Cloud Native Computing Foundation KubeCon + CloudNativeCon China 2019 (Shanghai), 5 Jul 2019

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: Deployment and Management in the Age of Cloud Integrated Applications - Neil Peterson, Microsoft

Description

Join us for Kubernetes Forums Seoul, Sydney, Bengaluru and Delhi - learn more at kubecon.io

Don't miss KubeCon + CloudNativeCon 2020 events in Amsterdam March 30 - April 2, Shanghai July 28-30 and Boston November 17-20! Learn more at kubecon.io. The conference features presentations from developers and end users of Kubernetes, Prometheus, Envoy, and all of the other CNCF-hosted projects

Deployment and Management in the Age of Cloud Integrated Applications - Neil Peterson, Microsoft

With the advancement in cloud service capabilities such as databases, cognitive services, and messaging queues, application architecture is extending to include not only code but also managed cloud services. In this talk we will examine the unique challenges in deploying and managing cloud integrated applications. Topics will include: - How to deploy and manage code and cloud services as a single unit - Managing connection strings and keys between application and cloud services - Comparison of emerging application deployment technology such as the Kubernetes Service Catalog, Terraform, and Cloud Native Application Bundles (CNAB) The goal of this talk is to not only consider how cloud integrated applications change deployment and management practices, but to also take a hands on look at the emerging technology that facilitates these deployments.

https://sched.co/Nrmg

A

All right, I'm gonna go ahead and get started. Thank you for coming out. My name is Niall Peters and I work with Microsoft and measure. This is not a talk about Asher. This is a talk about deployment in the age about distributed, applications, guys just 35 minutes and an absolute ton of information to go through. So a lot of its gonna be a surface level. I do have a bunch of demos as well.

B

A

Pre record all these demos, however, everything that I'm gonna walk through I've got available.

C

Github repo.

A

C

A

All of these slides have been uploaded the schedule builder as well, but so here's the things that we're gonna take a look at software deployment through the years throughout the years, particularly throughout my years. My career spans about 20 years and.

C

A

Primarily been doing software deployment for most of those years. Well then, look at like you know, what's changed, why? Why do? Why? Are we talking about this in a different context today and then we're gonna look at three options for deploying software.

A

These aren't the only three. These are just the three that I decided to show off here, but really you know, the goal is to talk about the challenges and what's changed and some different.

B

C

To deploying software.

A

And everything that that software needs to run it just so happens that these three kind of hit different like pros and cons, but there are others out there as well. um So that's that's the quick agenda again, there's a there's, a link to all the demonstration and that we're gonna see here so.

C

A

About software deployments, I touched this definition off of Wikipedia Wikipedia defined software deployment is all of the activities that make software systems available for use.

B

A

Like I said, I've been deploying the software or doing software deployment related activities for about 20 years and when I started honestly its bullet point three: here's where I started I, basically built my whole career. What.

C

Software updates.

A

I worked early.

C

On it works primarily in the windows-based.

A

Data centers, so Patch Tuesday was kind of what I did for a living, but then I expanded from there and did.

C

A lot of operating.

A

System deployment and then application on top of master.

B

Whether it's client.

A

Based applications or applications.

B

A

Data centers, however, you know as as I grow into the the cloud age. Things are changing and I found my activities not only encompassed applications operating system and software updates. But now it was up to me, as the deployment engineer, to think about things like deploying infrastructure.

B

A

Deployment time making.

B

Sure the right access controls around place here.

C

That infrastructure.

A

B

A

Applications and.

B

A

Finally, deploying policies and I'm sure there's some other things in here as well, but defining policies to make sure that that appointment stays in the state that, if an kind of real quick look at employment through the ages and how things have changed there. But now, let's take a look, a kind of a modern application- and this is kind of my canonical example of.

B

A

To describe a modern.

C

B

This is an actual application, in fact, in one of demos, we're going to.

A

Deploy this application.

B

The application interests you again there's a.

A

Link to a github repo with.

B

This application lives running.

C

Through what a judge.

B

On this side right here, we've got a Buster on.

A

That side over there, we have.

B

What I'm trying.

C

B

With this application, I've got a plot. That's basically.

A

Pulling the body of tweets on Twitter.

