DevOps and Change Mgmt

Our organization is slowly but surely moving into a DevOps world. In the existing ITIL base service delivery world we have an established and mature (bureaucratic) Change Mgmt process.

We are facing more and more changes from the DevOps side especially more Infrastructure as Code.

How best to handle change from the DevOps side without introducing antipatterns and such?

https://redd.it/pt6yyf
@r_devops

We analysed 30 job postings to understand what is expected from SRE role.

Hey, this is Kaushik from Spike.sh.

We worked on a fun project. The role of SRE has been in discussion for years but expectations still seem slightly mystical depending on every org's functions.


So, we took SRE job postings from Google, Twilio, Instacart, Gitlab, Spotify and more to understand some key expectations.

​

Here is an interesting figure from our article:

>Build tools and automations  (56% of job descriptions)

Only 56% of job descriptions talk about managing build tools and automations. I personally find this kinda shocking. Shouldn't this be in ALL job postings?


Do read and let me know what you think. We have some ideas on digging into more details about SRE, DevOps and more roles, that is if the community finds it useful !

https://spike.sh/blog/sre-role-2021-analysed-30-job-postings/

https://redd.it/pt7zd0
@r_devops

Training Plan For Devops.

As a team in our organization, we are new to devops and i am try to come up with a training plan for folks joining the team. You can expect that these new folks are new to devops and don't have much experience. Is there a training plan for devops?. Few things i can think off are

AWS, GCP, Azure, python, shell-scripting, terraform, jenkins, messaging technologies, docker, containers.

Are there any other things

​

Also, is there any links that you would recommend to above technologies. Udemy is also fine.

https://redd.it/psc1f1
@r_devops

Incite - Golang Library to write pretty Cloudwatch Log Insights queries

This one looks like a nice library to write pretty golang code to get Cloudwatch Log Insights queries.
https://github.com/gogama/incite


Anybody tried this yet? It takes care of streaming, polling, chunking, previewing and also has optional logging support.

https://redd.it/ptb52s
@r_devops

Distributed Load Testing Using Fargate | Open-Source AWS

Hacktober is right around the corner so i finally decided to make a series for working on Open-source projects. Would love any feedback good or bad.

https://youtu.be/LXKavTaeo4E

https://redd.it/pta7zb
@r_devops

k8s exposing multiple ports on the same domain name

I know that exposing multiple ports over the same domain name on k8s nginx ingress contorller is not possible. Any ideas for a good approach to accomplish this ?


example:
example.com:80 goes to service A and example.com:8000 goes to service B

https://redd.it/ptdh4z
@r_devops

App Deployment: CLI vs. Web UI?

I'm building a tool to deploy web applications to the cloud, and I'm wondering whether I should focus on building a CLI or web UI to manage and debug the apps. I would appreciate some feedback on what you enjoy using most!

https://redd.it/ptdtqb
@r_devops

learning on the job vs training

Hey all, I am new ish here. I started a sort of jr DevOps role a few years ago after about 20 years doing support and sys admin work. I always had an interest in DevOps but had none of the dev skills go get there. I transferred internally from a sr support position to a DevOps role so the interview did not have any code testing etc. Over the past few years I have gained a lot of skills and learned a lot of the tools of the trade, however none of it is deep knowledge. We are a Microsoft shop so 90% of what we do is in powershell and while that is an object-oriented language, it's not really used that much in the industry. I am really interested in Python but we don't use it in our team. I guess the question is, if I don't use a tool/language on a regular basis, how do I learn it and moreover how do I advance my career.?

https://redd.it/ptfzi8
@r_devops

Where do people host internal jobs?

I've run into situations where a job or script is needed by non-technical team members. These scripts might be as simple as one or two API calls to our backend, a SQL query, or a monthly report generator. Typically, we tell these folks to make the calls from Postman or have a Jenkins job they can input a few params into and run (we usually want to keep it simple and avoid building UI's). However, it feels weird giving this much access.

