Kimserey's blog

New contents are now posted to my personal website https://www.kimsereylam.com
Enjoy!

Load test your API with VegetaVegeta is a open source HTTP load testing tool. Today I’ll demonstrate how quickly and easily we can load test our API endpoint using it in three parts:Get VegetaSetup a target fileGenerate reports1. Get VegetaVegeta binaries are available on GitHub Releases. For Windows, all we need to do is to get the Windows executable and unzip it for example under C:\vegeta.
The vegeta.exe is the executable we will be using. To make sure it works as expected, we can display the usage guide by execute vegeta.exe without any arguments.> vegeta.exe Usage: vegeta [global flags] <command> [command flags] global flags: -cpus int Number of CPUs to use (default 4) -profile string Enable profiling of [cpu, heap] -version Print version and exit attack command: -body string Requests body file -cert string ... The main concept of Vegeta are the targets. A target represents an endpoint which will be load tested - or in Vegeta …

Create React App with Mobx in TypescriptI have been using state management frameworks for the past few years, mainly with Angular and NGRX. Today we will see how we can get started with Create-React-App using Mobx as a state management system.Bootstrap a fresh applicationCreate componentsCreate an observable state / storeCreate observers components1. Bootstrap a fresh applicationStart by installing the latest version of NPM then use npx to setup a fresh project.npx create-react-app my-app --typescript Then navigate to /my-app and run npm start. This will start the application in a development server with a live reload. The command run by npm start is defined under package.jsonscripts > start and runs react-scripts start.cd my-app npm start The application should now build and run properly and any changes done on the application should be reflected on the browser. We now have all the necessary tool to start writing code in React.2. Create componentsIn React with Typescript, creating…

Setup HTTPS with Nginx on Azure Ubuntu VMToday we will see how we can setup HTTPS on using Certbot Nginx configuration on an Azure Ubuntu VM. This post will be composed of three steps:Prepare the VMInstall NginxInstall Certbot1. Prepare the VMWe start first by creating an Azure VM on Ubuntu 18.04 with either password or SSH and allowing HTTP, HTTPS, SSH.Once done, we can select a custom DNS for our VM. This makes it easier to SSH but also it will be required for our SSL certificate setup.We set the Assignment as Static then we choose a DNS name label. Here we choose azure-test-vm therefore the VM will be accessible at azure-test-vm.southcentralus.cloudapp.azure.com.We should now be able to SSH into the VM using the command:ssh kimserey@azure-test-vm.southcentralus.cloudapp.azure.com 2. Install NginxNext once we are in the VM, we can install Nginx by installing the following:sudo apt-get update sudo apt-get install nginx Once installed, as we already have opened the HTTP port during cr…

Moving from chaining to piping in rxjs 6.xLast month I updated all my NPM packages and realised that everything was broken! The reason why was that rxjs decided to move out of extensions on observable in favor of pipable functions from 5.x to 6.x. A major breaking change where majority if not all codebase built using rxjs needs to be changed.
Today we will explore the reasoning behind this change.ReasonsExample of migrationLearnings1. ReasonsThe reasons given on the official documentation are:Any library importing the patch operator augmenting the observable prototype would also augment it for all consumers. For example, when a library imports map extension from import 'rxjs/add/operators/map', it makes map available for all code importing the library itself. This could bring confusion due to the inconsistent way of importing extensions. Some places would not need to important map (those were the library is used) while in other place of the code we would have to else the compi…

Entity Framework Core Performance OptimizationLast year I talked about Entity Framework Core. It is a easy and feature rich ORM which makes working with database in a .NET environment typesafe. But even though it makes things easy, there are ambiguous cases which can take us off guard. Today we will see four of this cases and how to deal with them.Client evaluationIterationInclude and ThenIncludeNoTrackingFor the following examples, I will be using SQLite with Entity Framework Core.tl;drMake sure that the query constructed in c# uses function that can be translated to SQL,Make sure that there isn’t an abnormal amount of queries created and that it does not iter item per item,Make sure to use Include and ThenInclude for object relation to include them after query execution, before query execution it is not needed,Use NoTracking for readonly queries to disable tracking on entity to yield better performance.1. Client evaluationThe following example illustrates the first case, client eval…

Create a Navigation loading bar for Angular with PrimeNGIn Angular, it is common practice to execute commands prior routing a page using guards and resolvers. Guards prevent the routing from occuring until the condition turns to true and resolvers prevent the routing from occuring until the data is returned. Those actions can take time to complete and during the time being, the component will not load leaving the user with an impression of unresponsiveness. Today we will see how we can implement a navigation loading bar for Angular using PrimeNG progress bar component in two parts:Setup an Angular projectPrimeNG Progress barIf you are unfamiliar with the Angular router, you can have a look at my previous blog post explaining the feature of the router https://kimsereyblog.blogspot.com/2017/06/how-to-use-angular-router.html.1. Setup an Angular projectWe start by creating a project and installing PrimeNG.npm install primeng --save npm install primeicons --save Next we also install Bootst…

