Along with cp and mv, both of which we talked about at length in the previous installment of this series, links are another way of putting files and directories where you want them to be. The advantage is that links let you have one file or directory show up in several places at the same time.

As noted previously, at the physical disk level, things like files and directories don’t really exist. A filesystem conjures them up for our human convenience. But at the disk level, there is something called a partition table, which lives at the beginning of every partition, and then the data scattered over the rest of the disk.

Although there are different types of partition tables, the ones at the beginning of a partition containing your data will map where each directory and file starts and ends. The partition table acts like an index: When you load a file from your disk, your operating system looks up the entry on the table and the table says where the file starts on the disk and where it finishes. The disk header moves to the start point, reads the data until it reaches the end point and, hey presto: here’s your file.

Hard Links

A hard link is simply an entry in the partition table that points to an area on a disk that has already been assigned to a file. In other words, a hard link points to data that has already been indexed by another entry. Let’s see how this works.

Open a terminal, create a directory for tests and move into it:

mkdir test_dir

cd test_dir

Create a file by touching it:

touch test.txt

For extra excitement (?), open test.txt in a text editor and add some a few words into it.

Now make a hard link by executing:

ln test.txt hardlink_test.txt

Run ls, and you’ll see your directory now contains two files… Or so it would seem. As you read before, really what you are seeing is two names for the exact same file: hardlink_test.txt contains the same content, has not filled any more space in the disk (try with a large file to test this), and shares the same inode as test.txt:

$ ls -li *test*

16515846 -rw-r--r-- 2 paul paul 14 oct 12 09:50 hardlink_test.txt 16515846 -rw-r--r-- 2 paul paul 14 oct 12 09:50 test.txt

ls‘s -i option shows the inode number of a file. The inode is the chunk of information in the partition table that contains the location of the file or directory on the disk, the last time it was modified, and other data. If two files share the same inode, they are, to all practical effects, the same file, regardless of where they are located in the directory tree.

Fluffy Links

Soft links, also known as symlinks, are different: a soft link is really an independent file, it has its own inode and its own little slot on the disk. But it only contains a snippet of data that points the operating system to another file or directory.

You can create a soft link using ln with the -s option:

ln -s test.txt softlink_test.txt

This will create the soft link softlink_test.txt to test.txt in the current directory.

By running ls -li again, you can see the difference between the two different kinds of links:

$ ls -li

total 8 16515846 -rw-r--r-- 2 paul paul 14 oct 12 09:50 hardlink_test.txt 16515855 lrwxrwxrwx 1 paul paul 8 oct 12 09:50 softlink_test.txt -> test.txt 16515846 -rw-r--r-- 2 paul paul 14 oct 12 09:50 test.txt

hardlink_test.txt and test.txt contain some text and take up the same space *literally*. They also share the same inode number. Meanwhile, softlink_test.txt occupies much less and has a different inode number, marking it as a different file altogether. Using the ls‘s -l option also shows the file or directory your soft link points to.

Why Use Links?

They are good for applications that come with their own environment. It often happens that your Linux distro does not come with the latest version of an application you need. Take the case of the fabulous Blender 3D design software. Blender allows you to create 3D still images as well as animated films and who wouldn’t to have that on their machine? The problem is that the current version of Blender is always at least one version ahead of that found in any distribution.

Fortunately, Blender provides downloads that run out of the box. These packages come, apart from with the program itself, a complex framework of libraries and dependencies that Blender needs to work. All these bits and piece come within their own hierarchy of directories.

Every time you want to run Blender, you could cd into the folder you downloaded it to and run:

./blender

But that is inconvenient. It would be better if you could run the blender command from anywhere in your file system, as well as from your desktop command launchers.

The way to do that is to link the blender executable into a bin/ directory. On many systems, you can make the blender command available from anywhere in the file system by linking to it like this:

ln -s /path/to/blender_directory/blender /home/<username>/bin

Another case in which you will need links is for software that needs outdated libraries. If you list your /usr/lib directory with ls -l, you will see a lot of soft-linked files fly by. Take a closer look, and you will see that the links usually have similar names to the original files they are linking to. You may see libblah linking to libblah.so.2, and then, you may even notice that libblah.so.2 links in turn to libblah.so.2.1.0, the original file.

This is because applications often require older versions of alibrary than what is installed. The problem is that, even if the more modern versions are still compatible with the older versions (and usually they are), the program will bork if it doesn’t find the version it is looking for. To solve this problem distributions often create links so that the picky application believes it has found the older version, when, in reality, it has only found a link and ends up using the more up to date version of the library.