B

A

And storing that in a message there you go to.

C

Do some message queue research talk about spider, then have another pod off.

B

Of that queue I truly it takes.

A

The body and evaluates it against the tech sentiment.

B

A

Any one of your cloud providers and I basically return.

C

B

A

Was this tweet positive in nature.

B

What is it negative in.

C

Takes those results and stores them in a database? It doesn't really matter what that database is and then finally I've got.

B

A broad that visualizes so very different type of application than the applications that I was going.

A

20 years ago,.

B

C

Years ago, I had like an executable, and these were MSI.

B

A

Would install that on a server and boom there's my application.

C

Solution still I.

A

Had that one installer application, a very separate, in fact, when I first wrote this session when I first built this application, you.

B

A

B

You don't annually.

A

Created the data store.

B

In village and then I had to do.

C

Things like a connection ticket and he'd.

B

Bring those over into.

A

Kubernetes I start.

B

A

Pod those connection was stringing his environment variables.

B

And that was very clunky at.

A

A point it took me a couple hours or.

B

Two and then I thought it was it.

C

And I wrote a bash script to.

A

Do it, but but what I want to show.

C

You is it without other options that are quite.

A

A bit more sophisticated now.

C

You might be asking yourself but wait what about CI CD? You know why.

A

Do I need some.

C

Reason vessels pipes for.

A

Their deployment pipelines in my environment and that's.

C

A that's a very valid.

A

C

A

You're working with an application.

B

If I've got one in this.

A

B

A

I wanted to play that to QA test and then finally to productions, and then maybe this talk is not relevant. I think.

B

However, if I've got a new, this is something that somebody can you.

A

Know open up their phone and maybe.

C

They don't know there, but.

A

This maybe they're just signing up for a service, no good every time they sign up for that service. They don't want to be with an instance of this application.

C

There boy that's.

B

A very different situation.

C

Something that.

A

I might use a.

C

Pipeline for and kind of, like I.

A

Only have one instance of this application scenario, so this is the application like.

C

I said we'll always get a deployment.

A

Of this application a little bit later on, but so what are you know? Let's examine like how this changes things and we've already seen that a single application can be built.

B

A

Deployed Robby Baker.

A

A

And this is pretty cool.

B

Like just applications like deliver an author or developer of that application, to do things.

A

Get a lot easier like that: let's take a sentiment. Analysis like I didn't have to go, build.

B

With sentiment.

A

Analysis I could just subscribe to as.

B

A

Application developer, like this.

B

Modern world, but as a deployment.

A

A

There's there's some new challenge: a multiple different deployment rotation.

B

I already talked about it like when I first.

A

Built that application.

C

So I was manually stuck in my cloud providers and.

A

Then manually deploying stuff in my environment.

A

I wanted to go. Look at this thing. I had to kind of like poke through my cloud provider to see.

C

Half of the application, when you read.

A

Eddie's cluster, to see the.

C

Other job keeping their kind of a.

A

Disjointed experience there, those are simple ones. It seems next to that I, particularly.

B

A

B

The first ones the secrets.

A

Manager, that's it so if.

B

We think about it.

B

A

Know get over the other Eddie.

C

So that actually.

A

C

For instance, you know I've got.

A

A connection to bring in a key to.

B

A

B

Todd needs that.

C

Need that information, this pod needs that information or get out of there and.

A

That's what I mean by.

B

Secrets management.

A

I'm and then finally instance management. We.

B

Talked about the point affirmation. But what.

A

About de provisioning for the same application, let's.

C

Say I have a thousand.

A

Instances of that Twitter.

C

B

Need to go to the poi one, let me it's a service.

A

Of Excel and.

B

Then somebody no.

A

Longer wants to pay me for that service humanization. One.

B

A

That application I need to make sure.

C

I get the right database and the right sentiment.

A

C

Service and the right pot, are you doing like.

B

C

These things together so that I pin it to be provision them right.

A

With a single cushion.

B

A

It may be given.

B

That that's what I need like.

A

Managing a large pool of it.

C

So these are kind of some of the challenges.

A

And next we're gonna look at three solution.

C

B

Or may not you.

A

B

The right solutions for everyone.