Monitor Upstream Response Time with Nginx and CloudWatchLast week we saw how we could Setup CloudWatch to push logs from our application to CloudWatch. Apart from the application logs, another type of logs that is worth looking into are the access logs from Nginx. Nginx being the entrypoint of the application, every traffic in and out goes through it and today we will see how we can leverage its access logs to monitor the response time of our application from CloudWatch in three parts:Setup Nginx to log upstream response timeSetup logs to be pushed to CloudWatchSetup metrics filter on CloudWatch1. Setup Nginx to log upstream response timeBy default Nginx does not log the upstream response time. In order to print it, we can create a new log format which includes it and configure logs to be written into a file using that format. We can add our log configuration in a file called log_format.conf which we can place under /etc/nginx/conf.d/ and nginx will include the configuration under htt…

Serilog with AWS Cloudwatch on UbuntuFew weeks ago we saw How to configure Serilog to work with different environment. At the end of the post, we saw briefly how to get the structured logs synced to Cloudwatch. Today we will explore the configuration in more details.Unified Cloudwatch agentLiterate and json logs with SerilogDebug the Cloudwatch agent1. Unified Cloudwatch agentThe Unified Cloudwatch agent can be installed by following the official documentation https://docs.aws.amazon.com/AmazonCloudWatch/latest/logs/UseCloudWatchUnifiedAgent.html. There is a previous version of the Cloudwatch agent, this new version, introduced in December 2017, unifies the collection of metrics and logs for Cloudwatch under the same configuration.To install the agent, execute the following commands:mkdir ~/tmp cd tmp wget https://s3.amazonaws.com/amazoncloudwatch-agent/linux/amd64/latest/AmazonCloudWatchAgent.zip unzip AmazonCloudWatchAgent.zip sudo install.sh This will install the agent in the /opt/…

Basic Authentication with NginxBasic authentication provides an easy way to password protect an endpoint on our server. Today we will see how we can create a password file and use it to enable basic authentication on Nginx.Create a passwordEnable basic authenticationReuse configurationIf you haven’t seen Nginx before, you can have a look at my previous blog post on Getting started with Nginx1. Create a passwordTo create a password, we will use the apache2-utils tool called htpasswd.sudo apt-get update sudo apt-get install apache2-utils htpasswd allows use to create passwords encrypted stored in a filesudo htpasswd -c /etc/myapp/.htpasswd user1 -c is used to create the file. If we want to add other users we can omit the parameter.sudo htpasswd /etc/myapp/.htpasswd user2 Now if we navigate to /etc/myapp/ we will be able to find our .htpasswd file containing encrypted passwords together with usernames.2. Enable basic authenticationEnabling basic authentication can be achieved by using th…

Gitlab CI/CD with pipeline, artifacts and environmentsAlmost a year ago I wrote about how we could setup CI/CD with gitlab pipeline. I showed a very simple 3 stages pipeline build/test/deploy. Since then Gitlab has improved considerably their CI tool with features simplifying releases management. Today we will revisit pipelines and introduce few concepts which will help in managing releases.PipelineReleasesArtifactsEnvironments1. PipelinePipeline are defines as jobs. Each job can be part of a stage in the pipeline and multiple jobs can run concurrently if part of the same stage.The pipeline is define in a .gitlab-ci.yml file placed at the root of the application.We can setup our own runner or use a shared runner from Gitlab. The shared runner runs on Docker therefore it’s possible to build the dotnet image and build our dotnet application.Here is the pipeline we will be using as example pipeline:image : microsoft/dotnet:latest stages: - build - test - package - deploy build:…

Setup a Jenkins Pipeline for local development environment in Docker containerCI/CD pipelines allow us to automatically build, test and deploy code changes. With Jenkins pipeline, the pipeline itself is generated from a file called the Jenkinsfile which, usually, is source controlled together with the source code repository. When we need to push new changes, what we would usually do, is test locally and then commit to the repository. From there, the Jenkins pipeline will trigger and build, test on the integration server and deploy to a testable environment (DEV/QA). But what do we do when the changes that we are making are on the Jenkinsfile itself? How do we test locally the validity of the Jenkinsfile or more simply, how do we try on a sandbox a Jenkins pipeline to learn how to write a Jenkinsfile? Today we will see how we can setup a sandbox with a full CI/CD deployment which can be quickly brought up and teared down for testing.Jenkins server via dockerJenkins pipelineSimulate dep…