Is Pulumi really open source?

Why do I need to login to use?

​

I do not want to create the account before using any open-source tool.

​

Is it really open source? Or just the CLI open-source and backend proprietary?

https://redd.it/drb96u
@r_devops

Career suggestions?

- I love coding (Python), mainly for automating things
- I really like Linux as well
- Right now the money is where the cloud is...... my concern is that I don’t want to “follow the money” and then be doing something I don’t enjoy.

https://redd.it/drcy0b
@r_devops

Best library to interact with Gitlab?

Is there any official or stable library to interact with Giltab and also with the .git directory? I need a program which will make commits in an automated way. I don't care about the programming language but I am used to use golang.

https://redd.it/dref8l
@r_devops

Jenkins: Parse HTML report and fail build based on criteria

I currently have some security scans within my pipeline which generates a HTML report, the report is published using the jenkins plugin 'publishHTML' . This report indicated the severity level of the scans. What i would like to do is introduce the ability for Jenkins to parse the report and fail the build if it detects anything in the category of High or critical.

​

I believe i can do this with a bash script and grep but it seems messy. Anyone got any recommendations or similar experience.

cheers

https://redd.it/drf651
@r_devops

CI - Do I build docker container with app and scp it? Or do I build app and scp the app?

I am building a CI with Github and CircleCI.

When a developer tags a commit and pushes, CircleCI will see that it is tagged and run it's workflow. This will involve:

1. Running Ant to compile the app into the \`built\`
directory.
2. scp the contents of the newly \`built\`
directory to the production server which is /home/user/foo-repo/tomcat/ROOT.

The production server users docker-compose and has docker containers: 1) nginx 2) tomcat 3) postgres.

Am I doing this correctly by scp'ing the contents of the \`built\` directory to the production server at home/user/foo-repo/tomcat/ROOT?

Or should I be building the tomcat docker container on CircleCI, along with the app, and scp the docker container to production server?

https://redd.it/dr9yxd
@r_devops

Wikileaks architecture?

I'm not sure if this is the right sub, i want to learn how wikileaks system architecture works. Particularly the online submission platform, any resources/links to the same would help. Thanks in advance.

https://redd.it/dpzxo9
@r_devops

When will Kubecon start selling tickets for 2020 Amsterdam?



https://redd.it/drh20i
@r_devops

Docker vs VM | simply explained

Hi there!

I did a video comparing **Docker vs Virtual Machine**. 🐳

Please check it out, any feedback appreciated: [https://youtu.be/5GanJdbHlAA](https://youtu.be/5GanJdbHlAA)

Playlist from the whole Docker & Kubernetes beginner series: [https://www.youtube.com/playlist?list=PLy7NrYWoggjwPggqtFsI\_zMAwvG0SqYCb](https://www.youtube.com/playlist?list=PLy7NrYWoggjwPggqtFsI_zMAwvG0SqYCb)

https://redd.it/drivpt
@r_devops

How do people manage IP/Host black/whitelisting for logging? (AKA: how do I restrict those who can log to me)

Backstory: I'm currently managing the equivalent of an ELK stack for logging, where the Logstash is actually a Syslog-NG server, and I'd love to be able to restrict certain IPs from logging to my system.

This can be achieved (and I actually do it this way nowadays) in Syslog-ng using filters, but it doesn't scale very well, since for every new IP that I'd like to add to my white/black list I need to update the configuration and restart the services.

Besides the obvious answer of "well tell those IPs to stop logging to your stack", is there any intelligent way of addressing this issue?

Has anyone ever found themselves in a similar situation?

Thanks in advance.

https://redd.it/drlsbe
@r_devops

Cloud Engineer Interview Prep

I have an interview coming up for a role as a Cloud Systems Engineer. I'm coming from a background in Software Consulting, where I've supported both on-prem and AWS/GCP solutions, but this is the first time I've applied for a "Cloud Engineer" role.

Coming from a mostly development background, I'm used to seeing leetcode and whiteboarding. Can I expect the same or similar for a cloud engineer role?

It seems to be a young field, so there aren't much resources available online. What kind of interview questions should I expect to be asked?

https://redd.it/drhxvi
@r_devops

Bleeding edge DevOps tools?

Can you suggest some latest tools or software that would be relevant to DevOps, and seem like they will have big impact in the years to come?

Need to write a paper on the same.

https://redd.it/drkc7m
@r_devops

Terraform move from single file to modules

Recently came across a question and was wondering the correct answer to it. Let's say you have everything in terraform split between two files, a [main.tf](https://main.tf) and a [varibales.tf](https://varibales.tf) fle. I know this is not the correct way of doing it, so assuming you get given this, is there a way of moving towards modules while still keeping the exact resources that have been created already?

​

My first thought is no you can't but then it seems like a common case that this would happen and that it would make sense for this type of thing to be supported.

https://redd.it/droofp
@r_devops

How to deal with db migrations in rolling updates on managed instance groups

Hi,

i'm studying for the Google Cloud Professional Architect Certification.
For update deployments, everywhere is recommended (or mandatory) the rolling update feature (on a managed instance group, or on kubernetes pods), so some instances begins to update while others remain on the old version. This is obviously good because you have continuity and no downtime.

But, if I have, let's say, a django (or ruby) application with a managed relational database, and I need to apply a database migration (meaning a change of structure), and without it my new version breaks and with it my old version does not work, how is it possible to deal with it?

