CI/CD pipeline from scratch



Hi guys,

I need some advice from you. I have been following an open source project called OpenEMR for quite some time now. It is a HIPPA compliant Electronic Medical Records system used in hospitals and clinics.

I would like to install and host this EMR system in an african rural hospital where i am originally from. The installation and hosting will be at my own cost as my small contribution to humanity.

I would also like to create an entire CI/CD pipeline from scratch on which i will be collaborating on with some friends of mine back in africa to continuosly develop and fix issues/bugs that will inadvertently pop up. I plan on initially hosting it on a local server and ultimately move it on to AWS.

Could you guys give me any guidlines on how i can do this? What will i need to set up this CI/CD Infrastructure? What costs can i expect to encounter being that i am based in Europe and the project will be in Africa? Do i need any other hardware other than a high end machine that i have at home?

Thanks a lot in advance for your inputs/comments/questions/advices.

https://redd.it/lfz8a1
@r_devops

Docker in Kubernetes is about to be gone. What now?

The dust created after the news about Docker Shim deprecation has settled.
Everyone had plenty of time to figure out how to deal with the disappearance of Docker from Kubernetes clusters.
Yet, Docker Shim deprecation is still a cause of confusion and, sometimes, panic.
Here's my take on the "situation" >>> https://youtu.be/xa453MkdaAk
Did you have to do anything? Was it "life as usual", or are you putting your fate in Mirantis taking over Docker Shim maintenance?

https://redd.it/lg5eiz
@r_devops

Azure DevOps Pipeline: To upload the artifacts to another repo (in same project)

Hey ADO experts,
I'm wondering how can I upload the artifact from the same pipeline to another repo in the same project. I'm looking for any task in ADO pipeline which can do it.


I've been doing it on my server using powershell to push files in Git, using git commands. I tried to do the same thing in ADO pipeline using PowerShell task but it does not work.


Your help will be deeply appreciated.

https://redd.it/lfcjx2
@r_devops

Terraform: Serverless Kubernetes Cluster with AWS EKS and Fargate

Running Pods on Fargate with EKS Cluster and exposing them for the outside world with the help of load balancer controller (elbv2) using Terraform.
Also shown how to create both Testing and Production environments with the same Terraform code. And code management with Terraform modules.

Read complete article here :-
https://link.medium.com/Dks0bmfhzdb

https://redd.it/lf8f8l
@r_devops

Hardware Virtualization within CI

Hey I’m wondering if anyone has some good reads about hardware virtualization? Basically I’d like to emulate hardware to flash firmware and run tests, ideally without needing hardware so my solution can be completely cloud based.

Anyone done anything similar before?

Thanks!

https://redd.it/lgofez
@r_devops

Kubernetes - shibboleth + ldap

Using shibboleth and ldap for auth and user management in our app running on k8s.

I'm building helm charts for these based on a legacy project. Wondering if anyone can set me on the track for using something more up to date, perhaps a helm chart combination that is widely used for this purpose.

Have looked at options like cognito but would prefer a k8s based solution.

I'm a bit nooby on this subject, any guidance will be appreciated 🙏

https://redd.it/lggdr3
@r_devops

I made an archive of default config files

Hi everyone!

​

I encountered the following problem:

When editing configuration files I sometimes reached a point where I messed up the config file pretty bad. No point in trying to fix things, when clearly a fresh start makes more sense.

Turns out finding the shipped original version of the configuration file is actually pretty hard. If you are lucky you find this original config file on some random GitHub Gist or accidentally indexed FTP server.
You probably know this situation.

My Solution:

I went ahead and built an archive of these clean, original configuration files. Now you and me got them all in one place.

They are collected from a fresh installation of the application (Apache, nginx, MariaDB, etc) on various Linux distros (Debian, CentOS, Alpine, etc). This is important because config files differ slightly between Linux distros.

​

The Tech:
Frontend: static site generated with Hugo.
Backend: applications are installed in Docker containers. The shipped configuration files are fetched and the containers are destroyed. This process is repeated for every Linux distro.

​

Let me know what you think, what I should improve and what applications/distros you are missing and so on!

TL;DR: My goal is to provide a collection of original, unaltered configuration files for multiple applications.

Link: https://exampleconfig.com

https://redd.it/lgpsbd
@r_devops

Kubernetes alternative to shibboleth

Currently try to build a helm chart around shibboleth, ldap setup.

Wondering if there are any good helm chart alternatives to manage application auth and users on our cluster.

I'm a bit nooby on this subject so doubting ambiguity of the question.

https://redd.it/lgg96u
@r_devops

On-demand self-hosted AWS EC2 runner for GitHub Actions

Hey folks!

As GitHub recently made it possible to create self-hosted runners using API, I've created a new action for creating on-demand self-hosted runners on AWS EC2. Automatically right before you need it.

Please check the repo: https://github.com/machulav/ec2-github-runner

We started using it for running jobs that need access to the private resources in the VPC like a database migration. Maybe it will be useful also for some of you.

I would really appreciate hearing your feedback and thoughts on that!

https://redd.it/lgg2ai
@r_devops

Prometheus Retention Period

Hi there,

What happens to Prometheus metrics data after the retention period ? How can we persist that data without increasing the retention period ? I understand increasing the retention period also increases the storage capacity of the TSDB ?

I'm a beginner with Prometheus, please help with recommended techniques. Thanks

https://redd.it/lgdgs1
@r_devops

Load balancing pods based on least connection on GKE

I have a kube cluster setup on GKE. I want to load balance my pods based on least connection algorithm. I saw that kube proxy has IPVS mode which supports least connection algo but thats not available on GKE.