Somewhat related is what happens with programs you compile yourself from the source code. Programs you compile yourself often end up installed under /usr/local: the program itself ends up in /usr/local/bin and it looks for the libraries it needs / in the /usr/local/lib directory. But say that your new program needs libblah, but libblah lives in /usr/lib and that’s where all your other programs look for it. You can link it to /usr/local/lib by doing:

ln -s /usr/lib/libblah /usr/local/lib

Or, if you prefer, by cding into /usr/local/lib…

cd /usr/local/lib

… and then linking with:

ln -s ../lib/libblah

There are dozens more cases in which linking proves useful, and you will undoubtedly discover them as you become more proficient in using Linux, but these are the most common. Next time, we’ll look at some linking quirks you need to be aware of.

Learn more about Linux through the free “Introduction to Linux” course from The Linux Foundation and edX.

Welcome back to this series on building faster web pages. The last article talked about what you can achieve just through image compression. The example started with 1.2MB of browser fat, and reduced down to a weight of 488.9KB. That’s still not fast enough! This article continues the browser diet to lose more fat. You might think that partway through this process things are a bit crazy, but once finished, you’ll understand why.

Preparation

Once again this article starts with an analysis of the web pages. Use the built-in screenshot function of Firefox to make a screenshot of the entire page. You’ll also want to install Inkscape using sudo:

$ sudo dnf install inkscape

If you want to know how to use Inkscape, there are already several articles in Fedora Magazine. This article will only explain some basic tasks for optimizing an SVG for web use.

Analysis

Once again, this example uses the getfedora.org web page.

Getfedora page with graphics marked

This analysis is better done graphically, which is why it starts with a screenshot. The screenshot above marks all graphical elements of the page. In two cases or better in four cases, the Fedora websites team already used measures to replace images. The icons for social media are glyphs from a font and the language selector is an SVG.

There are several options for replacing:

• CSS3
• Fonts
• SVG
• HTML5 Canvas

HTML5 Canvas

Briefly, HTML5 Canvas is an HTML element that allows you to draw with the help of scripts, mostly JavaScript, although it’s not widely used yet. As you draw with the help of scripts, the element can also be animated. Some examples of what you can achieve with HTML Canvas include this triangle pattern, animated wave, and text animation. In this case, though, it seems not to be the right choice.

CSS3

With Cascading Style Sheets you can draw shapes and even animate them. CSS is often used for drawing elements like buttons. However, more complicated graphics via CSS are usually only seen in technical demonstration pages. This is because graphics are still better done visually as with coding.

Fonts

The usage of fonts for styling web pages is another way, and Fontawesome is quiet popular. For instance, you could replace the Flavor and the Spin icons with a font in this example. There is a negative side to using this method, which will be covered in the next part of this series, but it can be done easily.

SVG

This graphics format has existed for a long time and was always supposed to be used in the browser. For a long time not all browsers supported it, but that’s history. So the best way to replace pictures in this example is with SVG.

Optimizing SVG for the web

To optimize an SVG for internet use requires several steps.

SVG is an XML dialect. Components like circle, rectangle, or text paths are described with nodes. Each node is an XML element. To keep the code clean, an SVG should use as few nodes as possible.

The SVG example is a circular icon with a coffee mug on it. You have 3 options to describe it with SVG.

Circle element with the mug on top

<circle style="opacity:1;fill:#717d82;fill-opacity:1;stroke:none;stroke-width:9.51950836;stroke-linecap:round;stroke-linejoin:round;stroke-miterlimit:4;stroke-dasharray:none;stroke-dashoffset:0;stroke-opacity:1;paint-order:markers fill stroke" id="path36" cx="68.414307" cy="130.71523" r="3.7620001" />

Circular path with the mug on top

<path style="opacity:1;fill:#717d82;fill-opacity:1;stroke:none;stroke-width:1.60968435;stroke-linecap:round;stroke-linejoin:round;stroke-miterlimit:4;stroke-dasharray:none;stroke-dashoffset:0;stroke-opacity:1;paint-order:markers fill stroke" d="m 68.414044,126.95318 a 3.7618673,3.7618673 0 0 0 -3.76153,3.76204 3.7618673,3.7618673 0 0 0 3.76153,3.76205 3.7618673,3.7618673 0 0 0 3.76206,-3.76205 3.7618673,3.7618673 0 0 0 -3.76206,-3.76204 z" id="path20" />

