Is testing part of a build process, or should they be seen as completely separate steps?

Disclaimer: All binaries should be fully tested before letting them touch production side. Period. This is something is fully believe in and won't deviate from.

There's basically 2 views on this:

1\. **Yes, the testing is part of the build.**

When you push something to your branch, this branch should be build and tested fully. This means the developer will have to wait for the entire process(includes unit testing and integration testing) to complete. The mindset here is that an untested binary is not a valid output and should not exist.

2\. **No, the testing is a separate**

Same thing as above, but if you push something to a branch ci/cd picks up your new changes and builds your code. This binary/output is then stored somewhere as an untested blob. *Then depending on your configuration*[1], you can run another job to test the output. This test job then does whatever the other method of working does and stores/deploys the tested output.

-----------------
My opinion:

Personally I'm in the mindset of the latter because it decouples the build from the testing, giving you the benefit of [1]. Not every branch needs to be tested on every push. I want to push and get results asap. I don't want to wait. The former mindset makes no sense to me since it just sounds like "tests are mandatory so every build should be tested", but that doesn't mean you should implement it on a technical level like that.

I'm curious about opinions from this sub.

https://redd.it/ezwztw
@r_devops

Is a devops job a good move for me?

I am looking for new jobs and I am currently fullstack developer (technically lead architect) for the startup I work at. I do a lot of devops in my current role and am about to sit the AWS CSAA exam and I do enjoy this aspect of my work. I am interviewing for a Head of Devops role at a challenger bank which I think I have a good shot at but this would likely not include any application development.

I would like my future roles to be things like Head/VP of Engineering do you think this is a good career move?

https://redd.it/eztr4s
@r_devops

Where to start ?

i'm pursuing my career towards cybersecurity and i don't know much about devops i thought it would helpful if i add devops to it but i don't know where to start so can you guys help me by giving good information

https://redd.it/ezswxz
@r_devops

Is there a way to hit a particular pod for a request.

So i have a kube cluster deployed for a service which runs around 50 pods. A few of them having been going out of memory and I am pretty sure there is memory leak in my application. I have written an endpoint in my application which takes the heap dump and then uploads it on s3. I want to take this heap dump on a long running pod so i can debug the mem leak. Is there way i can hit a particular pod using postman by passing in some selectors?

https://redd.it/ezsp4n
@r_devops

Application monitoring tool

I am looking for an Application monitoring, health check, and service map dashboard tool for node js and java based applications. I host applications in AWS EC2 servers.

Could you please tell me which is the most popular tool for this purpose?

Which tool can help ?

https://redd.it/ezs73j
@r_devops

[DevSecOps] Stick to CentOS 7.x for DevSec Linux Baseline or attempt to adapt it to CentOS 8.x?

I'm using https://github.com/dev-sec/linux-baseline to create hardened image basic templates. However, the repo hasn't been updated to support the newer distributions yet, and I need to stick to CentOS for a specific project.

In terms of best practices, should I stick to CentOS 7.x as a known safe default, or or try to lift up the DevSec Linux Baseline to support CentOS 8.x instead, without really knowing what I'm doing?

https://redd.it/ezq838
@r_devops

Build a Serverless appliance with cloud-init and faasd

Have you heard of faasd, [the new provider](https://docs.openfaas.com/architecture/faas-provider/) for OpenFaaS? It's just like OpenFaaS on Kubernetes, the same API, same UI, CLI and ecosystem, but without the complexity and stress of running a scale-out distributed system.

In this short tutorial I'll show you how to leverage cloud-init, a standard way to install packages and to configure cloud hosts to get a faasd host in less than 5 minutes flat. Or your money back.

[https://blog.alexellis.io/deploy-serverless-faasd-with-cloud-init/](https://blog.alexellis.io/deploy-serverless-faasd-with-cloud-init/)

If you're avoiding Kubernetes, or keen to use the right tool for the right job, would you use this? What would you like to see in terms of documentation and literature around it? If you try out the tutorial, please let me know what you think.

https://redd.it/ezpybt
@r_devops

Do you use Pull Panda or similar tool to solve the issue of unnoticed code reviews?



https://redd.it/f08uaa
@r_devops

Step by step guide for DevOps or any other Operations Role

\*Hello Noobs , who wants to begin this journey with me , i highly recommend Myself , a young sys administrator with a lot of motivation to succeed in this field .... check out the link to see the roadmap.

https://redd.it/ezqs5o
@r_devops

VPN *INTO* China

Hi guys,

I just received a very... unusual requests from one of my clients.

He's located in EU and needs a possibility to connect to an FTP server located in China.

Of course, port 21/22 is blocked from outside of China and of course a special whitelisting rule cannot be added.

Because... why would it be so simple?

So my question is: Do any of you know of some sort of VPN/Proxy/server in a basement that I can use in order to "get into China"?

Thanks...

https://redd.it/f0a07m
@r_devops

Looking for recommendations on a Orchestration tool for my specific use case

I wish to orchestrate the following

Terraform plan/apply to AWS
Packer build for AMI’s
Ansible for configuration within the client OS

It is for a new project, I already have good knowledge of the above but need to bring in an orchestration layer above, as currently ops are pushing manually to AWS

Goals
Changes are made within a Git repo by a ops engineer and are put through several automated steps to validate and obtain approval prior to deploy
LDAP integration
UI to be view deployments
Secure
Highly Available
Easy to manage (Team is made up of mainly infrastructure folks, not developers)

I was leaning towards Jenkins but not sure if its over kill for our requirements - we don’t develop applications only automate the deployment of infrastructure and deployment of instances etc

I appreciate any advise you are able to offer

https://redd.it/f0ehzo
@r_devops

Is Packer still the best choice for building images for multiple clouds?

