Kun 1ambda

standalone-kubelet-with-containerd.md

가끔 kubernetes 같은건 너무 무거워서 단일 호스트에 docker만 띄워서 운영하는 경우같은걸 고민 하게 됩니다. 하지만 그냥 docker만 가지고 사용하면 컨테이너가 죽었을때 관리해 줘야 하는 불편함과 같이 기능을 보다 더 필요로 하게 됩니다. 그래서 이 방법을 소개해 드립니다.

이건 standalone kubelet 이라고 불리는 master 없이 node만 관리해주는 kubelet을 이용해서 container를 관리하는 방법입니다. 이건 k8s 초기 부터 원래 존재했던 컨셉이나 많이 알려져 있지 않아서 모르시는 분들이 많아서 이글을 작성해 봤습니다.

이 방법을 이용하면 실제 master 없이 kubelet 만으로 서비스를 할 수 있습니다. 컨테이너 자동 재시작, port forwarding, resource limit 등 기본적인 kubelet에서 제공하는 다양한 방법으로 컨테이너들을 관리 할 수 있습니다.

React context vs Redux in 2020

The [React docs][condoc] give some example use cases for context:

Context is designed to share data that can be considered “global” for a tree of React components, such as the current authenticated user, theme, or preferred language.

The common property of these use cases is that data like the current theme doesn't change often and needs to be shared deep down the component tree, which would be cumbersome with "[prop drilling][drill]". Something else that needs to be shared everywhere is the application state when using a "single source of truth" pattern, so it would follow that the context API would help with that as well, but there's a catch: components that use context will rerender every time that the provided value changes, so sharing the whole application state through context would cause excessive render lifecycles.

marp

Marp

Markdown presentation ecosystem

이 글의 코드는 저자의 Github에서 확인할 수 있습니다.

때늦은 React Hooks 시리즈 1탄 - 개요/useState

나온지가 언제인데 지금에서야 이걸...?

그건 제가 이제서야 이걸 알았기 때문입니다. 늦게라도 알았으니 시작해야죠 별수 있나요.

	package com.spring.coroutine

	import io.r2dbc.spi.ConnectionFactory
	import org.springframework.boot.autoconfigure.SpringBootApplication
	import org.springframework.boot.runApplication
	import org.springframework.context.support.beans
	import org.springframework.core.io.ClassPathResource
	import org.springframework.data.annotation.Id
	import org.springframework.data.r2dbc.connectionfactory.init.CompositeDatabasePopulator
	import org.springframework.data.r2dbc.connectionfactory.init.ConnectionFactoryInitializer

	import Document, { Head, Main, NextScript } from 'next/document'

	import { renderToString } from 'react-dom/server'
	import { extractCritical } from 'emotion-server'
	import { ServerStyleSheet } from 'styled-components'
	import flush from 'styled-jsx/server'

	export default class MyDocument extends Document {
	static async getInitialProps(ctx) {
	const styledComponentsSheet = new ServerStyleSheet()

	#! /usr/bin/env bash

	# If we have an error, bail out!
	set -e

	# If we're debugging, be really loudly verbose
	[[ ! -z $DEBUG ]] && set -x

	# Attempt to load up the /etc/lsb-release file to determine our OS. If none is
	# found, we default our settings to RHEL/Amazon Linux as the original behavior

	$ docker run k8s.gcr.io/cluster-autoscaler:v1.2.2 /cluster-autoscaler -h
	Usage of /cluster-autoscaler:
	pflag: help requested
	--address string The address to expose prometheus metrics. (default ":8085")
	--alsologtostderr log to standard error as well as files
	--application-metrics-count-limit int Max number of application metrics to store (per container) (default 100)
	--azure-container-registry-config string Path to the file containing Azure container registry configuration information.
	--balance-similar-node-groups Detect similar node groups and balance the number of nodes between them
	--boot-id-file string Comma-separated list of files to check for boot-id. Use the first one that exists. (default "/proc/sys/kernel/random/boot_id"

	import React, {Component} from 'react'
	import Router from 'next/router'
	import AuthService from './AuthService'

	export default function withAuth(AuthComponent) {
	const Auth = new AuthService('http://localhost')
	return class Authenticated extends Component {

	static async getInitialProps(ctx) {
	// Ensures material-ui renders the correct css prefixes server-side

	dfConverted3 = dfConverted1\
	.selectExpr("*",
	f"CASE WHEN education == '{educationInvalid}' THEN '{educationDefault}' ELSE education END as education"
	)