Has anyone else run into this problem? I'm wondering if there's any better solutions out there.

https://redd.it/ptdgpl
@r_devops

What do you, as a DevOps, use for infra downtime prevention, APM, and log management?

Hello all!


I'm new to the DevOps world and want to understand what solutions are worth relying on when the thing comes down to anomalies tracking, log analyzing, APM, alerting.

I had an experience with Zabbix, Datadog, InsightCat. You've probably heard about them. According to my colleagues, these solutions are the leaders. Also, we don't forget about ELK stack, obviously :)


I would appreciate your help. Thank you!

https://redd.it/pt4juj
@r_devops

What is DevOps Life Cycle?

DevOps practice that aims at merging development, quality assurance, and operations (deployment and integration) into a single, continuous set of processes.

DevOps roles include development, IT, operations, security, and support.

DevOps is an approach to culture, automation, and platform design intended to deliver increased business value and responsiveness through rapid, high-quality service delivery.

DevOps is complementary with Agile software development; several DevOps aspects came from the Agile methodology.

📷

Plan: Requirements and feedback are gathered from stakeholders and customers and it is used to build a product roadmap to guide future development.

Code: Involve software designs to writing code, designing infrastructure, automating processes, defining tests, and implementing security.

Build: The Build phase is where DevOps really kicks in. Once a developer has finished a task, they commit their code to a shared code repository.

Test: This phase Involves continuous testing (manual or automated) to ensure optimal code quality.

Release: The Release phase is a milestone in a DevOps pipeline — it is the point at which we say a build is ready for deployment into the production environment.

Deploy: The build is ready for its release into production. There are several tools & processes to automate the release & to make it reliable with no outage window.

Operate: The operations team makes sure that everything runs perfectly. The environment scales automatically based on hosting service configuration.

Monitor: In this phase, monitoring, identifying, and collecting information about issues from a specific software release in production is taken care of.

https://redd.it/pt5m4n
@r_devops

The new "cat /etc/passwd" on developer machines is... ?

find . -name ".env" -exec cat {} \\;

https://redd.it/psqua2
@r_devops

Do you have to find work in your team or someone assigns it to you?

I am in a weird position where my manager does not assign tasks to anyone. I am of an introverted nature and do not have the habit of latching on to tasks during sprint planning, which others seem to do.

When I don’t have many tasks, I ask other members if they need help with something, most of the times they don’t as already 2 people are working on one thing.

This is causing me to become depressed. I have been pairing with one teammate, but he manages to do things alone, so he doesn’t need my help. I already mentioned it to my manager and he mentioned that its okay, still keep working with him. I don’t want to upset my manager, but seems like he doesn’t think well of me, which is happening for the first time for me.

I could really use some perspective on this. I am planning to move to backend because of this, preparing on my own, not sure if the same happens there as well.

Thank you

https://redd.it/ptp86d
@r_devops

Retry curl command in jenkins pipeline

I'm trying to run my integrationtests through Jenkins but I have trouble. My full step look like this:

sh 'docker-compose -f docker-compose.dev.yml up -d'
sh 'curl -sfv --connect-timeout 30 --retry 15 --retry-connrefused --retry-delay 2 https://127.0.0.1:8181/health'
sh './gradlew integration'

But when I call the curl command I get: Recv failure: Connection reset by peer
* Closing connection 0

If I run a sh 'sleep 30' before my curl it passes which added to the fact that it fails in around 5 seconds leads me to believe that it doesn't actually retry but reads the exit code from the first curl command.

My question is how can I either retry my curl or await my containers dynamically without a sleep or similar hardcoded solutions?

https://redd.it/ptqgt6
@r_devops

A new voyage in the sea of observability

In the observability market, new tools based on open-source projects as well as commercial solutions are emerging and it can be difficult to understand differences and advantages. So if you want to learn more, check out this new article and find out why NexClipper is embarking on this exciting new voyage!

https://nexclipper.io/nexclipper-a-new-voyage-in-the-sea-of-observability/

https://redd.it/pto42a
@r_devops

Securing Serverless Applications with Critical Logging

In this article, how serverless architecture can benefit your application; graceful scaling, cost efficiency, and a fast production time are just some of the things you think of when talking about serverless.

