Kubernetes provides great plugin architecture. You can extend kubectl with your plugins.

If you are an author of a plugin, you need to do the following steps on each release.

1. Build the plugin.
2. Upload the binary to GitHub Releases or etc.
3. Write a plugin manifest.
4. Open a pull request to kubernetes-sigs/krew-index repository.
5. (optional) Upload a formula to the Homebrew tap repository.

How boring! You may want to automate this process.

This article assumes you write a plugin in Go and you are using GitHub and CircleCI.

Make

Let’s create a Makefile.

TARGET := kubectl-hello
VERSION := v1.0.0

(1) Build

Build the plugin for…

gradleupdate is a service which provides continuous update of Gradle in your GitHub repositories.

Recently new version of Gradle is released every 1~2 months. gradleupdate allows automated tests with the latest version by CI. It is useful for Gradle plugin authors.

Getting Started

You need to add the following badge to README in a repository.

[![Gradle Status](https://gradleupdate.appspot.com/YOUR/REPO/status.svg)](https://gradleupdate.appspot.com/YOUR/REPO/status)

Here is an example of badge.

And then gradleupdate will send a pull request for the latest version of Gradle wrapper if it is out-of-dated.

Here is an example of pull request.

GitHub provides GraghQL API as well as REST API. This article introduces some example queries.

You can try a query on https://developer.github.com/v4/explorer/.

Users

We can know about the current user by the following query:

{
  viewer {
    login
    name
    avatarUrl
    url
  }
}

It will return the following response:

{
  "data": {
    "viewer": {
      "login": "int128",
      "name": "Hidetake Iwata",
      "avatarUrl": "https://avatars0.githubusercontent.com/u/321266?v=4",
      "url": "https://github.com/int128"
    }
  }
}

Pull Requests

We can find pull requests by head and base branch name:

{
  repository(owner: "octocat", name: "Spoon-Knife") {
    pullRequests(states: [OPEN, CLOSED, MERGED], first: 1, headRefName: "example", baseRefName: "master") {
      nodes {
        number
      }
    }
  }
}

…

TL;DR

Kubernetes supports EBS Persistent Volumes by default. It also supports EFS Persistent Volumes by the external efs-provisioner.

This article introduces EFS Persistent Volumes (EFS PVs) and explains how we can use it and migrate to it.

EBS vs EFS

The article “When to Choose Amazon EFS” says:

Amazon EFS provides shared file storage for use with compute instances in the AWS Cloud and on-premises servers. Applications that require shared file access can use Amazon EFS for reliable file storage delivering high aggregate throughput to thousands of clients simultaneously.

Amazon EBS is a cloud block storage service that provides direct access from a single…

Atlassian JIRA and Confluence server supports single sign on (SSO) by using plugins.

JIRA and Confluence have user management and LDAP integration, but you can integrate them with an identity manager such as Keycloak. It allows that an user types the password just once and login to many sites such as GitLab or Mattermost. As well as it allows that an administrator can set a password policy.

This article introduces how to integrate JIRA and Confluence with Keycloak.

JIRA SSO

There are many SAML plugins on Atlassian Marketplace. This article introduces the following plugin:

https://marketplace.atlassian.com/plugins/com.bitium.jira.SAML2PluginJira/server

It is free and open source. The…

kops does not support changing size or type of etcd volumes after cluster creation. It is inconvenient when you created too large volumes.

If your etcd volumes are gp2 or io1 type, you can extend the volumes easily in the AWS management console.

You can shrink the volumes by the following steps:

1. Stop the Kubernetes master services.
2. Create the new volumes.
3. Copy data from old to new.
4. Add tags to the new volumes.

Steps to shrink volumes

It assumes your cluster has single master. If your cluster has multiple masters, shrink to single master in advance.

(1) Stop Kubernetes master

Connect to the master instance:

ssh -i your_ssh_key…

Atlassian JIRA Software and Confluence are a nice agile management tool. We can start using JIRA and Confluence on Atlassian Cloud in a few minutes but we still need a private server in some cases.

I just published Kubernetes Helm charts for JIRA and Confluence. You can easily deploy JIRA and Confluence on your Kubernetes cluster.

Getting Started

Install tools

Make sure you can access to the cluster using helm command.

brew install kubernetes-helm
helm init

In this article, we use Helmfile for configuration management.

curl -L -o ~/bin/helmfile https://github.com/roboll/helmfile/releases/download/v0.17.0/helmfile_darwin_amd64
chmod +x ~/bin/helmfile

Deploy JIRA

Get the repository of the chart.

git clone https://github.com/int128/devops-kompose
cd devops-kompose

…

Golang provides a simple HTTP server in http package.

Now let’s think how to shutdown the HTTP server when it receives a request to the endpoint such as http://localhost:8000/shutdown.

1. Simply shutdown HTTP server

http.Server provides Shutdown method for graceful shutdown.

We can simply shutdown the HTTP server by calling Shutdown method in the handler as follows:

func main() {
    m := http.NewServeMux()
    s := http.Server{Addr: ":8000", Handler: m}
    m.HandleFunc("/shutdown", func(w http.ResponseWriter, r *http.Request) {
        s.Shutdown(context.Background())
    })
    if err := s.ListenAndServe(); err != nil && err != http.ErrServerClosed {
        log.Fatal(err)
    }
    log.Printf("Finished")
}

Run a server and send a request to /shutdown, then the server…