Reddit DevOps
271 subscribers
45 photos
31.7K links
Reddit DevOps. #devops
Thanks @reddit2telegram and @r_channels
Download Telegram
Should devs have access to production?

I'm trying to move my org towards a devops culture and one thing I'm struggling with getting across to leadership is that it is okay for devs to be able to at least have read-access to production. If devs are to be responsible for their code, it seems obvious that they should understand the production environment, and be able to investigate issues there - at least that's how its worked at my previous gigs.

​

How do you manage competing concerns of developer autonomy and security/safety?

Do devs have access to prod? How about contractors?

What safety nets do you have?

https://redd.it/u2xz7e
@r_devops
Is it a good idea to version the platform deployments?

I manage a large number of kubernetes clusters, and deployments are done using ansible and gitlab pipelines.
Each cluster has its unique properties in a ansible inventory folder.
What I wanted to know is, just like software releases, is there an approach to version IAAC deployments too.
It is more of platform release management question, rather than a kubernetes question.
For example, let's say I want to change a property in one cluster, so I create a branch and apply that change using ansible to that one cluster. Now, I want to name that change with a release name so that I can communicate the same to other stakeholders and at the same time propagate that change to other clusters.
Maybe I am doing/visualising it completely wrong, thus looking for suggestions.

https://redd.it/u35hh2
@r_devops
What is your company/consultancy's hourly billable rate (not their hourly salary) for a Senior DevOps Engineer? Or what rates are standard in your region?

I'm trying to gauge the market across the US and world for the above engineer. For example, here in NYC, I have companies charge anywhere from $225-300.

https://redd.it/u35g1n
@r_devops
Things I Wish I Knew Earlier About Pulumi

Hey folks! I've been able to check out Pulumi in a real-world use case over the past few months, and wrote up the high-level things I learned which seem to be missing from basic docs and tutorials. You can check it out here.

My takeaway is that it's pretty cool, but if you want to make use of the Terraform ecosystem you would be better off using Terraform.

To those of you who have used Pulumi already: have I missed any good-to-know-early gotchas? To any IaC practitioners: are there more-advanced Pulumi or Terraform resources you could recommend? I couldn't find much around Pulumi, and can't help but feel like I must have missed something. Thanks!

https://redd.it/u3b3wf
@r_devops
I made a website that helps you do Cloud practical projects for free

I made a website that aims to help those wishing to set up practical projects through an interactive and detailed documentation, and this while taking advantage of the Free Tier from AWS.

This site was made for those who want to get started in the Cloud or have little practical experience (Juniors for example), and wish to set up practical projects to learn by doing and especially to present those projects during interviews.

No more fear of receiving a hefty bill, the 3 projects currently published all use Free Tier tools, which allows to create projects at reduced costs or even completely free of charge (everything is indicated on the website).

I have made sure to write each project in as much detail as possible so that everyone, regardless of your level and knowledge, can successfully complete the projects.

By the way, the website is totally free and open source, it's my way to help the community as it helped me.

Let's recap: An open source website, free to access and use, and helping to create almost free projects using the Free Tier... The best name to give it could not be other than "Cloud is free".

Here is the link to the website: https://cloudisfree.com/

And the link to the Github: https://github.com/ydamni/cloud-is-free

I am open to any feedback, and would love to hear from you.

https://redd.it/u3fo3p
@r_devops
what is the difference between ci and cd ?

hey guys, i'm working on a project where my data input is coming from google sheets automatically using sheets api then processed in order to get my final product which is a dashboard describing the total of sheets in the spreedsheet every week, i used for this python and streamlit and i want to deploy it on heroku, but i'm new to this, my question is how i can make changes later on my source code if i need to add a new feature without affecting the prod web app (dashboard).

thank you

https://redd.it/u3i5a6
@r_devops
Who wants to learn Kubernetes this weekend?

I had put a lot of effort in creating this video series (https://youtube.com/playlist?list=PLVx1qovxj-akr\_3XqQQgpqRyQw4GYuS4h) on Kubernetes for beginners. It consist of basics on kubernetes, creating minikube, kubernetes cluster, k3d cluster, Deployments, services, replicaset, HPA, Integration with Jenkins to deploy your application to Kubernetes cluster, kubernetes screts and more. .

I am sure this will help you to enhance your skills. I would appreciate a look on it :)

https://redd.it/u414dr
@r_devops
The best UI for log monitoring?

In terms of functionality, visual, etc. Is it Kibana, Grafana, or something else for you?

https://redd.it/u44f9n
@r_devops
Monitoring and Metrics replacement project - looking for suggestions opinions

​

Hi all,

I'm a System Engineer with limited knowledge around monitoring and metrics working on a small team. I was handed a project to work on cleaning up what we have (SolarWinds) and gathering basic metrics for business critical items to be observed and shared within IT. I wanted to present my overall landscape and see what more knowledge folks had suggestion wise.