How do you keep that continuity?

https://redd.it/drl5yl
@r_devops

New DevOps platform for Game Engines!

Hello fellow humans, we’re launching our CI/CD platform that focus on the game industry get an early access right now to check it out.
[https://www.wololoci.com/](https://www.wololoci.com/)

We’re going to be interviewing at YC this winter so any feedback will be appreciated it.PD: and if you can show us some love at product hunt will be awesome

[https://www.producthunt.com/posts/wololo](https://www.producthunt.com/posts/wololo)

https://redd.it/drrvts
@r_devops

Does anyone use terraform for azure?

I know you can use Terraform to spin up different kinds of resources on Azure and I can find some info on the topic on Microsoft's website, I just don't find much on medium or on other types of blogs. Is there anywhere on here that uses Terraform to manage/ and spin up Azure infrastructure?

https://redd.it/drgjve
@r_devops

Embedded devops

I was asked to figure out a ci/cd pipeline for various embedded systems. I'm familiar with the tool chains involved however there is a huge problem that I haven't really found a good solution for.

Embedded systems need to be tested on actual hardware. This means having the board(s) plugged in.

Currently they have a few physical machine to handle this but it's not scalable and it wastes a lot of time from everybody involved.

I do have some ideas I want to try but they're kind of a long shot. If this is a solved problem I couldn't find it.

What are your experiences in this area?

https://redd.it/drehj9
@r_devops

Considering moving off Azure DevOps

For a number of years we’ve mainly been a Microsoft shop and using DevOps made sense at the time. It’s fairly easy to configure and for the most part no cost since it’s basically included with MSDN. That being said, in the last few years we’ve become a lot less Microsoft centric and considering moving to a different platform.

There are some things I don’t want to leave behind though. I really like how everything is tied together, almost like an ERP system. If you setup the boards and pipelines it’s really nice. For example, create a branch to fix a bug, associate the bug with that branch, then your commits get linked to that ticket. Later, when you merge to master and build/release, those subsequent items are linked as well. Long story short, I can create a release and very easily see what is different compared to the last one.

My question is, what is a typical setup outside DevOps that would provide that same type of functionality? We only have a handful of developers and while cost isn’t a huge factor, I really don’t want to go from paying $40/month (total) to hundreds.

https://redd.it/drw5tz
@r_devops

Issue connecting to Consul from inside container

I'm working on Dockerizing some applications and currently Consul runs locally on my laptop but eventually will be on Kubernetes. I have a Node.js application that even though I can connect and application works fine when I run the application outside the container, when I run it in a container I get this error message back

​

Error message:

Error: bad request
at create (/opt/node_app/app/node_modules/papi/lib/errors.js:14:5)
at Object.response [as Response] (/opt/node_app/app/node_modules/papi/lib/errors.js:38:15)
at IncomingMessage.<anonymous> (/opt/node_app/app/node_modules/papi/lib/client.js:592:26)
at IncomingMessage.emit (events.js:215:7)
at endReadableNT (_stream_readable.js:1183:12)
at processTicksAndRejections (internal/process/task_queues.js:80:21) {
isPapi: true,
isResponse: true,
message: 'bad request',
statusCode: 400
}

&#x200B;

Node.js code:

const CONSUL_ID = require('uuid').v4();
const ip = require('ip');
const my_IP = ip.address();
let options = {
name: service,
address: `${my_IP}`,
port: port,
id: CONSUL_ID,
check: {
ttl: '10s',
deregister_critical_service_after: '1m'
}
};
consul.agent.service.register(options, function(err) {
if (err) throw err;
console.log(`Registered service with ID of ${CONSUL_ID}`);
});

&#x200B;

I've checked and it's getting the correct IP for my laptop like it normally would if ran outside the container. It is also has the proper `service` and `CONSUL_ID` so I am good there too.

&#x200B;

Command used to start container:

docker run --net=host -p 8091:8080 -e mongo_uri=mongodb+srv://<user>:<password>@<subdomain>.mongodb.net -e CONSUL=127.0.0.1 myapp:b424cab

&#x200B;

Any ideas what might be the issue currently or suggestions on how I can make this work for both local development as well as on Kubernetes?

https://redd.it/dpvf04
@r_devops

First DevOps Interview

Hey guys. I'm having my first on site interview for a devops engineer position.

I'm surprised to have made it this far, because my experience with jenkins is basically none, and quite basic experience with Ansible. My main strengths are in programming, docker, and I guess having the mental urge to automate tasks and try new things.

That being said, what are some of the things I should look out for, or have knowledge of before hand?

I'm not expecting to get the job, however I am extremely excited to get experience and learn more through the interview.

https://redd.it/dpue2z
@r_devops

How to get listed in CNCF?

Does an early stage startup have any chance to get listed to CNCF landscape? Does it have to be a sponsor? Any recommendations on the process? Thanks.

https://redd.it/dpt697
@r_devops

generating adhoc plots

When I do system benchmarking I create stats for each time I run the benchmark/test. Is there a tool which I can post the data to to generate graphs? I prefer web based. At the moment I use python matplotlib but not all machines have particular version of Python and especially Matplot lib and its dependencies.

Ideally, Grafana would be awesome for this but I don't think I can create adhoc graphs. There is a JSON plugin but I couldn't get it to work. Are there any similar setups like that?

https://redd.it/drykec
@r_devops