single path

<path style="opacity:1;fill:#717d82;fill-opacity:1;stroke:none;stroke-width:1.60968435;stroke-linecap:round;stroke-linejoin:round;stroke-miterlimit:4;stroke-dasharray:none;stroke-dashoffset:0;stroke-opacity:1;paint-order:markers fill stroke" d="m 68.414044,126.95318 a 3.7618673,3.7618673 0 0 0 -3.76153,3.76204 3.7618673,3.7618673 0 0 0 3.76153,3.76205 3.7618673,3.7618673 0 0 0 3.76206,-3.76205 3.7618673,3.7618673 0 0 0 -3.76206,-3.76204 z m -1.21542,0.92656 h 2.40554 c 0.0913,0.21025 0.18256,0.42071 0.27387,0.63097 h 0.47284 v 0.60099 h -0.17984 l -0.1664,1.05989 h 0.24961 l -0.34779,1.96267 -0.21238,-0.003 -0.22326,1.41955 h -2.12492 l -0.22429,-1.41955 -0.22479,0.003 -0.34829,-1.96267 h 0.26304 l -0.16692,-1.05989 h -0.1669 v -0.60099 h 0.44752 c 0.0913,-0.21026 0.18206,-0.42072 0.27336,-0.63097 z m 0.12608,0.19068 c -0.0614,0.14155 -0.12351,0.28323 -0.185,0.42478 h 2.52336 c -0.0614,-0.14155 -0.12248,-0.28323 -0.18397,-0.42478 z m -0.65524,0.63097 v 0.21911 l 0.0594,5.2e-4 h 3.35844 l 0.0724,-5.2e-4 v -0.21911 z m 0.16846,0.41083 0.1669,1.05937 h 2.80603 l 0.16693,-1.05937 -1.57046,0.008 z m -0.061,1.25057 0.27956,1.5782 1.34411,-0.0145 1.34567,0.0145 0.28059,-1.5782 z m 1.62367,1.75441 -1.08519,0.0124 0.19325,1.2299 h 1.79835 l 0.19328,-1.2299 z" id="path2714" inkscape:connector-curvature="0" />

You probably can see the code becomes more complex and needs more characters to describe it. More characters in a file result, of course, in a larger size.

Node cleaning

If you open an example SVG in Inkscape and press F2, that activates the Node tool. You should see something like this:

Inkscape – Node tool activated

There are 5 nodes that aren’t necessary in this example — the ones in the middle of the lines. To remove them, select them one by one with the activated Node tool and press the Del key. After this, select the nodes which define this lines and make them corners again using the toolbar tool.

Inkscape – Node tool make node a corner

Without fixing the corners, handles are used that define the curve, which gets saved and will increase file size. You have to do this node cleaning by hand, as it can’t be effectively automated. Now you’re ready for the next stage.

Use the Save as function and choose Optimized svg. A dialogue window opens where you can select what to remove or keep.

Inkscape – Dialog window for save as optimized SVG

Even the little SVG in this example got down from 3.2 KB to 920 bytes, less than a third of its original size.

Back to the getfedora page: The grey voronoi pattern used in the background of the main section, after our optimization from Part 1 of this series, is down to 164.1 KB versus the original 211.12 KB size.

The original SVG it was exported from is 1.9 MB in size. After these SVG optimization steps, it’s only 500.4KB. Too big? Well, the current blue background is 564.98 KB in size. But there’s only a small difference between the SVG and the PNG.

Compressed files

$ ls -lh insgesamt 928K -rw-r--r--. 1 user user 161K 19. Feb 19:44 grey-pattern.png -rw-rw-r--. 1 user user 160K 18. Feb 12:23 grey-pattern.png.gz -rw-r--r--. 1 user user 489K 19. Feb 19:43 greyscale-pattern-opti.svg -rw-rw-r--. 1 user user 112K 19. Feb 19:05 greyscale-pattern-opti.svg.gz

This is the output of a small test I did to visualize this topic. You should probably see that the raster graphic — the PNG — is already compressed and can’t be anymore. The opposite is the SVG, an XML file. This is just text and can compressed, to less then a fourth of its size. As a result it is now around 50 KB smaller in size than the PNG.

Modern browsers can handle compressed files natively. Therefore, a lot of web servers have switched on mod_deflate (Apache) and gzip (nginx). That’s how we save space during delivery. Check out if it’s enabled at your server here.

Tooling for production

First of all, nobody wants to always optimize SVG in Inkscape. You can run Inkscape without a GUI in batch mode, but there’s no option to convert from Inkscape SVG to optimized SVG. You can only export raster graphics this way. But there are alternatives:

• SVGO (which seems not actively developed)
• Scour

This example will use scour for optimization. To install it:

$ sudo dnf install scour

To automatically optimize an SVG file, run scour similarly to this:

[user@localhost ]$ scour INPUT.svg OUTPUT.svg -p 3 --create-groups --renderer-workaround --strip-xml-prolog --remove-descriptive-elements --enable-comment-stripping --disable-embed-rasters --no-line-breaks --enable-id-stripping --shorten-ids

This is the end of part two, in which you learned how to replace raster images with SVG and how to optimize it for usage. Stay tuned to the Fedora Magazine for part three, coming soon.