**The current landscape: \~**

* 200 servers
* 80 different physical locations (offices)
* fair amount of networking equipment
* 17,000 elements (data points/thingymajigs) in SolarWinds speak

We are using SolarWinds to monitor and alert via email and teams, nothing groundbreaking here. We have a mix of on-prem hardware, AWS, and Azure. For the most part it’s basic out Monitoring of servers and network equipment like up/down, CPU, Mem, Disk, bandwidth, with a few oddballs sprinkled in but nothing critical on the oddball stuff. We do have a bit of SQL monitoring and IIS monitoring but none of it is tuned in the slightest.

**The need:**

* Agentless (I know this is a limiting factor for greater detail)
* General cost on par with SolarWinds (not terrible because we are relatively small)
* Ability to the monitor basics above
* Need the ability to pull in data from other sources into a presentation tool such as Grafana. Some of the metrics I'm being tasked to pull in will not be easy or practical to pull from SolarWinds or Zabbix, or whatever we choose from the monitoring side. So the ability to pull in data from other systems, CSVs, APIs, or just text would almost be a requirement.
* The big "WANT" is, 1 central place to display & present the data.

I’ve started looking into things like Zabbix + InfluxDB + grafana, PRTG, (Datadog soon), and generally lurking in this forum for the last few weeks researching stuff. But as with a lot of us out there in the business this isn't my primary responsibility so there is only so much brain space to devote to it :).

As an aside, SolarWinds is pretty quirky and clunky but does things “OK” as my org needs them. I’m pretty sure with a stupid amount of work i could make SW do most of what needs to be done. But the level of effort involved there might be 2-3x more than dropping in multiple other systems.

https://redd.it/u3y6d8
@r_devops
Cake v2.2.0 released 🎉

Highlights of this release

Opt-in script caching for significant performance gains

New aliases for using dotnet workload search

Fixes in UNC and relative path collapsing

Improved handling of .NET runtime reference assemblies

More details in the release blog post

https://redd.it/u47o82
@r_devops
Service Meshes - Reasons to avoid them?

I've read a lot about Service Meshes and like a lot of the possibilities but have seen a lot of things saying they should be avoided until absolutely needed.

I haven't seen good descriptions of why, however, outside of "it takes a lot of effort".

Can someone explain why it wouldn't make sense to start out with a service mesh with defaults and then invest in individual features when it makes sense?

Are there legit horror stories or is it just scars from early adopters that have been mostly worked out?

https://redd.it/u4ae46
@r_devops
I'm Implementing DevOps in my Organization. Which CI/CD tool to prefer

I'm currently asked to choose from one of the three Jenkins, GitHub actions and Travis CI to use as CI/CD tool but I don't know much about any other tool except for jenkins.
My organization is a growing startup, all the codebase is in GitHub and currently I'm the only DevOps person who itself is an intern. I really don't wanna do something that will haunt me back later.
Any suggestions??

Edit: I'm highly familiar with Jenkins though

https://redd.it/u4nw0x
@r_devops
most cost-effective backend

my backend runs fine on a high-end consumer-grade computer. It doesn't need a GPU. I am wondering what is a cost effective backend service. AWS ec2 seems too expensive.

https://redd.it/u4n3eq
@r_devops
Azure Devops > Simple File Copy with Deployment Groups

I am defining a Release Pipeline - and I want to simply copy files (the deployment assets) to a directory on the remote agents (on-prem). I do not want to specify credentials. The deployment agent should pull the assets from Azure, similar to how the "IIS Web App Deployment" task group works.

This is a simple console app that I am deploying. I cannot find an appropriate task to do this. Anyone have any ideas?

https://redd.it/u4j5q0
@r_devops
Roles on continuous delivery

Hi every one, due to a university protect I was investigating about the roles in a software company or software development team, that implements continuous delivery, all I can find was the roles in scrum, and what I need is a more general set of roles.

I really appreciate any kind of help, and sorry for my broken English.

https://redd.it/u4h67v
@r_devops
New Route53 Cli release in - Get info about your records from the terminal - quickly!

New Release - r53

Example:

r53 -q
my.company.domain.com

It will return a list:

Hosted Zone ID + Web URL
The target behind (Load balancer, Lambda, etc) + Web URL to target
Recursively expand records
Verify NS match with dig

Install:

$ brew tap isan-rivkin/toolbox

$ brew install r53

New features:

\- Exposed SDK on top of CLI

\- Now support JSON output --output-json

https://github.com/Isan-Rivkin/route53-cli

​

\#aws #route53 #golang #go #dns #networking

https://redd.it/u4u0it
@r_devops
Jenkins vs gitlab

Hi people. About CI/CD, I would like to know your opinion between Jenkins and gitlab, the pros and cons. Can we talk about?

https://redd.it/u4kxj0
@r_devops