But what about serverless security? What do I need to do to ensure my application is not prone to attacks?

Read here: https://towardsaws.com/securing-serverless-applications-with-critical-logging-2a2d0da8697d

https://redd.it/ptoznz
@r_devops

Terraform loops are improved be still complete rubbish

Fight me

https://redd.it/ptuq21
@r_devops

GITLAB CI/CD PROCESS NOT STARTING

Notes:

* CI/CD is already set up across all repos

* new repo is made following the same structure and has the correct .gitlab-ci.yml file

* runners are available

yet, after code is pushed, the pipeline doesn't start. It says the pipeline isn't setup... So i went to the settings page for it --> turned off CI/CD --> saved --> turned on CI/CD --> saved --> no luck

can anyone suggest next steps I can take? The CTO is away and i'm slowly learning this but have hit a roadblock

https://redd.it/ptxedx
@r_devops

SilverSurfer - An OpenSource project to check ApiVersion Status and provide Migration path for Kubernetes objects when upgrading Kubernetes to 1.22 or any other.

Hey Everyone, as Kubernetes released its version 1.22, Developers and DevOps Engineers need to upgrade Kubernetes Objects before upgrading Kubernetes. It's a tedious task to know whether the current ApiVersion of the Object is Removed, Deprecated or Unchanged. We thought why not build a tool, Silver-Surfer, that would not only validate against the given Kubernetes version but also provides a Migration Path to a newer ApiVersion. It shows exact errors which need to be resolved for migration to newer ApiVersion as well as specification errors with respect to the provided ApiVersion in case it is not removed.

https://github.com/devtron-labs/silver-surfer

I know that there are other tools that provide similar functionality but they seemed Inadequate

Kubeval \- Well known, It only validates against the given Kubernetes Version but doesn’t provide a migration path

Kubent \- Only provides Info about removed and deprecated api but no schema validation

Kubepug \- Only checks whether the existing objects have any Deprecated ApiVersions

Silver-Surfer uses openapi spec provided by the Kubernetes with releases, for eg. in case of target kubernetes version 1.22 openapi spec for 1.22, to validate the Kubernetes objects for depreciation or non-conformity with openapi spec. This allows it to validate against any kubernetes version as soon as its specs are released on the github kubernetes project.

It's a side project we took up as these issues were a hassle upgrading one of our Platforms, Devtron.

Would love to know what you think of this, any suggestions and Contributions are welcome :)

https://redd.it/pty4bh
@r_devops

Any Unity (Game) DevOps Engineers Have Advice For CI/CD?

I am setting up the CI/CD for game development in a repository. I am using gameci and trying to get it to run in AWS but it is proving to be more difficult than expected. Our current stack is GitHub actions, AWS, and Github.


if you work on game development with Unity, what is your current CI/CD workflow and what advice would you have for me?

https://redd.it/pu035w
@r_devops

Why does Bazel not get more love?

We just ran a PoC to port over all our dev, build, and deploy tooling to run on Bazel, and so far it’s been an absolute dream. We have a medium-sized codebase, probably a bit small for Bazel’s ‘sweet spot’. But, forgetting all the reasons why Google made it, having one set of tools that compiles and deploys all your code to any environment, and with only the changes that are necessary, feels like the future.

We’re thinking about rolling this out to the whole of our org, but we’re getting pushback because, frankly, it’s an unknown. There aren’t that many articles on it outside of the C++/Java world, and we’re a JS/PHP/Go shop. But why? There seems to be plenty of support in the actual code. It feels like a diamond in the rough.

I could talk about it for hours but I’d love to hear your opinions. Did you consider adopting it? What did you ultimately end up choosing?

https://redd.it/pu30to
@r_devops