C

And, and also just.

A

To kind of really be very transparent, this.

C

A

C

B

C

Lot of what we're gonna see is not exactly what I would consider being brilliant production.

B

A

C

Bunch you're getting our heads around.

B

A

You know this new world of distributed application.

B

Emergent technology, so the first one is terraformer. Anybody using a couple.

A

Hands, unfortunately, I'll have enough time to go like.

C

So I'll really.

A

Just talk about.

B

A

To offer here honestly, I didn't even think about terraform when I first started down this path, I actually looked at the fur, the the remaining two that we're gonna see. First, so then it occurred to me like man. Terraform is actually relatively mature, like.

C

I just made a stage, though not production-ready,.

A

But of everything that we're gonna see this your bike, the most.

C

Solution, that's it.

A

Also has some camille its.

C

Own challenges.

A

C

A

Allowing us to deploy in structure.

B

C

Whatever it may be,.

C

A

Terraform zero points 12.0 was released a couple weeks ago.

B

A

Terraform works on the turn. Writer got.

B

C

Terraform and then.

A

A provider and there's a bunch of providers, providers.

C

For all your major.

A

Cloud services or cloud providers there's.

C

B

A

One of the like kind of the brilliant pieces around terraform.

B

That allows us a.

B

Bunch of instances same thing.

A

Is something called terraformed state to enter.

B

One where the springs on the state, but every time I display our form, shows.

C

Something called a state file is created basically like the.

A

Record of what was deployed I'll.

C

Call out right now it's actually a clear spot to representation.

A

Of that configuration I call out that clear text piece because you've got to be careful.

B

Eight miles and you can any other secrets.

A

B

Now we'll see.

A

Kind of a way to secure that so.

B

A

The state file is the record of what it's been to blow in a.

B

A

Space is the second piece here, shakin.

C

Is like a virtualized machine because.

B

She knew funding, the cutting display is terraformed.

A

Configuration works very similar to name.

B

So I can have single terraform configuration Boyet.

A

C

One of those works.

B

A

All all three of these I'm gonna kind of kind of give my impression you.

B

B

A

This is not like I'm not trying to make these complete.

C

Against each other or I'm not trying to tell.

A

You, which one is it's.

C

Good or bad they've all got.

A

So instance, management.

B

Review state file works.

C

In the tariff or.

B

Tariffs, one has this thing called.

A

B

Which one allows us boy? The.

A

B

A

B

A

These last two that I really kind of want to focus on these.

B

Are the two that really kind of like helped.

A

Me decide what I want.

B

A

The maturity and legibility involvement- and this is.

C

Where we just like, destroys everything else, we're able to look at and.

A

In a good way, Tara.

B

Forms a relatively mature product only.

C

Been around for like five.

A

Years but I mean.

C

It's going gangbusters a lot of people are using its ability, is, you know, I, have it like a medium.

A

C

Can write your own providers or in some.

A

C

A

Azure provider, if it doesn't have a capability you can. You can write that capability source project, so you can get that in. If that's.

B

Something you're willing to do.

A

So with that said, let's take a look at a demo and, like I said, these are all pre-recorded.

B

C

I do apologize.

A

I know this is hard to see, unfortunately, I pre-recorded it and I can't really like zoom in or anything like that, but I'll talk through what's going on. um So this is a this is a terraform configuration. It doesn't really matter what its gonna deploy. There's just a bunch of stuff in here that we're gonna deploy, but one thing that I want to call out.

A

Something that I will call out is right here inside of the configuration and I'll pause on it here. In just a moment, you can see that I'm using this is an interpolation from expression, 5 terraform, dot workspace. What that's gonna do is take the workspace name and we'll see how to create named workspaces and like use that as a value in my deployment.

A

um So in this case, I'm gonna create a resource group in Azure, which is a logical containers for my resources, I'm gonna name that I'm gonna, give it an identical name as the workspace that I'm working in I mean that's kind of how I can like logically, keep this stuff sorted.

C

A

Then lastly, this configuration up here is a back-end storage, and this takes that state file and stores it in Azure, so I kind of made that call out that you've got clear text in there. That's one way to kind of secure that alright, so now, I've run this command. Terraform workspace list which showed I had two workspaces, I did default and then I created a new one called test. Environment, I just ran terraform plan and then terraform apply, which is going to deploy an incident of this.

