Which DevOps tool you would like to replace in 2020 and why.

Even though ours is a startup, I would like to replace Docker Swarm with Kubernetes.
I would like to see it’s power that everyone else is talking about. Our containers and workloads are increasing in number and yes I think it’s better to make use of Kubernetes at this point.

Which tool your world like to replace from your existing stack and why?

https://redd.it/eitfsi
@r_devops

How do you guys handle rollbacks with automated CI/CD pipelines?

Title. If you have a bad release, what do you do? Revert master to the last good commit, then let the pipeline take care of things? An alternative process?

https://redd.it/eid1ak
@r_devops

Help: Having trouble with routing urls

For a school project, we need to deploy our spring boot apps on kubernetes servers. Right now I am testing locally, but I am having some trouble getting everything to work together. At this point, I have a frontend, with a custom nginx docker image. A backend docker image and an ingress. I am testing everything locally right now. I can access the frontend, through localhost (as defined in the ingress). I can use the IP address of the backend service, to do the url GET's (like /stock/{id}). But somehow my ingress doesn't route it correctly. localhost/stock/id doesn't shows an 404. Could you guys tell me what I am doing wrong?

I also don't know what url to use in the frontend, to refer to these calls. Normally you could use the dns name for the requests, but since we probably don't have a dns, I don't know if you can dynamically get the server ip to use? Right now I could use localhost, but what when I run it on the servers we get to use?

If you need any other files, please let me know. Thanks in advance.

The ingress yaml:

```
apiVersion: extensions/v1beta1
kind: Ingress
metadata:
name: gateway-ingress
annotations:
nginx.ingress.kubernetes.io/enable-cors: "true"
nginx.ingress.kubernetes.io/cors-allow-origin: "*"
nginx.ingress.kubernetes.io/rewrite-target: /

spec:
backend:
serviceName: default-http-backend
servicePort: 8080
rules:
- host: localhost
http:
paths:
- path: /stock*
backend:
serviceName: shop
servicePort: 2230
- path: /shopItem*
backend:
serviceName: shop
servicePort: 2230
- path: /transaction*
backend:
serviceName: shop
servicePort: 2230
- path: /
backend:
serviceName: frontend
servicePort: 80
```
The nxinx config file:
```
upstream shop {
server shop;
}


server {
listen 80;
add_header 'Access-Control-Allow-Origin' "$http_origin" always;
add_header 'Access-Control-Allow-Origin' '*';
add_header 'Access-Control-Allow-Credentials' 'true' always;
add_header 'Access-Control-Allow-Methods' 'GET, POST, PUT, DELETE, OPTIONS' always;

location / {
include /etc/nginx/mime.types;
root /usr/share/nginx/html;
index index.html index.htm;
try_files $uri $uri/ /index.html =404;
add_header 'Access-Control-Allow-Origin' "$http_origin" always;
add_header 'Access-Control-Allow-Credentials' 'true' always;
add_header 'Access-Control-Allow-Methods' 'GET, POST, PUT, DELETE, OPTIONS' always;
gzip on;

}
location /stock {
proxy_pass https://shop;
}
location /shopItem {
proxy_pass https://shop;
}
location /transaction {
proxy_pass https://shop;
}
}

```

https://redd.it/eiyhfp
@r_devops

What are your devops goals for 2020?

Any tech you want to use?
Anything you don’t like?
Things you wish existed to make our lives easier?

My personal goal is to explore different CICD tools as I’ve only used CircleCI in the past.

https://redd.it/ej3oid
@r_devops

Creating a CI/CD

My team and I are relatively inexperienced with DevOps. We have started to create a CI pipeline for a product. First step is to build on commits (on commit triggers). Before, this product would take over 2 hours to build. What we have done (not sure if this is the best way), is to build individual components of this product (only what has changed) and package it using the existing other components to create the whole product.

These separate component builds are run in parallel, so anytime a developer commits to one component, it will generate a new version of that component without having to wait for the previous build to finish. The latest of each component is then packaged together.

Now some details on how we are doing this:

We are using Jenkins pipelines, one pipeline for each component. The commit trigger is linked to GitLab. Note, we are limited to using Jenkins and GitLab due to security restrictions, access, and political reasons. To change these tools is whole other discussion.

It is a .NET build, using visual studio so we have multiple Windows VMs on a server waiting idle for when it is needed to do a build. When a commit happens, one VM is picked to build and generate artifacts. After a component is built successfully and unit tests are run, it is put into Artifactory, replacing the previous version.

Goal of this was to reduce the time it took to build so that the developers can be aware if their latest commit broke the build or not.

Looking for suggestions on improvements and changes we can do while this project is in the early stages. It would be nice to reduce the number of pipelines and the number of VMs.

https://redd.it/ej3mc2
@r_devops

Guidance on getting into DevOps from Sys Admin

Hello,

I've been reading around and flopping back and forth between different certs every time I stumble on new information. I'm a bit lost on what to ultimately pursue and wanted to ask for your advice on how to proceed.