Today we’re very happy to announce that the third preview of the next minor release of ASP.NET Core and .NET Core is now available for you to try out. We’ve been working hard on this release, along with many folks from the community, and it’s now ready for a wider audience to try it out and provide the feedback that will continue to shape the release.

How do I get it?

You can download the new .NET Core SDK for 2.2.0-preview3 (which includes ASP.NET 2.2.0-preview3) from https://www.microsoft.com/net/download/dotnet-core/2.2

Visual Studio requirements

Customers using Visual Studio should also install and use the Preview channel of Visual Studio 2017 (15.9 Preview 3 or later) in addition to the SDK when working with .NET Core 2.2 and ASP.NET Core 2.2 projects. Please note that the Visual Studio preview channel can be installed side-by-side with existing an Visual Studio installation without disrupting your current development environment.

Azure App Service Requirements

If you are hosting your application on Azure App Service, you can follow these instructions to install the required site extension for hosting your 2.2.0-preview3 applications.

Impact to machines

Please note that is a preview release and there are likely to be known issues and as-yet-to-be discovered bugs. While the .NET Core SDK and runtime installs are side-by-side, your default SDK will become the latest one. If you run into issues working on existing projects using earlier versions of .NET Core after installing the preview SDK, you can force specific projects to use an earlier installed version of the SDK using a global.json file as documented here. Please log an issue if you run into such cases as SDK releases are intended to be backwards compatible.

What’s new in Preview 3

For a full list of changes, bug fixes, and known issues you can read the release announcement.

Routing

We’ve introduced the concept of Parameter Transformers to routing in ASP.NET Core 2.2. A parameter transformer customizes the route generated by transforming parameter’s route values, and gives developers new options when generating routes. For example, a custom slugify parameter transformer in route pattern blog\{article:slugify} with Url.Action(new { article = "MyTestArticle" }) generates blog\my-test-article. Parameter transformers implement Microsoft.AspNetCore.Routing.IOutboundParameterTransformer and are configured using ConstraintMap.

These features are specific to the new endpoint routing system used in MVC by default in 2.2.

Parameter transformers are also used by frameworks to transform the URI to which an endpoint resolves. For example, ASP.NET Core MVC uses parameter transformers to transform the route value used to match an area, controller, action, and page.

routes.MapRoute(
 name: "default",
 template: "{controller=Home:slugify}/{action=Index:slugify}/{id?}");

With the preceding route, the action SubscriptionManagementController.GetAll() is matched with the URI /subscription-management/get-all. A parameter transformer doesn’t change the route values used to generate a link. Url.Action("GetAll", "SubscriptionManagement") outputs /subscription-management/get-all.

ASP.NET Core provides API conventions for using a parameter transformers with generated routes:

• MVC has the Microsoft.AspNetCore.Mvc.ApplicationModels.RouteTokenTransformerConvention API convention. This convention applies a specified parameter transformer to all attribute routes in the app. The parameter transformer will transform attribute route tokens as they are replaced. For more information, see Use a parameter transformer to customize token replacement.
• Razor pages has the Microsoft.AspNetCore.Mvc.ApplicationModels.PageRouteTransformerConvention API convention. This convention applies a specified parameter transformer to all automatically discovered Razor pages. The parameter transformer will transform the folder and file name segments of Razor page routes. For more information, see Use a parameter transformer to customize page routes.

Link Generation

Added a new service called LinkGenerator, it is a singleton service that supports generating paths and absolute URIs both with and without an HttpContext. If you need to generate links in Middleware or somewhere outside of Razor then this new service will be useful to you. You can use it in Razor, but the existing APIs like Url.Action are already backed by the new service so you can continue to use those.

return _linkGenerator.GetPathByAction(
 httpContext,
 controller: "Home",
 action: "Index",
 values: new { id=42 });

For now this is useful to link to MVC actions and pages from outside of MVC. We will add additional features in the next release targeting non-MVC scenarios.