A

This configuration again, it's just a simple application here. So if I flip over to Azure and refresh we'll see that I've got a resource group that matches the name of the test workspace and then, if I go into the storage account, we can see. I've got a state file that also matches the name. We can see terraformed aiya, yes stain, she has an environment and that's the name of my workspace. I've created an instance of this configuration and I can see that instance and the name matches the the workspace that I've created.

B

A

Next thing I'm gonna do- and this is this- goes pretty quickly, because I didn't want to waste time watching everything is I'm going to create a new workspace called production. Environment I just ran a terraform workspace list, so we can say: I've got a default workspace which we always have test environment with the first workspace I created and in production environment, which is the workspace that I'm working in now I'll.

C

A

Terraform plan terraform apply, which deploys an instance of the application. We can now see that I've got two state files in.

B

My state store.

A

If I look at my resource groups, we can see that I have two resource groups test and production. So now I've got two instances of this application, and now I can start to destroy them. So what I've done here is I've run a terraform workspace, select test environment I want to delete this test, one so I've selected. The workspace I then did terraform workspace list.

A

They can now see, but via that asterisk, that I'm working in the context of that test, workspace, I'm, gonna, run terraform, destroy, which is then gonna go and destroy just that instance of the application. So it's a pretty simple configuration they're using terraform for this stuff, I.

B

A

The workspace stuff, a little clunky, you can't see kind of like the workspace inside of the code. You've got to use the terraform commands to do it, but it's definitely definitely an option so.

C

The next one that I want to talk about is.

A

Keeping any service catalog anybody seen or work with different any service catalog one hand in a couple: hands cool a couple: more hands, so cool. So this is the one that uh you know. I'll spend some time. I'll go a little deeper on this. One is a cute on. um So what look before we talk about kubernetes Service, Catalog I want to talk about open service broker. This is a part of it and if I do it in the other order, it doesn't really make a whole lot of sense. I mean I.

A

Do that there's gonna be a lot of information here. I've got some diagrams to kind of pull it all together. So what open service broker is? Is an API specification for a standard cloud provider interface, meaning if you run like write an API that looks like this put it in front of AWS or Azure, and then you can write scripts in code that can interface with that API to provision things in that cloud provider.

A

It basically specifies five operations provision find unbind, D provision.

B

A

Owed open service broker just keep that in your mind, we'll get back to that kubernetes service catalog is a KPI extension for kubernetes that kind of enlightens kubernetes so that it knows how to talk to an open service broker compliant API, so out-of-the-box kubernetes cannot interface with an open service broker. Api service catalog allows it to do so, and basically service catalog adds five types to kubernetes. So just like a pod where a deployment as a type when you install Service Catalog, we get these five new types.

A

Cluster service broker service class or his plan service instance and service bindings will see a bunch of these in the demonstration. I actually usually do about an hour talk on service catalog. Don't have enough time to fit all that in here. So I'm not gonna define all of those we'll see those ones a demonstration. So let's see how this looks right here, so we've got kubernetes with the service catalog installed.

B

A

Then have a service broker in the case of the demonstration we're gonna see today, it's called open service broker for a sure. So it's just an it's. A open service broker, compliant API for interfacing, with Azure.

A

And that service broker can interface with Asher.

C

A

The cool thing here is that I can have multiple service brokers fronting like a single contribute, IDs clustered, so we can see right now. Service broker for AWS.

C

A

Just to kind of like prove out like how valuable these open service brokers are. I don't actually need super Nettie's here, like once, I have a service broker that interface with these cloud providers. I, can you know exchange paper Nettie's with something like clouds down? That's kind of how Service Catalog and the open service broker is kind of line up with cloud providers.

B

A

Let's kind of look even deeper kind of at a kubernetes like API level, how this all works, so installing a Service Catalog you can install it right into the cluster using a home church and then installing a service broker. You can also install it right into the cluster using a home chart method. So this dotted line right here represents.

C

A

Nettie's cluster I'm.

B

So I've got kubernetes.

B

A

Installed a service catalog and here's these five new types.

