and I'm looking to start off with suggestions for how to present code coverage from unit tests as well as automatically generated documentation from code comments.

Firstly I would like to say I've seen codecov and I really like their offering but their $10/person/month price tag is just not going to work for us and it's a private repository. I am perfectly willing to self-host this solution and would love to hear if there are free self hosted packages that do what I need.

I am hoping to achieve the following goals with my CI:

* HTML code coverage generated by the project is available linked to from the README.md or PR comment in a way that is specific to the current branch. This could be also done by annotating code views in GitHub if necessary but I would like this to be specific to each branch rather than having one HTML coverage artifact for the whole project which would need to be like the main branch or something
* Documentation generated from the build is placed somewhere that's easily linked to from the README.md
* Code coverage and documentation HTML is removed when a branch is deleted so that every live branch has artifacts but storage isn't building up forever.

There are some solutions I've considered that I don't love:

1) Auto generate documentation and HTML code coverage in a GitHub action and post it to the the GitHub pages for the repo. This has the disadvantage of there being only one coverage and documentation artifact for the entire repo, not one per active branch. But it's pretty easily done.

2) Check in the code coverage and documentation into each branch and view the HTML by navigating to the code. Something about this just weirds me out from a developer perspective. I don't love build artifacts cluttering the repo and we don't need to maintain historic changes for the HTML coverage report. Also I cringe at the thought of this causing merge conflicts that might have to be dealt with.

3) Build my own web host that has an API where in GitHub actions can upload the build results of code coverage and documentation to the server I create and provide a few simple links for users to go look at it. While this is exactly what I want, and also not exactly a very difficult project, I feel like something like this probably exists and I don't want to maintain this in the future if there's a good solution with a community.

https://redd.it/z91y9q
@r_devops

Containerizing devops tools with docker compose

Recently I've been thinking about containerizing our devops tools in my current work place. The dev and devops teams currently have everything installed locally to do deployments. We use terraform, terragrunt, ansible, packer, kube etc.

I have been toying with docker compose and feel it would be a good fit for us as we can control the versions of the tools at a repo level. I can get individual docker compose services running fine e.g. just terraform. When I introduce multiple services with the same docker compose file, things start to fall apart.

We use terraform and terragrunt together, packer and ansible so having the 2 related tools within the same docker compose file makes sense.

Does anyone have experience of this type of situation or would suggest any material?

https://redd.it/z90nv3
@r_devops

Access variables in Jenkins trigger section

I am trying to use a shared library pipeline (if that's the correct way to put it) from two different Jenkins files. One to do nightly builds and one for deliveries. This may not be the best setup (i'm just learning) but I am having trouble accessing a ?parameter? passed in. I want the nightly build to get scheduled for a certain time and the delivery to only be run if started manually. The rough outline is:

shared lib

**pipe.groovy**

def call(body)

triggers{

cron(env.target == 'nightly' ? '00 8 * * *' : '') // env.target is null?

....

stages{

stage{

steps{

sh('printenv') // env.target is set



**Jenkins.nightly**

pipe{

target = 'nightly'
...


**Jenkins.delivery**

pipe{

target = 'delivery'
...

I seem to be able to access the ?build in? env vars in the trigger section but not this way. I've tried many combinations of trying to do so.

Any suggestions?

https://redd.it/z96891
@r_devops

SWE joined devops, no coding for months

Promised the world, found OPs instead.
Smells like OPs, is OPs and I want out.

Afraid to have an only 6 months Job in CV.

Recruiters only propose Devops roles already but I was supposed to learn and I am not right now. Essentially I’ve given up SWE for devops and there’s no dev in it.

Made my mind to return to SWE but see above.

Anyone with same experiences?

How did you get out?

https://redd.it/z98kgx
@r_devops

Git Portfolio: GitHub or Gitlab?

I want to use GitHub or Gitlab for a personal portfolio. I only want to upload POCs or Demos.
I migrated from Github to Gitlab 3 years ago. However, In the last time, in some jobs I had, we migrated from Gitlab to Github because of the new pricing.
Nowadays, I believe maybe Github it's better for a personal Portfolio. I don't care about Github/Gitlab CI; I can adapt to that part.
So, GitHub vs. Gitlab? What it's better for a personal portfolio?

https://redd.it/z98lfq
@r_devops

Tools for determine when a PR is safe to automatically merge

I swear I’ve seen a post about this recently, but can’t find it for the life of me.

I’m not asking about running security scans on code, rather Im asking for a tool that allows me to define safety criteria (i.e., file X wasn’t modified, dir Y was modified, etc) and automerge a PR if that criteria is met.

My current solution using bash is becoming a bit annoying to manage.

Any suggestions welcome.

https://redd.it/z9bwjz
@r_devops

How do you find the lambda function from the HCL config file?

I see something like:

resource "aws_labda_function" "lambda01" {
filename = "lambda.zip"
function_name = "lambda01function"
}

I didn't find anything named [lambda.zip](https://lambda.zip), lambda01 or lambda01function elsewhere across the whole codebase. How do you find the lambda function that the HCL config refers to? HCL is used by Hashicorp Nomad.

https://redd.it/z9dy7o
@r_devops

Ansible:

Trying to update values in the below json file.

{
“Host”: “ttrg97”,
“Vm “: “test”
}


When I update any value , the square brackets are missing.

{
“Host”: “updatedvalue ”,
“Vm “: “test”
}
I’m new to Ansible can someone help me how to update this file

https://redd.it/z9775i
@r_devops