Some background on my current experience. I've worked a cumulative of 3 years doing help desk, PowerShell scripting, O365 management, SharePoint administration, and a little bit of server work. The places I worked at primarily employed Windows environments. I've worked a little bit with Linux in an academic environment for programming classes during college.

I'm slowly being given tasks to prepare me for a DevOps role at my current job. We use an Azure environment, so I'm being shown how are VM web servers and databases are being built as well as some of the processes that we employ such as database refreshes. Before being offered this direction an opportunity, I was studying the AWS Solutions Architect course on CloudGuru. I've taken a pause and switched to studying Developing Solutions for Microsoft Azure (AZ-203) on Pluralsight. I know I'm going to be staying at this current company for another 18-20 months because I'll be moving afterwards, so that's why I'm hesitant on my choices because I feel that AWS, Kubernetes, and the typical tools listed on most resources is where I should be studying, but my current work doesn't employ any of that. I'm going to be moving to the Bay Area in Northern California, so I want to prepare as best I can to enter that tech hub.

Any advice is much appreciated. Thank you!

https://redd.it/ej6ja2
@r_devops

Coding certifications for Devops?

I work on a small team of devops engineers some of whom are just beginning their careers in the devops field. One thing that I feel very strongly about is that in order to succeed in this field, you need to be able to code. We have been doing team mentoring on this subject but I was wondering if there are any good formal training/certification courses that folks here would recommend concerning coding. Not coding with the goal of being a full on developer but to be able to write scripts and basic coding algorithms for CI/CD pipelines and to be able to stitch together services that we use.

https://redd.it/eiwhk0
@r_devops

CircleCI New UI

Anybody know what's going on with CircleCI's "new" UI?!

It seems to be a half-baked rewrite in React that they are trying to force down everybody's throat now. Been in progress for a long time but there is still tons of missing information and features. It only applies to a small section of the site as well; going to org settings or insights switches back to the "old" experience.

A new post on the forum indicates they won't be spending time fixing bugs in the legacy, feature complete, UI in favor of allocating all resources on the new UI. The way everything is going down makes it feel very political even from the outside. I can't imagine it's a very popular project externally or internally.

The user hostility in the way they are now trying to get people to use it is concerning if it's indicative of the future IMHO.

https://redd.it/ej1rnz
@r_devops

AWX Linode API error

I'm getting this error when trying to run a simple playbook in awx.

>ERROR! the Linode dynamic inventory plugin requires linode\_api4.

the play is nothing special just runs apt or yum update against some VPS hosts

`---`
`- hosts: all`
`tasks:`
`- name: update packages`
`apt:`
`name: "*"`
`state: latest`
`update_cache: yes`
`when: ansible_distribution == 'Debian' or ansible_distribution == 'Ubuntu'`

`- name: update yum`
`yum:`
`name: "*"`
`state: latest`
`update_cache: yes`
`when: ansible_distribution == 'CentOS' or ansible_distribution == 'Red Hat Enterprise Linux'`
Has anyone seen this b4? I haven't been able to find answers anywhere, I even went as far as installing the linode\_api4 pip module

https://redd.it/eiypzo
@r_devops

Repeat builds for Build Validation and on merge

So we're got Build Validation on to ensure our things build before the PR can be completed.

We've also got builds triggered on commit to our main branch, which then gets deployed to our QA server.

The problem is when someone puts in a PR, it builds to verify that it's okay to merge, then once merged, it builds the same thing over again, wasting time. Sometimes it's needed if there's multiple PRs (validations last for 12 hours), but there are often times when it's not. Is there a way to tell it that it's building the same thing again and cut down on this duplication of work?

https://redd.it/ej9eqm
@r_devops

Does your company use a geolocation service?

We are looking to evaluate geolocation services to replace a legacy hosted third party service and was wondering what others may have experience with.

https://redd.it/ej8jwd
@r_devops

Issue with spinnaker and automated trigger

I am trying to get Spinnaker to have a pipeline use an automated trigger which is an image pushed to an ECR repo. I can add different repos fine but can't seem to see the images. I am following this guide and below is the example code I followed plus a second variation I tried.

​

ADDRESS=123456789123.dkr.ecr.us-west-2.amazonaws.com
REGION=us-west-2


hal config provider docker-registry account add my-ecr-registry \
--address $ADDRESS \
--username AWS \
--password-command "aws --region $REGION ecr get-authorization-token --output text --query 'authorizationData[].authorizationToken' | base64 -d | sed 's/^AWS://'"

hal deploy apply

Variation I tried

ADDRESS=123456789123.dkr.ecr.us-west-2.amazonaws.com/api

​

Currently I can see the repo in registry name then I load the image section and nothing shows up. I have checked many times and I do have images in the repo - I've even deployed them to my kubernetes cluster and tested they work when building my helm charts.