B

Inside of my cluster I've.

A

Installed my service broker again, these are just tells her that I can install inside the cluster.

C

A

Now, actually, I'll describe some of these types, so this service class represents like a service inside a MicroBot provider. Maybe it's a storage account. Maybe it's the text sentiment API.

B

A

Then the plan is kind of like this skew, like you like a lot of these cloud providers. You'll have like p0 for like a small instance and p5 for like a large instance. That's what the plan is. The service isn't kubernetes.

B

Type, that represents in.

A

One of these cloud services, and then the binding, is what the binding is. Is they reached about? Do that service? That.

C

We created and.

A

Keys and stores them in kubernetes as a kubernetes secret so that my pods can access those secrets and then access the services. So this is how this works once I have everything installed: I.

B

A

Create a service instance excited kubernetes, which provisions some piece of infrastructure to.

C

A

My cloud provider I then create a service binding which gets the pieces, whatever I need to connect to that services and stores them inside of kubernetes as a new brunetti secret I commenced.

C

To start my pods.

A

Which can consume those secrets, and now my pod knows how to interface.

B

With those services.

A

That I've created in the cloud provider and can do things like whatever it needs to do um so that is Service Catalog and.

A

So here's my impressions of instance, management and I guess one thing I didn't call out is like I, just threw a bunch of stuff out there service in social service classes. Well, when you write this stuff, you're just writing occur in any manifest, so we're extending kubernetes there's five new tights and we can you access them with all the same. You know calls that we do things like COD, so I can do like cube CTL.

C

A

Instance, whatever, and when I write the the manifest file to deploy this stuff, it looks like this: you burn any manifest file, it is a kubernetes.

A

Because of that, I can do things like write home charts that lets go back to the first couple. Slides I have that Twitter application I've, actually I, actually have a color chart that deploys all of that not.

C

Just the pod, but also a cloud.

B

C

A

That process all the secrets from.

B

Those cloud services.

C

Herstory infinity.

A

Buster and my pod to be taken care of so because of that I can be like helm, install helm, install helm, install now I've.

B

Got three above it.

A

Then I can do a helm, delete.

B

A

I can delete.

B

A

Management, we haven't written any secrets for that maturity and community I. Would college kind of medium I mean it's a project? That's a lot of organizations are building.

B

And kind of you.

A

Know putting some emphasis on using these open service reference from.

C

Kubernetes but I was so called a.

A

Relatively immature project and.

C

Then flexibility.

A

Extensibility, it's all open source, so you can definitely write your own providers or your own service broker and or contribute to them. But you still have to kind of go through that.

A

And I'm gonna dig. You know we're going to look at quite a few things here, because this is cube con.

A

The first thing I want to do is look at some of the infrastructure stuff. So I said that we install all of this with home charts, so I'm just doing cutesy tail get pods all namespaces, you can see. I've got two pods, that's the catalog and then I've got two bought two pods. That's the service broker I've just installed this stuff. Using a helmet arts. I said that it was a API extension but I just run Q CTL get API service and we can see in here. I've got an extended.

A

The kubernetes api too much food service and catalog. The next thing, I'm going to do is run cute. Ctl get service broker. This is just gonna return. A list of service brokers installed in my cluster. You can see: I've got O's ba', which is the open service broker for Azure.

B

A

I can describe that service broker, who can actually see an event here that it successfully fetched catalog entries for the broker, that's kind of the infrastructure pieces, that's kind of a set it and do it once type of stuff. But now, let's look at like classes plans and how to deploy stuff like Service, Catalog, I'm gonna run a cute CTL hich service classes, and here we come actually. We've actually got a catalog now inside the pyramid ease of services that I can deploy in Azure. So we can see things like red.

A

Is cash, eight, app insights, sequel so now I can see of this list of things. I can deploy directly computer Medi cube, CTL gets service plans. If you recall the plan was kind of excuse or the size of that service and notice here that it's a little hard to read, because I can see like basic free but I, don't have a relationship between visit plan and and the class itself. So there's some rough edge of the chair. Definitely we've got this tool called SP cat.

