The great mystery on how to make durable, asynchronous functions for asio.
Many C++ developers seems confused about the secret sauce of asynchronous completions - or composed operations - in Boost.asio.
Software & Coffee at the Edge of the Universe
Ramblings from the Viking who run the Cafe at the end of the universe.
Hi, my name is Jarle. I am a Freelance C++ developer. I have spent quality time with code for 35+ years, and I still love it!
This is my professional blog, where I write about stuff I research or discover during my daily work on various projects.
Many C++ developers seems confused about the secret sauce of asynchronous completions - or composed operations - in Boost.asio.
Some times, especially during development, I need to connect to lot's of k8s pods using command-line curl (or other tools I have on my PC). The traditional solutions is to install a "jump-pod", and execute the commands there. Today I had enough and installed a socks proxy pod in the k8s cluster.
It is surprisingly hard to deploy a private docker registry anywhere, and kubernetes is no exception.
For C++ developers, more cores to build on means faster builds. If your project is big, more cores can make a significant boost. If you have a local k8 cluster with spare CPU, it's amazingly simple to turn it into a build-machine.
My experience and spells to make the Lenovo Legion Y730 work well on Debian Buster and Ubuntu 18.4 and 18.10
What does it take to make a simple Qt/QML application and deploy it on iOS, Android, macos, Linux, Windows? Let's find out!
Latest articles
How do you ensure that, for example, a database wrapper, only hands out one instance of any given data object?
How to install Docker Swarm on a Linux machine in a minute.
How do we call a C++ std::function<> based callback from Android, in the simplest possible way?
A very naive Ansible Playbook to install Docker and Docker Compose on CentOS.
Some times it's required to copy a docker container between machines, or even developers. One option is to use a hosted repository and pay some money for this opportunity. Another option is to use a machine that has a reachable IP address and Docker to run a private and secure Docker Registry.
Building something in Jenkins, using a declarative pipeline and deploying it to a Docker container is common, and you would expect it to be well documented and trivial to do right? Well, it is now.
C++ is a popular Object-oriented language. Json is a popular format for storage of Objects. Why is it so hard to make those two things play ball?