Health Checks

DbContextHealthCheck

We added a new DbContext based check for when you are using Entity Framework Core:

// Registers required services for health checks


services.AddHealthChecks()
 // Registers a health check for the MyContext type. By default the name of the health check will be the
 // name of the DbContext type. There are other options available through AddDbContextCheck to configure
 // failure status, tags, and custom test query.
 .AddDbContextCheck<MyContext>();

This check will make sure that the application can communicate with the database you configured for MyContext. By default the DbContextHealthCheck will call the CanConnectAsync method that is being added to Entity Framework Core 2.2. You can customize what operation is run when checking health using overloads of the AddDbContextCheck method.

Health Check Publisher

We added the IHealthCheckPublisher interface that has a single method you can implement:

Task PublishAsync(HealthReport report, CancellationToken cancellationToken);

If you add an IHealthCheckPublisher to DI then the health checks system will periodically execute your health checks and call PublishAsync with the result. We expect this to be useful when you are interacting with a push based health system that expects each process to call it periodically in order to determine health.

Tags

In preview3 we added the ability to tag health checks with a list of strings when you register them:

services.AddHealthChecks()
 .AddDbContextCheck<MyContext>(tags: new[] { "db" });

Once you’ve done this then you can filter execution of your checks via tag:

app.UseHealthChecks("/liveness", new HealthCheckOptions


{
 Predicate = (_) => false


});





app.UseHealthChecks("/readiness", new HealthCheckOptions


{
 Predicate = (check) => check.Tags.Contains("db")


});

We see tags as a way for consumers of health checks, application authors, to use as a convenient grouping and filtering mechanism for their health checks. Not something that health check authors will pre-populate.

You can also customize what status a failure of this check means for your application, for example if your application is written such that it can handle the database not being available then a database being down might mean Degraded rather than UnHealthy.

Validation Performance Improvements

MVC’s validation system is designed to be extensible and flexible allowing developer to determine on a per request basis what validators apply to a given model. This is great for authoring complex validation providers. However, in the most common case your application only uses the built-in validation pieces such as DataAnnotations ([Required], [StringLength] etc, or IValidatableObject) and don’t require this extra flexability.

In 2.2.0-preview3, we’re adding a feature that allows MVC to short-circuit validation if it can determine that a given model graph would not require any validation. This results in significant improvements when validating models that cannot or do not have any associated validators. This includes objects such as collections of primitives (byte[], string[], Dictionary<string, string> etc), or complex object graphs without many validators.

For this model – https://github.com/aspnet/Mvc/blob/release/2.2/benchmarkapps/BasicApi/Models/Pet.cs – the table below compares the difference in Requests Per Second (RPS) with and without the enhancement:

HTTP Client Performance Improvements

Some significant performance improvements have been made to SocketsHttpHandler by improving the connection pool locking contention. For applications making many outgoing HTTP requests, such as some Microservices architectures, throughput should be significantly improved. Our internal benchmarks show that under load HttpClient throughput has improved by 60% on Linux and 20% on Windows. At the same time the 90th percentile latency was cut down by two on Linux. See Github #32568 for the actual code change that made this improvement.

Requests Per Second Linux (higher is better)

Requests Per Second Windows (higher is better)

Request Latency Linux (lower is better)

Request Latency Windows (lower is better)

ASP.NET Core Module

We added support for the ability to detect client disconnects when you’re using the new IIS in-process hosting model. The HttpContext.RequestAborted cancellation token now gets tripped when your client disconnnects.

The ASP.NET Core Module also features enhanced diagnostics logs that configurable via the new handler settings or environment variables that expose a higher fidelity of diagnostic information.

<?xml version="1.0" encoding="utf-8"?>


<configuration>
 <location path="." inheritInChildApplications="false">
 <system.webServer>
 <handlers>
 <add name="aspNetCore" path="*" verb="*" modules="AspNetCoreModuleV2" resourceType="Unspecified" />
 </handlers>
 <aspNetCore processPath="dotnet" arguments=".\clientdisconnect.dll" stdoutLogEnabled="false" stdoutLogFile=".\logs\stdout" hostingModel="inprocess">
 <handlerSettings>
 <handlerSetting name="debugFile" value="debug.txt" />
 <handlerSetting name="debugLevel" value="TRACE" />
 </handlerSettings> 
 </aspNetCore>
 </system.webServer>
 </location>


</configuration>

SignalR Java Client

Preview 3 includes a few notable changes to the SignalR Java Client as we progress towards a 1.0 release:

The “groupId” for the Maven package has changed to com.microsoft.signalr. To reference the new package from a Maven POM file, add the following dependency:

<dependency>
 <groupId>com.microsoft.signalr</groupId>
 <artifactId>signalr</artifactId>
 <version>1.0.0-preview3-35501</version>


</dependency>

Or in Gradle:

implementation 'com.microsoft.signalr:signalr:1.0.0-preview3-35501'

In Preview 3 we’ve changed all the APIs to be asynchronous, using RxJava. Our Java Client documentation will be updated to show the new usage patterns. We also have support for the invoke method, allowing the client code to wait for the server method to complete. This version also includes support for serializing custom types in method arguments and return values.