​

What am I missing or can I try?

https://redd.it/ej7wj1
@r_devops

Tools for Server Monitoring

Any recommendations for tools for basic server monitoring, paid or free? I'm looking for something that will show me storage capacity remaining, memory utilization, CPU utilization, server up/down. Ideally I can monitor multiple servers, and have alerts for when thresholds are crossed, like storage capacity at 95%.

https://redd.it/ej7b1f
@r_devops

A php based newsletter tool that supports template emails, dynamically inserted data, send rate limits and campaigns?

Good day everyone.

I'm looking for possible solutions / options - one time paid or free (maybe subscription based) that offers a complete tool for sending out emails with the following features:

1) Sending rate limits to accommodate for allotment quita by your SMTP server provider.

2) Creation of Email Templates

3) Creation of campaigns (Sending the same email to all your subscribers)

4) Dynamically insert content into the page like thumbnails, titles, content, links.

​

Now I don't expect all of this to be GUI based, API calls are just fine as well. I just want to see if there's something already out there before I start creating my own with PHP, Cron and mysql.

https://redd.it/ej1dfc
@r_devops

What certification looks good on a devops resume ? AWS SA professional, AWS devops or CKA

I have 2.5 years of ops experience and I'm trying to move to a devops role. I already have aws SA associate certificate and looking for the next certificate. I'm aware that certifications are only used to screen resumes by recruiters and their value is not much as it used to be. My employer is ready to pay for certification cost. Which one is worth spending the time on and which has a good value?

Thanks

https://redd.it/ej1c92
@r_devops

Editing Kops masters via the cli

I was wondering if it was possible to edit the number of kops masters and nodes via the cli instead of using the kops edit ig <group name> where it brings up a text editor. I'd like this so I can script scaling the cluster both for the masters and the nodes, idea is I want to cut costs on a cluster that is for dev reasons only and doesn't get used often but can't rebuild cluster because of other tooling needing to connect to the same cluster.

https://redd.it/ej010s
@r_devops

Best Certification Prep classes to recommend on the Company's dollar?

My company was recently acquired, and the parent company until that point had no well defined DevOps team concept. The company has since gathered the DevOps people from it's acquired companies, promoted us and solidified a Site Reliability team. Now they want to get us certifications that support our new role.

Currently, the company pays for Pluralsight, which is great for skills training, but for exam and cert prep seems very sparse.

What courses of study would you recommend for the following subjects:

* AWS Certified DevOps Engineer
* Microsoft Sysadmin

In addition to actual classes that will get us ready for certification exams, I'm interested in getting us books to do for a book club with my teammates, and adding those to the proposal. Thanks for any recommendations!

https://redd.it/eizmwt
@r_devops

Web Dashboard for Process Scheduling

Hi Everyone,

I am building a system (largely in python) that will need to start, stop, and moderate processes periodically.

&#x200B;

The use case could be:

\- I have a script I want to run everyday at 12 AM

\- I have a executable I want to run constantly (so in this case should just start the executable on bootup and have the executable infinitely run)

\- I have a process I want started at 10am and terminated at 2pm.. daily

&#x200B;

In the past I have used Jenkin's for task scheduling like this. I liked how you could see the progress of each build. However I've been told to not use jenkins for long running processes. I just figure there has to be something out there like this... that has Web Dashboard access.

https://redd.it/eizhkw
@r_devops

DevOps Blogs + Reddit

Sadly I don't have time to read all of these. Which 3 do you recommend? Already reading AWS, CloudTech News & following most of these on Twitter.

https://redd.it/eiw9fy
@r_devops

Consultant and company rates difference

Hi,

I work as a devops consultant with 6 years of experience and I started thinking about the difference in my md rates vs rates of a company that does the same.

Usually the companies that have several consultants and several references charge way more than my md rate, lets say double the amount. I find it weird, as I have worked fir these companies in the past and there wasn't much of a difference in the delivered services between theirs and mine.
Still, I might be a no-name to the client even though I do have good references if my own, so I charge less.

Is there a coeficient of how much less should I as a freelancer charge in comparison to companies that do the same of the same quality and experience (basically everything the same)? Because if it is the same, why exactly should I get paid less (I don't want to sound greedy, I just don't know).

Could you please advise me?

https://redd.it/ejfi2q
@r_devops

Book recommendations on Architechting for deployability

As per Nicole Forsgren and Jez Humble’s Accelerate. The biggest factor that impacts the deployability of a system (measured by how often it is deployed) Is it's architecture and I really do agree with it (reading it was line was one of those “aha!” momments as I’ve been repeating the same thing for years). Now comes the tough part, how can i learn to architect for deployability? I mean i think I know how to do it, I’ve been “doing” it, but I also feel that it may not be the best way to do it and my method is absurdly tainted by my biases. Is there any book or training or articles or case studies you would recommend to learn how to do this from other people? Like i feel Im versed on event-driven archs and kafka, but im sure there is a lot more things out there.

Thanks!

https://redd.it/ejgyqk
@r_devops