z2017 Session Incubation
Got an idea or want to contribute to someone else's? Put it on this page, and put your session at the top of the following list, separated by horizontal bars (use --- in Markdown).
Hosted by Your Name
In this session we will be working on some amazing things. And talking a lot. You should be there.
Interested: A Pretty Cool Person, And Another, …
Hosted by Franzi
I want to set up notifications of new meetups on the Queer Code London Slack and found this Heroku app: https://github.com/jeffcampbell/meetup-slack
I've deployed Heroku before but it might be nice for people to see how that works so they can deploy their own apps.
I'll do a live configuration and deployment while trying to encrypt the secrets on Heroku.
Interested:
Hosted by Esko Luontola
With the current trend of implementing much of web site features on the client side, I've found many sites to be easily hackable with just Chrome DevTools. In this session I'll show a couple examples of live sites that send to the browser the data which they are selling or which they don't want others to see. I'll share some DevTools tricks for debugging and modifying the JavaScript and hopefully others also have tricks to share.
Interested:
- ?
Hosted by Zsolt Sz. Sztupák
A short, hands-on tutorial to show how the various pieces of a cloud infrastructure fit together, and how someone can set up a full system with continuous deployment capabilities with open source tools in just a few hours.
This is a half-finished tutorial you can access from here showing you how one could deploy a three-tiered app onto a single Digital Ocean droplet, and then continuing from there how to modify the infrastructure to be highly available, and resilient to outages. Also since we’re deploying our app to the public internet, we take the necessary steps to make sure our infrastructure is as secure as possible from the start. We’ll be using Digital Ocean as the cloud provider, terraform as a tool to provision our cloud, and kubernetes to manage our cloud and orchestrate the containers of our application.
It would be good to gather and try to finish the tutorial later to include a fully working CD pipeline as well, that would mean we went from Zero to CD
Interested:
- ?
Hosted by Zsolt Sz. Sztupák
While most[citation needed] agree that introducing more and more automation, Continous Delivery and other DevOps buzzwords can help companies overcome difficulties in delivering clean and maintainable code and a consistent release experience. However sometimes it's hard to start on this road, especially in larger companies, with lots of legacy baggage to hold onto. While there are usually a fair amount of good tutorials on how to use technologies like docker, puppet, ansible, etc. what is occasionally missing is how you can introduce them inside your company, and how to get the buy-in from everyone in using them.
A few years ago we started a journey on modernising our platform. While the road taken was a long one, filled with bumps, potholes and landmines, in retrospective it did fulfill most of its promises. In this talk I'm going to talk about some of the decisions we've made, including how we managed to push our ideas further to other teams. However we still believe we could have done better in some regards, and we constantly try to improve upon our processes. I'd also like to start a discussion on what other people's experience is in transforming their company, what they believe was a good idea, and they believe is probably an antipattern in moving onto DevOps.
Interested:
- ?
Hosted by Markus Deibel
Alpha test of my planned XPDays talk
We all know them: "Legacy" systems. We don't like them, because we'll get our hands dirty when touching them. Worse, sometime they react to small changes like porcelain to an elephant. And then we receive the long dreaded task of extending one of these systems.
But where to start? How can we get closer to a solution that is within a reasonable time and with reasonable effort? How can we implement a solution that is sufficiently robust? And will my colleague understand what I checked in when she returns from her holidays three weeks later? Oracles, TDD refactoring and patterns are a few examples of the toolkit I used for implementing a "small" requirement
I'll show - using production code - in which ways these tools can/must interact to allow someone who is new to a codebase making changes without too much risk. Based on code examples and a step-by-step description of the implementation I also want to show that is important having seen tools work in a certain context to make good use of them in your own. As an example I'll extend a production software with a (simple) scheduler.
Target audience
Software developers quite new to the job with little experience in large code bases, devs who inherited a codebase new to them and those who want to make the first small steps to move their code away from "legacy".
Hosted by David Heath and Raul Garcia
How is your CI pipeline configured? What works well? What doesn't work so well? What improvements have you made over time? We'll tell you about ours if you tell us about yours :-)
In particular, we're interested in making CI/CD pipelines work well for microservices, how it fits with deployment, application monitoring etc.
We'd love to talk about specific problems and compare solutions. One issue which we're actively working on right now is trying to move from a large end-to-end test suite to more focused application level integration tests and unit tests (aka rebalancing the testing pyramid).
We also have a challenge with coordinating changes to individual applications with changes to the end-to-end suite.
Are you interested? Write your name/comments/suggestions:
- Zsolt Sz. Sztupak
Place for your idea.