We need to run custom images on a number of clouds and currently this is mostly done through bash scripts and the like. I was looking into using Packer to consolidate all this under one tool, is it a good choice? Any competing tools that show promise?

https://redd.it/f09m2o
@r_devops

Passing Bitbucket credentials and Personal Access Token via script.

Is there any way to pass a credential and personal access token for doing a git pull from BitBucket via script.

I'm not able to use an SSH key for this script because I can only connect via https, but I'm able to use a personal access token.

https://redd.it/f0bgo3
@r_devops

Self hosted CI/CD option

I'm looking for a tool that could replace Jenkins. What are the good CI/CD that could be hosted on AWS. I'm the only devops engineer in a team and I don't have enough time for managing jenkins

https://redd.it/f0a6ko
@r_devops

Deploying using a batch to multiple machines, what tools will help me?

So right now I'm working with a legacy system, I know a big surprise, and our system is distributed across multiple machines that are identical. For example we have 4 servers of type x, 4 servers of type y, and 4 servers of type z all which work to run our application, each of these servers uses a batch file that pulls down some code from a repo and builds out the repo it pulled down.

Now what I want to know is what type of tools I can use to make our deployments faster, centralized, and to avoid as much human error as possible, currently our deployment requires logging onto each machine and running the batch file which is laboring , not to mention very prone to human error. I imagine I could write some scripts to call out to each machine and have it run all my batch file as long as it is in a common path across machines but this loses me a few things and overall just really doesn't seem like a good solution I could go further into why but I think we understand. My next thought was to use something like octopus and have that installed on each machine and then using a process within Octopus to trigger the batch on each machine and in that way I could avoid human error, have a centralized tool that I have some experience with in the past.

While I believe Octopus solves the issue I'm working on I'm curious of any other ideas for how I could approach this problem. Any and all suggestions / criticism about my current solutions are appreciated.

https://redd.it/f08fre
@r_devops

Rundeck To jenkins migration.

Can anyone help me..,,

My all existing jobs are running on run-deck. Right now i want to run everything from Jenkins Can anyone help me how to migrate the jobs from Rundeck to Jenkins.

https://redd.it/f068i0
@r_devops

CI/CD configuration management

Hi,


How do you guys solving the configration release and application release issue when the application haveto be restarted multiple times but the pipeline is the same for all environment?


I mean if config changes apps restarts
If app changes a restart required aswell.


which means when you reach the point on the last deployment prod the app will restart twice...

https://redd.it/f0ic21
@r_devops

Finally...

Been 5 years in a multinational in an ops capacity. Rotting, professionally and mentally. No time to change any of the legacy bullshit procedures. Spending more time in meetings with overenthusiastic PMs trying to justify their existence than doing actual work. I'm sure many of you have been there.

A semi-surprise org change just pushed me to a managed services team that wants me to get their stuff ready for CI/CD. I finally have the autonomy to flex my python and automate the fuck out of all those stupid manual tasks that have been annoying me for years. I'm excited to go to work again.

https://redd.it/f0g1tq
@r_devops

Strategies for deploying multiple application versions in a DR scenario

Hey all,

Been stumped thinking about this recently. In my situation, we have an application that needs to have multiple versions deployed side-by-side.

Via a branching strategy, we create a long-living branch off of master for each version release. Along with that is a infrastructure-as-code file to create the services needed. (In this case, it is a Azure ARM template, but you could plug anything in here).

Now where I get confused is in thinking about a Disaster Recovery scenario. There's no one branch I can depend on to deploy all my application versions again. I would have to go through each version branch to redeploy everything.

In my opinion, that seems pretty ugly. But I am not sure of any better strategies. It is almost like I want the master branch to know about all the versions, so its ARM template can deploy everything needed.

I'm really interested to hear what others have done to solve this (if it needs solving).

Thanks!

https://redd.it/f0fmq1
@r_devops

IT Help Desk -> Junior DevOps / Entry level Cloud roles?

Help Desk to Junior DevOps?

I graduated with a Bachelors Degree in Computer Information Systems and spent the past 3 years in IT Support / Help Desk roles. I gained a great bit of knowledge and experience working in those roles as an intern (at a Fortune 500, at a Wealth/Finanace Company, and at a MSP).



I was lucky enough to accept a great job out of college doing IT & Software Support. About 5 months in, I realize that my heart really isn't in doing Tier 1 type Help Desk stuff (resetting passwords, program installs, recovering emails). I am starting to work with the software support team and realize all of these guys have a lot of coding knowledge in work in DevOps type roles. I figured I worked too hard in college and spent most of that time busting my butt interning to just be complacent with a job doing low level work. I am thankful for where I ended up... but I want to continue learning and climbing the ladder rather than be complacent.



My goal was to start studying for the AWS Certified Solutions Architect - Associate certification. After that, I wanted to grab a better understanding and study up on Linux (I have minor knowledge in Linux atm). After I got those 2 down - I wanted to dedicate my time learning Python and automation.



Is this a good learning track towards a Junior role? Has anyone moved out of a Help Desk role into a Junior DevOps role?



All suggestions help!! Thanks!

Resume: https://m.imgur.com/gallery/u51h9wa

https://redd.it/f04fxj
@r_devops