The Java Client currently requires Android API Level 26 (or higher). We are investigating moving down to a lower API level before RTM. If you are planning to use SignalR in an Java-based Android application, please comment on the GitHub issue tracking our Android API level support so we know what API level would work well for our users.

Migrating an ASP.NET Core 2.1 project to 2.2

To migrate an ASP.NET Core project from 2.1.x to 2.2.0-preview3, open the project’s .csproj file and change the value of the the element to netcoreapp2.2. You do not need to do this if you’re targeting .NET Framework 4.x.

Giving Feedback

The main purpose of providing previews is to solicit feedback so we can refine and improve the product in time for the final release. Please help provide us feedback by logging issues in the appropriate repository at https://github.com/aspnet or https://github.com/dotnet. We look forward to receiving your feedback!

By Michael Coffer 17 Oct 2018

In Part 2 of our series, we deployed a Jenkins pod into our Kubernetes cluster, and used Jenkins to set up a CI/CD pipeline that automated building and deploying our containerized Hello-Kenzan application in Kubernetes.

In Part 3, we are going to set aside the Hello-Kenzan application and get to the main event: running our Kr8sswordz Puzzle application. We will showcase the built-in UI functionality to scale backend service pods up and down using the Kubernetes API, and also simulate a load test. We will also touch on showing caching in etcd and persistence in MongoDB.

Before we start the install, it’s helpful to take a look at the pods we’ll run as part of the Kr8sswordz Puzzle app:

• kr8sswordz – A React container with our Node.js frontend UI.

• puzzle – The primary backend service that handles submitting and getting answers to the crossword puzzle via persistence in MongoDB and caching in ectd.

• mongo – A MongoDB container for persisting crossword answers.

• etcd – An etcd cluster for caching crossword answers (this is separate from the etcd cluster used by the K8s Control Plane).

• monitor-scale – A backend service that handles functionality for scaling the puzzle service up and down. This service also interacts with the UI by broadcasting websockets messages.

We will go into the main service endpoints and architecture in more detail after running the application. For now, let’s get going!

Read all the articles in the series:

 

This tutorial only runs locally in Minikube and will not work on the cloud. You’ll need a computer running an up-to-date version of Linux or macOS. Optimally, it should have 16 GB of RAM. Minimally, it should have 8 GB of RAM. For best performance, reboot your computer and keep the number of running apps to a minimum.

Running the Kr8sswordz Puzzle App

First make sure you’ve run through the steps in Part 1 and Part 2, in which we set up our image repository and Jenkins pods—you will need these to proceed with Part 3 (to do so quickly, you can run the part1 and part2 automated scripts detailed below). If you previously stopped Minikube, you’ll need to start it up again. Enter the following terminal command, and wait for the cluster to start:

minikube start

You can check the cluster status and view all the pods that are running.

kubectl cluster-info kubectl get pods --all-namespaces

Make sure the registry and jenkins pods are up and running. 

So far we have been creating deployments directly using K8s manifests, and have not yet used Helm. Helm is a package manager that deploys a Chart (or package) onto a K8s cluster with all the resources and dependencies needed for the application. Underneath, the chart generates Kubernetes deployment manifests for the application using templates that replace environment configuration values. Charts are stored in a repository and versioned with releases so that cluster state can be maintained. Helm is very powerful because it allows you to templatize, version, reuse, and share the deployments you create for Kubernetes. See https://hub.kubeapps.com/ for a look at some of the open source charts available. We will be using Helm to install an etcd operator directly onto our cluster using a pre-built chart.

1. Initialize Helm. This will install Tiller (Helm’s server) into our Kubernetes cluster.

helm init --wait --debug; kubectl rollout status deploy/tiller-deploy -n kube-system

2. We will deploy an etcd operator onto the cluster using a Helm Chart.  

helm install stable/etcd-operator --version 0.8.0 --name etcd-operator --debug --wait

An operator is a custom controller for managing complex or stateful applications. As a separate watcher, it monitors the state of the application, and acts to align the application with a given specification as events occur. In the case of etcd, as nodes terminate, the operator will bring up replacement nodes using snapshot data.

3. Deploy the etcd cluster and K8s Services for accessing the cluster.

kubectl  create -f manifests/etcd-cluster.yaml
kubectl  create -f manifests/etcd-service.yaml

You can see these new pods by entering kubectl get pods in a separate terminal window. The cluster runs as three pod instances for redundancy.