Is there any way I can load balance pods based on connections on Google kubernetes engine.

https://redd.it/lgap13
@r_devops

Serverless Framework with Terraform: A firsthand experience

Hey all - thought I'd share some learnings and experiences we've had getting the Serverless Framework and Terraform to work in sync. Was quite a journey and hope this helps anyone else trying to get started with a serverless application.

Let me know if you have any questions! Happy to help where I can.

https://redd.it/lg9k7t
@r_devops

Packer - vsphere-iso + Windows 2019 + VMTools service not installing correctly

Hello,

I have a frustrating issue with Packer, specifically with VMware Tools installation on the latest ISO from Microsoft for Windows Server 2019.

During the Packer install, I load up a script and have VMware Tools 11.2.1.17243207 installed. It seems to install successfully, I can't see anything in the logs but I notice VMTools service is not running. I have to re-run setup64.exe via the GUI and do a repair, then I see the service exist and runs, and Packer can discover the IP address of the VM to finish it. I opened a ticket with VMware to see what might not be working right.

This is how I try to run it in a script.

e:\\setup64 /s /v "/qb REBOOT=R"

Anyone ever come across this?

https://redd.it/lgvbxa
@r_devops

What are some Kubernetes concepts or gotchas that beginners should watch out for?

Hey devops peeps. How's the helm treating you? I'm pretty fresh to Kubernetes, but I live on the outskirts of the ecosystem. Conceptually it's pretty awesome, though as I understand it, in reality it's not as easy as the promise dreams. This is a good article on some of the concepts for someone like me. When I close my eyes, I think of an octopus with that elegantly controls its arms to pull the right levers and push the right buttons with impeccable timing. But maybe it's more like Octodad :)

If you were to design (or recommend) a good course for someone like me, what would it look like?

https://redd.it/lgwrrr
@r_devops

Unit testing registry metrics

Hi,

I am wondering if there is a way to unit test metrics, I am using micrometer library with springboot.

In Class A I use a Timer and counter.

In Class B I would like to do simple unit test to assert correct counter increments and such.

A MeterRegistry is instantiated in the constructor of class A, then a timer and counter are initialized in Class A to monitor certain value.

I have visibility of the MeterRegistry in Class A in my Test Class (Class B). Is this visiblity enough to get the necessery information of the timer and counter to use in my Assert statement of Class B? I'm not certain how to approach this problem and if it is the right way.

I am not sure how to proceed, and since I find no information about this online I am wondering if it makes sense? Or am I missing something?

Thank you.

p.s. Sorry for the title in all caps I have a weird bug, tried copy pasting lowercase and it stills shows up as all caps

https://redd.it/lh3fud
@r_devops

Kubernetes Monitoring

Hello All,

In our environment the production servers have resource constraints, we are in testing phase of deploying K3s. For the cluster monitoring, initial perception is Prometheus will be heavy on the servers and might not be potential comptabile candidate. I am looking for some light weight open source tools for monitoring the cluster and all objects (pods, deployments, replicasets, etc). Can you guys suggest any suitable tools for my use case ?

Thanks and appreciate your assistance

https://redd.it/lh9jeg
@r_devops

Thoughts on eduonix devops courses?

I’m a total noob when it comes to devops. I just started kodekloud a few weeks ago and I’m really liking it, but am looking for other material too. Eduonix has a big sale on their devops courses, but I haven’t used their site before. If anyone has taken any of these courses, would you recommend them?

https://redd.it/lh98dt
@r_devops

Looking to change from Bash files that transform text files to a CI/CD system

Hi,

Currently I have a bunch of shell files running bunch of Python scripts that receive a zip file, which I manually unpack, and transform data in different stages.

The bunch of Python scripts part is necessary. I would like to set up this in a CI/CD system where myself and other devs can alter the python scripts on our git repo. Then at the end of the day scripts are pulled, the data gets processed in stages, and if one of the stages fails, we get an e-mail or similar.

I guess I described it too much, but I was looking at Jenkins. Would this be the ideal solution?

Thanks in advance

https://redd.it/lh5c82
@r_devops

What are your best tips for avoiding alert fatigue?

Is it a matter that can be fixed by choosing the right system, or just by company culture and a dedication to removing noise?

https://redd.it/lh3wkw
@r_devops

Knowing what you know now if you could best advise junior devs on how to be competitive for dev ops what would you say is number 1 thing?

Im a typical junior dev but a big picture with solid communication skills who enjoys planning code with goals and i enjoy leading others.

Id like to take the route that is the most logical and also the strongest competition wise to enter into the dev ops engineering field.

Which of these is number top of your list if you are a junior dev aiming for dev ops?

View Poll

https://redd.it/lh6yeg
@r_devops

Resources to get the networking fundamentals I need?

I've been working in the DevOps world for about half a year now. Got some certs and experience at a client on azure. Currently diving into openshift for the next client but i keep realising my networking knowledge is sub-par.

Did this fundamentals course on udemy but it was basically what is a router, switch etc but not how to configure all that stuff. Some colleagues don't think this is relevant, but i just think different and wanna have a solid foundation.

So i learn best with video/reading + labbing. Do you guys have some advice on how to get up to date in networking?

Saw David bombals CCNA for instance? Also heard that it the CCNA might go too deep?

I really wanna go quite deep and really understand networking (eventually want to cloud architect in 3-5 years) and don't wanna half-ass my job.

TLDR: Work in DevOps role , wanna get a solid foundation in networking and looking for resources. Any suggestions?

https://redd.it/lhet5u
@r_devops