A

It's attempting the secret kind of it answers some of those rough edges, but it's definitely not a necessary tool, but just to show it off when we're an SSD cat get plans. And now you can actually see like as your sequel is the class and then I've got a plan of basic or as your sequel premium, but I can kind of browse that stuff using the SP cat tool, but again we're just kind of looking at the stuff. That's loaded here for us, let's actually deploy something here.

A

So the first thing I'm going to do is create a service than shoot so I want to deploy that thing into my cloud provider, so I'm running with keep CTL, create and then I'm just best, applying a yamo file I'm starting that right now, because it does take a couple of couple seconds to run, but now, let's take a look at the mo file. So what you can see here is an I've got kind of service instance.

A

I've got some metadata like the namespace that I wanted to ploy it into I've, got the class name so out here: storage, the plan name all in one and then some parameters that are needed, like what locate as your location go on to deploy. Thank you and what resource group do I wanted to play that into, but if you'll notice in a minute.

B

A

Can run cube, CTL, get service instances to see the state of that provision, so this thing is being provisioned but I'm. Seeing this from kubernetes.

C

A

I come jump over to a sure. We can actually see my resource group and there's my storage account now I'm gonna create that service binding. The instance is great, but I can't access it from my pods unless I've got those connections free. So again, we've got the manifest file kind service, bindings I'm, giving it a name like what instance do I want to bind to and then what? What's the name of the secret that I want to create and.

B

I've, just given it that name I.

A

Can you know, return information about the service binding.

B

We can see that it's ready I can.

A

Then run with detail, get secrets very secret, see that my secret has been created and then finally I'm.

B

A star Kapaa, the only case, that's gonna,.

A

Consistent variables and then that pot is gonna, do something now.

B

I'm not doing the cinnamon.

A

Here I've just got a very simple little application orange account that was created, but.

B

You can see it's.

A

Just a standard renamed manifest file, I'm, creating a container and.

B

I'm setting these environment variable, but setting.

A

Those kind of getting the values for them in those environment variables from with secrets that were.

B

Created by the servo trying to take over I.

A

Did this in kind of three files here.

C

Like it has a manifest file or like.

B

A

A helmet chart to do it, which.

C

I've got some demonstration.

B

You're a bit of how actually, if those.

A

For the sake of time, but you can see that.

C

We've actually created a file.

B

C

In my storage, odd did that for.

A

C

Created that you had started up it.

A

Consumed the secrets created.

B

A

All the helmet demo was that I just use television deploy a couple more and then I ran tell them.

B

To leave holy deleted.

B

A

B

B

A

B

About terrible, we'll talk.

A

About service catalog, this thing is like Brandis banking to.

B

See does anybody use native.

A

Application bundles are read about that, so I'm not gonna, be able get to the demo here.

B

B

Formatting good.

A

B

A

And stolid, knowing.

B

A

Many things I really think about it.

B

Which is like a make file on that you're using seen a package.

A

Up everything we need with a stall or.

B

C

Oci compliant format like.

A

Gordon, your painter ready no-go retreats into using command-line.

B

A

B

C

A

Cloud native application funny.

A

Wayne there's a couple different.

B

A

A cloud native application bundle he's.

B

B

A

Home install maybe.

B

After I'm deployed.

A

In my application, install.

B

A

Whatever else made.

B

All your application.

A

B

C

B

A

Is just the kind of the entry point now seen app is just the spec I was.

B

Synonym to war it's what kind of a format it's.

A

Just respect for the scene, a bundle itself.

C

Producer even have several different implementations.

A

That are actually kind of like.

B

A

Deploy our scene a bundle like first.

B

One in these are the story that I know I'm sure.

A

There's other being others being built.

B

A

The bundle runtime is hand off. They.

C

A

Don't let this thing called Porter and.

C

I am short on time later. I would.

B

Recommend your.

B

C

A

Handles a lot of stuff for you like creating.

B

C

A

Writing the the spec and then.

B

Symphonies managing the deployment.

A

Of those bundles to.

B

The meeting or registries and.

C

Then docker app, probably.

A

The most mature.

A

I do apologize that I wasn't able to get into the demo.

A

And one of the great things about scene out.

A

B

A

That I needed it to blue I needed to make a pull request, see.

B

Map help do that by on tour.