4. The crossword application is a multi-tier application whose services depend on each other. We will create three K8s Services so that the applications can communicate with one another.

kubectl apply -f manifests/all-services.yaml

5. Now we’re going to walk through an initial build of the monitor-scale application.

docker build -t 127.0.0.1:30400/monitor-scale:`git rev-parse 

 --short HEAD` -f applications/monitor-scale/Dockerfile 

 applications/monitor-scale

To simulate a real life scenario, we are leveraging the github commit id to tag all our service images, as shown in this command (git rev-parse –short HEAD).

6. Once again we’ll need to set up the Socat Registry proxy container to push the monitor-scale image to our registry, so let’s build it. Feel free to skip this step in case the socat-registry image already exists from Part 2 (to check, run docker images).

docker build -t socat-registry -f applications/socat/Dockerfile 

 applications/socat

7. Run the proxy container from the newly created image.

docker stop socat-registry; docker rm socat-registry; docker run 

 -d -e "REG_IP=`minikube ip`" -e "REG_PORT=30400" --name 

 socat-registry -p 30400:5000 socat-registry

This step will fail if local port 30400 is currently in use by another process. You can check if there’s any process currently using this port by running the command lsof -i :30400

8. Push the monitor-scale image to the registry.

docker push 127.0.0.1:30400/monitor-scale:`git rev-parse --short HEAD`

9. The proxy’s work is done, so go ahead and stop it.

docker stop socat-registry

10. Open the registry UI and verify that the monitor-scale image is in our local registry.

minikube service registry-ui

11. Monitor-scale has the functionality to let us scale our puzzle app up and down through the Kr8sswordz UI, therefore we’ll need to do some RBAC work in order to provide monitor-scale with the proper rights.

kubectl apply -f manifests/monitor-scale-serviceaccount.yaml

In the manifests/monitor-scale-serviceaccount.yaml you’ll find the specs for the following K8s Objects. Role: The custom “puzzle-scaler” role allows “Update” and “Get” actions to be taken over the Deployments and Deployments/scale kinds of resources, specifically to the resource named “puzzle”. This is not a ClusterRole kind of object, which means it will only work on a specific namespace (in our case “default”) as opposed to being cluster-wide. ServiceAccount: A “monitor-scale” ServiceAccount is assigned to the monitor-scale deployment. RoleBinding: A “monitor-scale-puzzle-scaler” RoleBinding binds together the aforementioned objects.

12. Create the monitor-scale deployment and the Ingress defining the hostname by which this service will be accessible to the other services.

sed 's#127.0.0.1:30400/monitor-scale:$BUILD_TAG#127.0.0.1:30400/

 monitor-scale:'`git rev-parse --short HEAD`'#' 

 applications/monitor-scale/k8s/deployment.yaml | kubectl apply -f -

The sed command is replacing the $BUILD_TAG substring from the manifest file with the actual build tag value used in the previous docker build command. We’ll see later how Jenkins plugin can do this automatically.

13. Wait for the monitor-scale deployment to finish.

kubectl rollout status deployment/monitor-scale

14. View pods to see the monitor-scale pod running.

kubectl get pods

15. View services to see the monitor-scale service.

kubectl get services

16. View ingress rules to see the monitor-scale ingress rule.

kubectl get ingress

17. View deployments to see the monitor-scale deployment.

kubectl get deployments

18. We will run a script to bootstrap the puzzle and mongo services, creating Docker images and storing them in the local registry. The puzzle.sh script runs through the same build, proxy, push, and deploy steps we just ran through manually for both services.

scripts/puzzle.sh

19. Check to see if the puzzle and mongo services have been deployed.

kubectl rollout status deployment/puzzle

kubectl rollout status deployment/mongo

20. Bootstrap the kr8sswordz frontend web application. This script follows the same build proxy, push, and deploy steps that the other services followed.

scripts/kr8sswordz-pages.sh

21. Check to see if the frontend has been deployed.

kubectl rollout status deployment/kr8sswordz

22. Check to see that all the pods are running.

kubectl get pods

23. Start the web application in your default browser.

minikube service kr8sswordz

Giving the Kr8sswordz Puzzle a Spin

Now that it’s up and running, let’s give the Kr8sswordz puzzle a try. We’ll also spin up several backend service instances and hammer it with a load test to see how Kubernetes automatically balances the load.   

1. Try filling out some of the answers to the puzzle. You’ll see that any wrong answers are automatically shown in red as letters are filled in.

