Skip to main content
  1. Article Categories/

Kubernetes

Trying Distributed Tracing with Jaeger

··1467 words·7 mins
Let’s say we have a set of applications that was designed to be a set of microservices. Each of the applications would generally be designed to be focused on one specific domain and in order to achieve the overall goal of the platform. However,for the platform to work properly, the applications would generally need to work together as one which would involve the application contacting each other.

Attempting to setup Kubernetes on Ubuntu VMs

··906 words·5 mins
This post details my naive attempt to bring up a Kubernetes cluster on a VM. These steps to try out Kubernetes in a bare Google Virtual Machine (but the following steps should work for most Debian/Ubuntu virtual machines). This deploys a single node Kubernetes cluster (naturally don’t think of using this for production)

Trying out skaffold

··497 words·3 mins
When developing application that are meant to be deployed to the Kubernetes platform, it involves a bunch of steps on top of your usual local development work: Writing a Dockerfile to package the application (Multi stage applications are optional here - useful for compiled based languages) Build and tagging the docker image of the application with the target repository Either use kubectl commands or use kubernetes config resource files to define the resources required for deploying the applications. Use those commands/configurations to define the resources on the staging/production application Repeat the process for each update of the application (Repeat second point onwards) As you see from above, it starts to be pain to do so after each iteration of the application development. The building of the docker containers process as well as the applying of the new images to each cluster, (sometimes with slightly changed configuration files) - the kubernetes secret and config files can change across different environments.