#221 — June 19, 2019

Read on the Web

Web Operations Weekly

..soon to be Statuscode Weekly

HAProxy 2.0 Released — I have a huge soft spot for HAProxy, so this is really exciting. HAProxy is a powerful and easily configurable TCP and HTTP load balancer that you can put in front of almost anything when wiring up Web infrastructure. HAProxy 2.0 features:

  • end-to-end HTTP/2 support
  • full gRPC support
  • a ‘Data Plane API’ for on-the-fly configuration
  • traffic shadowing
  • native support for providing metrics to Prometheus
  • end-to-end TCP Fast Open support

Daniel Corbett

▶  Demystifying Web Speed Tooling — Paul Irish and Elizabeth Sweeny of Google gave a talk at Google I/O that dug deep into the world of user experience, Web performance, and the tools available to improve matters on your own site.

Google I/O ’19

Powerful CI/CD Pipelines. Pay Only for What You Use — Built for developer productivity, Semaphore 2.0 allows you to model any CI/CD workflow with fully customizable, powerful pipelines. With the new pricing model you pay only for what you use. Sign up free with GitHub and give it a run-through.

Semaphore 2.0 sponsor

Docker ❤️ WSL 2: The Future of Docker Desktop for Windows — Microsoft has announced a major new version of their Linux compatability layer, WSL. WSL 2 will eschew emulation and actually provide a real Linux kernel which means Docker will be able to integrate much better with Windows.

Simon Ferquel

Some Digital Ocean and AWS IPs Banned in Russia? — A developer noticed that his blog was inaccessible from Russia and used the ‘Is It Blocked In Russia’ tool to confirm it (it seems 206.189.0.0/16 is blocked). It might be worth trying on your own sites.

Ibrahim Diallo

DNS Cookies: Using DNS Caches to Identify Related Network Flows — An interesting side-channel to be aware of.

Daniel Dent

Quick bytes:

💻 Jobs

Senior Node.js + DevOps Engineer - (Remote) — Engineers wanted to deploy Node.js applications that depend upon individual cryptocurrency daemons using AWS servers.

Exodus

Land a New Dev Job on Vettery — Vettery specializes in tech roles and is completely free for job seekers.

Vettery

📖 Tutorials & Stories

Teaching A Cheap Ethernet Switch New Tricks — An interesting story of hacking a cheap router that involves running Go code on it. If you’re a networking/hardware geek who loves a good story, this is for you.

Ben Cox

▶  Scaling an Erlang Cluster to 10,000 Nodes — Maxim Fedorov demonstrates an example of a live Erlang cluster being scaled from just a few nodes to 10,000 machines with no service interruption.

InfoQ

How to Deploy a Resilient Go Application to DigitalOcean Kubernetes — Learn how to deploy a dockerized Go application to a Kubernetes cluster in this step-by-step tutorial.

DigitalOcean sponsor

Alerting on SLOs like Pros — SLOs, alerting, and error budgets are central concepts of Site Reliability Engineering (SRE).

Björn Rabenstein (SoundCloud)

Server-Sent Events (SSE): A Conceptual Deep Dive — SSE is a lightweight, subscription-based protocol for event-driven data streams that’s supported by most major browsers (Edge will gain support once the switch to the Chromium engine is complete).

Ably Realtime

Web Single Sign-On, the SAML 2.0 Perspective — SAML is a standard for different parties to trade authorization data so that a user could, say, sign in with a single provider but then access numerous other services.

Guillaume Klaus

Seven Surprising Bash Variables — The TMOUT one was a real eye opener to me.

Ian Miell

'Why Google Refunded $200 Because I Missed 5 Lines of Code' — A developer forgot to acknowledge payments in his app. Oops. Always good to hear a nice customer support story though.

Aswin Mohan

9 SRE Best Practices for Incident Management

Gremlin sponsor

How Spotify Paints a Picture of Its Infrastructure in Minutes — A quick look at how Spotify has been able to create graphs of its Google Cloud-based infrastructure by using Cloud Asset Inventory and Neo4j, the graph database.

Carly Schneider (Spotify Labs)

What Redditors Think About The State of GraphQL

Robert Matyszewski

▶  Bringing Scalable Real-Time Analytics to The Enterprise — A 35 minute podcast interview with two guys from Rockset, a company working on low latency big data analytics.

O'Reilly Data Show podcast

🛠 Code & Tools

Pika CDN: A CDN for Modern JavaScript — An interesting proxy that serves up JavaScript packages that use the ES Module syntax so you can use them directly on the Web using import.

Pika

API Platform: A REST and GraphQL Framework on Top of Symfony and React

Kévin Dunglas

macOS-Simple-KVM: Tools to Set Up a Quick macOS VM in QEMU — Running Linux but got some testing to do on macOS..?

Foxlet

YIG: A New Object Storage Server with an S3 Compatible API

Journey Midnight

Clojerl: Clojure, but for the Erlang VM — Clojure is a popular Lisp dialect but it’s (mostly) limited to the JVM. This project ports it to Erlang’s famously stable and scalable VM.

Clojerl