2. Click Submit. When you click Submit, your current answers for the puzzle are stored in MongoDB.

3. Try filling out the puzzle a bit more, then click Reload once. This will perform a GET which retrieves the last submitted puzzle answers in MongoDB.

Did you notice the green arrow on the right as you clicked Reload? The arrow indicates that the application is fetching the data from MongoDB. The GET also caches those same answers in etcd with a 30 sec TTL (time to live). If you immediately press Reload again, it will retrieve answers from etcd until the TTL expires, at which point answers are again retrieved from MongoDB and re-cached. Give it a try, and watch the arrows.

4. Scale the number of instances of the Kr8sswordz puzzle service up to 16 by dragging the upper slider all the way to the right, then click Scale. Notice the number of puzzle services increase.

If you did not allocate 8 GB of memory to Minikube, we suggest not exceeding 6 scaled instances using the slider.

In a terminal, run kubectl get pods to see the new replicas.

5. Now run a load test. Drag the lower slider to the right to 250 requests, and click Load Test. Notice how it very quickly hits several of the puzzle services (the ones that flash white) to manage the numerous requests. Kubernetes is automatically balancing the load across all available pod instances. Thanks, Kubernetes!

​6. Drag the middle slider back down to 1 and click Scale. In a terminal, run kubectl get pods to see the puzzle services terminating.

7. Now let’s try deleting the puzzle pod to see Kubernetes restart a pod using its ability to automatically heal downed pods

a. In a terminal enter kubectl get pods to see all pods. Copy the puzzle pod name (similar to the one shown in the picture above).

 b. Enter the following command to delete the remaining puzzle pod. 

kubectl delete pod [puzzle podname]

c. Enter kubectl get pods to see the old pod terminating and the new pod starting. You should see the new puzzle pod appear in the Kr8sswordz Puzzle app.

What’s Happening on the Backend

We’ve seen a bit of Kubernetes magic, showing how pods can be scaled for load, how Kubernetes automatically handles load balancing of requests, as well as how Pods are self-healed when they go down. Let’s take a closer look at what’s happening on the backend of the Kr8sswordz Puzzle app to make this functionality apparent.  

1. pod instance of the puzzle service. The puzzle service uses a LoopBack data source to store answers in MongoDB. When the Reload button is pressed, answers are retrieved with a GET request in MongoDB, and the etcd client is used to cache answers with a 30 second TTL.  

2. The monitor-scale pod handles scaling and load test functionality for the app. When the Scale button is pressed, the monitor-scale pod uses the Kubectl API to scale the number of puzzle pods up and down in Kubernetes.

3. When the Load Test button is pressed, the monitor-scale pod handles the loadtest by sending several GET requests to the service pods based on the count sent from the front end. The puzzle service sends Hits to monitor-scale whenever it receives a request. Monitor-scale then uses websockets to broadcast to the UI to have pod instances light up green.

4. When a puzzle pod instance goes up or down, the puzzle pod sends this information to the monitor-scale pod. The up and down states are configured as lifecycle hooks in the puzzle pod k8s deployment, which curls the same endpoint on monitor-scale (see kubernetes-ci-cd/applications/crossword/k8s/deployment.yml to view the hooks). Monitor-scale persists the list of available puzzle pods in etcd with set, delete, and get pod requests.

We do not recommend stopping Minikube (minikube stop) before moving on to do the tutorial in Part 4. Upon restart, it may create some issues with the etcd cluster.

 a. curl -sL https://deb.nodesource.com/setup_7.x | sudo -E bash -

 b. sudo apt-get install -y nodejs

 a. cd ~/kubernetes-ci-cd

 b. npm install

npm run part1 (or part2, part3, part4 of the blog series)

Up Next

Now that we’ve run our Kr8sswordz Puzzle app, the next step is to set up CI/CD for our app. Similar to what we did for the Hello-Kenzan app, Part 4 will cover creating a Jenkins pipeline for the Kr8sswordz Puzzle app so that it builds at the touch of a button. We will also modify a bit of code to enhance the application and enable our Submit button to show white hits on the puzzle service instances in the UI.  

Curious to learn more about Kubernetes? Enroll in Introduction to Kubernetes, a FREE training course from The Linux Foundation, hosted on edX.org.

This article was revised and updated by David Zuluaga, a front end developer at Kenzan. He was born and raised in Colombia, where he studied his BE in Systems Engineering. After moving to the United States, he studied received his master’s degree in computer science at Maharishi University of Management. David has been working at Kenzan for four years, dynamically moving throughout a wide range of areas of technology, from front-end and back-end development to platform and cloud computing. David’s also helped design and deliver training sessions on Microservices for multiple client teams.